notebooks/OSDR-evaluation.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "0218187e-16be-4d4f-af8f-bf16270f2794",
   "metadata": {},
   "outputs": [],
   "source": [
    "%reload_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "9f9a7deb-e1e4-4214-91e6-69e92b132de3",
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install openpyxl tenacity loguru  python-dotenv rake-nltk seaborn -qq\n",
    "!pip install git+https://git@github.com/NASA-IMPACT/evalem.git@develop -qq"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "a0dcc700-a774-469c-92bd-78a085931103",
   "metadata": {},
   "outputs": [],
   "source": [
    "import ast\n",
    "import glob\n",
    "import json\n",
    "import os\n",
    "import random\n",
    "import re\n",
    "from typing import List"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e00f4048-be53-4cfa-819e-25ec66a8e98d",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pprint import pprint\n",
    "from IPython.display import display, HTML"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "4ba3bbc0-5572-4af7-a2f4-dd2053b32327",
   "metadata": {},
   "outputs": [],
   "source": [
    "from loguru import logger"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "8754ddab-a661-47c0-98cf-da4a00265f02",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "8f49bd28-5009-43e1-90f6-483ac8f794b6",
   "metadata": {},
   "outputs": [],
   "source": [
    "from dotenv import load_dotenv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "f5a5d561-792d-49d9-b7fd-f1a8e0185bf3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "load_dotenv()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "276c87f8-95c6-4845-95a2-da00a257205d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Archive:  ../data/osdr-eval.zip\n",
      "   creating: ../data/osdr-eval/\n",
      "  inflating: ../data/osdr-eval/responses-gpt-3.5-turbo.csv  \n",
      "  inflating: ../data/osdr-eval/osdr-eval-retrieval-kws.csv  \n",
      "  inflating: ../data/osdr-eval/OSDR_LLM_QA_correctFormat.xlsx  \n",
      "  inflating: ../data/osdr-eval/responses-simplified-gpt-3.5-turbo-rephrased.csv  \n",
      "  inflating: ../data/osdr-eval/osdr-eval-retrieval-semantic.csv  \n"
     ]
    }
   ],
   "source": [
    "!unzip ../data/osdr-eval.zip -d ../data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ff194b8d-1240-4489-94e6-bdba80632d79",
   "metadata": {},
   "source": [
    "# Load Validation Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "7be67049-2832-4d4e-a413-2cbede699a3e",
   "metadata": {},
   "outputs": [],
   "source": [
    "FILE_PATH = \"../data/osdr-eval/OSDR_LLM_QA_correctFormat.xlsx\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "d49eb939-2b6b-451f-acaf-78cc9c1b665c",
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_data(path) -> pd.DataFrame:\n",
    "    data = pd.read_excel(path)\n",
    "    data = data.rename(columns={data.columns[0]: 'context'})\n",
    "    return data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "aec1b9b5-9c05-454e-b178-d47fbdacb1bf",
   "metadata": {},
   "outputs": [],
   "source": [
    "def simplify_columns(df: pd.DataFrame) -> pd.DataFrame:\n",
    "    df = df.copy()\n",
    "    transformed_data = []\n",
    "    for index, row in df.iterrows():\n",
    "        context = row['context']\n",
    "        url = row['URL']\n",
    "        for i in range(1, 6): # Assuming there are 5 question-answer pairs\n",
    "            question = row[f'Question {i}']\n",
    "            answer = row[f'Question {i} Answer']\n",
    "            if question and answer:\n",
    "                transformed_data.append(dict(\n",
    "                    context=context,\n",
    "                    context_hash=abs(hash(context)) % (10 ** 8),\n",
    "                    url=url,\n",
    "                    question=question,\n",
    "                    answer=answer,\n",
    "                ))\n",
    "    df =  pd.DataFrame(transformed_data)\n",
    "    return df.dropna(subset=['question', 'answer'], how='all')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "464f6e4b-7701-403c-bf30-2286092b8cbd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>context</th>\n",
       "      <th>URL</th>\n",
       "      <th>Question 1</th>\n",
       "      <th>Question 1 Answer</th>\n",
       "      <th>Question 2</th>\n",
       "      <th>Question 2 Answer</th>\n",
       "      <th>Question 3</th>\n",
       "      <th>Question 3 Answer</th>\n",
       "      <th>Question 4</th>\n",
       "      <th>Question 4 Answer</th>\n",
       "      <th>Question 5</th>\n",
       "      <th>Question 5 Answer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Genome-wide transcriptional profiling showed t...</td>\n",
       "      <td>https://sciencediscoveryengine.nasa.gov#/previ...</td>\n",
       "      <td>Where were the gene expression alterations in ...</td>\n",
       "      <td>reducing gravity levels in the International S...</td>\n",
       "      <td>What kind of experiments showed differential r...</td>\n",
       "      <td>simulation experiments on ground testing space...</td>\n",
       "      <td>What is overrepresented in the different exper...</td>\n",
       "      <td>the same GO groups including a large multigene...</td>\n",
       "      <td>what analytical methodology provided integrati...</td>\n",
       "      <td>Self-organizing maps of the Gene Expression Dy...</td>\n",
       "      <td>What are the implications of the findings in t...</td>\n",
       "      <td>The findings indicate that the finely adjusted...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>The objective of the Rodent Research-5 (RR-5) ...</td>\n",
       "      <td>https://sciencediscoveryengine.nasa.gov#/previ...</td>\n",
       "      <td>What was the main goal of the Rodent Research-...</td>\n",
       "      <td>to evaluate bone loss in mice during spaceflig...</td>\n",
       "      <td>What type of mice and how old were they when f...</td>\n",
       "      <td>a cohort of forty 30-weeks-old female BALB/cAn...</td>\n",
       "      <td>How many days after launch were the mice treat...</td>\n",
       "      <td>Six days after launch</td>\n",
       "      <td>How was the Earth recovery period for the subj...</td>\n",
       "      <td>Following their return to Earth, the experimen...</td>\n",
       "      <td>What specific type of samples and from which c...</td>\n",
       "      <td>GeneLab received RNA later preserved dorsal sk...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>The radiation bystander response is an importa...</td>\n",
       "      <td>https://sciencediscoveryengine.nasa.gov#/previ...</td>\n",
       "      <td>What is a key component of the overall respons...</td>\n",
       "      <td>The radiation bystander response</td>\n",
       "      <td>Which methods were used to measure the bystand...</td>\n",
       "      <td>using chromosomal aberration and micronucleus ...</td>\n",
       "      <td>What was shown to increase chromosomal structu...</td>\n",
       "      <td>depletion or mutation of RAD9</td>\n",
       "      <td>what cellular pathways were altered as a resul...</td>\n",
       "      <td>The depletion of RAD9 caused changes in numero...</td>\n",
       "      <td>How were differences in the activation of tran...</td>\n",
       "      <td>The use of network analysis allowed the discov...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Martian regolith (unconsolidated surface mater...</td>\n",
       "      <td>https://sciencediscoveryengine.nasa.gov#/previ...</td>\n",
       "      <td>What is considered a potential medium for plan...</td>\n",
       "      <td>Martian regolith (unconsolidated surface mater...</td>\n",
       "      <td>What is the level of hydrated magnesium sulfat...</td>\n",
       "      <td>hydrated magnesium sulfate mineral levels in t...</td>\n",
       "      <td>What was characterized in the study to identif...</td>\n",
       "      <td>The early Arabidopsis root transcriptome respo...</td>\n",
       "      <td>how was the genetic comparison approached to d...</td>\n",
       "      <td>A comparative analysis was conducted between t...</td>\n",
       "      <td>What distinctive characteristic does the cax1-...</td>\n",
       "      <td>The cax1-1 mutant line demonstrated relative t...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>The objective of the Rodent Research-23 missio...</td>\n",
       "      <td>https://sciencediscoveryengine.nasa.gov#/previ...</td>\n",
       "      <td>What was the objective of the Rodent Research-...</td>\n",
       "      <td>to better understand the effects of spacefligh...</td>\n",
       "      <td>How many mice and of what type were involved i...</td>\n",
       "      <td>twenty male, C57BL/6J, 16-17 weeks-old mice</td>\n",
       "      <td>How long were the mice in the RR-23 study main...</td>\n",
       "      <td>maintained in microgravity for 38 days</td>\n",
       "      <td>What was the procedure followed for the mice f...</td>\n",
       "      <td>The mice, after landing in the Atlantic, were ...</td>\n",
       "      <td>How was RNA extraction and analysis managed fo...</td>\n",
       "      <td>Adrenal glands of the mice were preserved by i...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                             context  \\\n",
       "0  Genome-wide transcriptional profiling showed t...   \n",
       "1  The objective of the Rodent Research-5 (RR-5) ...   \n",
       "2  The radiation bystander response is an importa...   \n",
       "3  Martian regolith (unconsolidated surface mater...   \n",
       "4  The objective of the Rodent Research-23 missio...   \n",
       "\n",
       "                                                 URL  \\\n",
       "0  https://sciencediscoveryengine.nasa.gov#/previ...   \n",
       "1  https://sciencediscoveryengine.nasa.gov#/previ...   \n",
       "2  https://sciencediscoveryengine.nasa.gov#/previ...   \n",
       "3  https://sciencediscoveryengine.nasa.gov#/previ...   \n",
       "4  https://sciencediscoveryengine.nasa.gov#/previ...   \n",
       "\n",
       "                                          Question 1  \\\n",
       "0  Where were the gene expression alterations in ...   \n",
       "1  What was the main goal of the Rodent Research-...   \n",
       "2  What is a key component of the overall respons...   \n",
       "3  What is considered a potential medium for plan...   \n",
       "4  What was the objective of the Rodent Research-...   \n",
       "\n",
       "                                   Question 1 Answer  \\\n",
       "0  reducing gravity levels in the International S...   \n",
       "1  to evaluate bone loss in mice during spaceflig...   \n",
       "2                   The radiation bystander response   \n",
       "3  Martian regolith (unconsolidated surface mater...   \n",
       "4  to better understand the effects of spacefligh...   \n",
       "\n",
       "                                          Question 2  \\\n",
       "0  What kind of experiments showed differential r...   \n",
       "1  What type of mice and how old were they when f...   \n",
       "2  Which methods were used to measure the bystand...   \n",
       "3  What is the level of hydrated magnesium sulfat...   \n",
       "4  How many mice and of what type were involved i...   \n",
       "\n",
       "                                   Question 2 Answer  \\\n",
       "0  simulation experiments on ground testing space...   \n",
       "1  a cohort of forty 30-weeks-old female BALB/cAn...   \n",
       "2  using chromosomal aberration and micronucleus ...   \n",
       "3  hydrated magnesium sulfate mineral levels in t...   \n",
       "4        twenty male, C57BL/6J, 16-17 weeks-old mice   \n",
       "\n",
       "                                          Question 3  \\\n",
       "0  What is overrepresented in the different exper...   \n",
       "1  How many days after launch were the mice treat...   \n",
       "2  What was shown to increase chromosomal structu...   \n",
       "3  What was characterized in the study to identif...   \n",
       "4  How long were the mice in the RR-23 study main...   \n",
       "\n",
       "                                   Question 3 Answer  \\\n",
       "0  the same GO groups including a large multigene...   \n",
       "1                              Six days after launch   \n",
       "2                      depletion or mutation of RAD9   \n",
       "3  The early Arabidopsis root transcriptome respo...   \n",
       "4             maintained in microgravity for 38 days   \n",
       "\n",
       "                                          Question 4  \\\n",
       "0  what analytical methodology provided integrati...   \n",
       "1  How was the Earth recovery period for the subj...   \n",
       "2  what cellular pathways were altered as a resul...   \n",
       "3  how was the genetic comparison approached to d...   \n",
       "4  What was the procedure followed for the mice f...   \n",
       "\n",
       "                                   Question 4 Answer  \\\n",
       "0  Self-organizing maps of the Gene Expression Dy...   \n",
       "1  Following their return to Earth, the experimen...   \n",
       "2  The depletion of RAD9 caused changes in numero...   \n",
       "3  A comparative analysis was conducted between t...   \n",
       "4  The mice, after landing in the Atlantic, were ...   \n",
       "\n",
       "                                          Question 5  \\\n",
       "0  What are the implications of the findings in t...   \n",
       "1  What specific type of samples and from which c...   \n",
       "2  How were differences in the activation of tran...   \n",
       "3  What distinctive characteristic does the cax1-...   \n",
       "4  How was RNA extraction and analysis managed fo...   \n",
       "\n",
       "                                   Question 5 Answer  \n",
       "0  The findings indicate that the finely adjusted...  \n",
       "1  GeneLab received RNA later preserved dorsal sk...  \n",
       "2  The use of network analysis allowed the discov...  \n",
       "3  The cax1-1 mutant line demonstrated relative t...  \n",
       "4  Adrenal glands of the mice were preserved by i...  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = load_data(FILE_PATH)\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "91c44192-e4d2-480e-9bfd-08f765b80822",
   "metadata": {},
   "source": [
    "## Data Peek\n",
    "\n",
    "Here we pick a sample row to inspect the nature of data.\n",
    "- What each column looks like"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "0fdfb6da-a192-4188-8ff6-fc0e895c5a0b",
   "metadata": {},
   "outputs": [],
   "source": [
    "def peek_data(data: pd.DataFrame, row_index: int = 0):\n",
    "    print(f\"Row = {row_index}\")\n",
    "    row = data.iloc[row_index]\n",
    "    for col, value in zip(data.columns, row):\n",
    "        display(HTML(f\"<h4>{col}</h4>\"))\n",
    "        print(value)\n",
    "        display(HTML(\"<hr/>\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "7f75d398-58ce-4b4e-acc6-3f08a7028074",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(60, 12)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "194713b7-dd62-46f2-a2c5-84458dbca3dd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Row = 0\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>context</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Genome-wide transcriptional profiling showed that reducing gravity levels in the International Space Station (ISS) causes important alterations in Drosophila gene expression intimately linked to imposed spaceflight-related environmental constrains during Drosophila metamorphosis. However, simulation experiments on ground testing space-related environmental constraints, show differential responses. Curiously, although particular genes are not common in the different experiments, the same GO groups including a large multigene family related with behavior, stress response and organogenesis are over represented in them. A global and integrative analysis using the gene expression dynamics inspector (GEDI) self-organizing maps, reveals different degrees in the responses of the transcriptome when using different environmental conditions or microgravity/hypergravity simulation devices. These results suggest that the transcriptome is finely tuned to normal gravity. In regular environmental conditions the alteration of this constant parameter on Earth can have mild effects on gene expression but when environmental conditions are far from optimal, the gene expression is much more intense and consistent effects.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>URL</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "https://sciencediscoveryengine.nasa.gov#/preview?id=%2FSDE%2Fgenelab_lsda%2F%7Chttps:%2F%2Fosdr.nasa.gov%2Fbio%2Frepo%2Fdata%2Fstudies%2FOSD-70%2F&query=%7B%22name%22:%22query-smd-primary%22,%22scope%22:%22All%22,%22text%22:%22osd-70%22%7D\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Question 1</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Where were the gene expression alterations in Drosophila observed during their metamorphosis?\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Question 1 Answer</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "reducing gravity levels in the International Space Station (ISS)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Question 2</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "What kind of experiments showed differential responses in gene expression?\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Question 2 Answer</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "simulation experiments on ground testing space-related environmental constraints\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Question 3</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "What is overrepresented in the different experiments despite particular genes not being common?\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Question 3 Answer</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the same GO groups including a large multigene family related with behavior, stress response and organogenesis\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Question 4</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "what analytical methodology provided integrative insights on transcriptome variations?\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Question 4 Answer</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Self-organizing maps of the Gene Expression Dynamics Inspector (GEDI) were utilized to reveal transcriptome variations under various environmental conditions or microgravity/hypergravity simulation devices.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Question 5</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "What are the implications of the findings in terms of the tuning of the transcriptome to Earth's gravity and its variations in extreme environmental conditions?\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Question 5 Answer</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The findings indicate that the finely adjusted transcriptome is tuned to normal Earth gravity, showing a more marked and coherent genetic response when environmental conditions significantly deviate from optimal.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# index = random.randint(0, len(data))\n",
    "peek_data(data, row_index=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6d907c7a-b5dd-4021-b3e6-f2559ad0999d",
   "metadata": {},
   "source": [
    "## Simplify Data\n",
    "\n",
    "Here we separate out individual question/answer pair into separate row for each context.\n",
    "\n",
    "\n",
    "This is used for simplifying the validation process so that each row would have consistent view of data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "9e51eb58-ad91-485a-8eae-a29e409c660d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>context</th>\n",
       "      <th>context_hash</th>\n",
       "      <th>url</th>\n",
       "      <th>question</th>\n",
       "      <th>answer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Genome-wide transcriptional profiling showed t...</td>\n",
       "      <td>27905003</td>\n",
       "      <td>https://sciencediscoveryengine.nasa.gov#/previ...</td>\n",
       "      <td>Where were the gene expression alterations in ...</td>\n",
       "      <td>reducing gravity levels in the International S...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Genome-wide transcriptional profiling showed t...</td>\n",
       "      <td>27905003</td>\n",
       "      <td>https://sciencediscoveryengine.nasa.gov#/previ...</td>\n",
       "      <td>What kind of experiments showed differential r...</td>\n",
       "      <td>simulation experiments on ground testing space...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Genome-wide transcriptional profiling showed t...</td>\n",
       "      <td>27905003</td>\n",
       "      <td>https://sciencediscoveryengine.nasa.gov#/previ...</td>\n",
       "      <td>What is overrepresented in the different exper...</td>\n",
       "      <td>the same GO groups including a large multigene...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Genome-wide transcriptional profiling showed t...</td>\n",
       "      <td>27905003</td>\n",
       "      <td>https://sciencediscoveryengine.nasa.gov#/previ...</td>\n",
       "      <td>what analytical methodology provided integrati...</td>\n",
       "      <td>Self-organizing maps of the Gene Expression Dy...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Genome-wide transcriptional profiling showed t...</td>\n",
       "      <td>27905003</td>\n",
       "      <td>https://sciencediscoveryengine.nasa.gov#/previ...</td>\n",
       "      <td>What are the implications of the findings in t...</td>\n",
       "      <td>The findings indicate that the finely adjusted...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>The objective of the Rodent Research-5 (RR-5) ...</td>\n",
       "      <td>14559633</td>\n",
       "      <td>https://sciencediscoveryengine.nasa.gov#/previ...</td>\n",
       "      <td>What was the main goal of the Rodent Research-...</td>\n",
       "      <td>to evaluate bone loss in mice during spaceflig...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>The objective of the Rodent Research-5 (RR-5) ...</td>\n",
       "      <td>14559633</td>\n",
       "      <td>https://sciencediscoveryengine.nasa.gov#/previ...</td>\n",
       "      <td>What type of mice and how old were they when f...</td>\n",
       "      <td>a cohort of forty 30-weeks-old female BALB/cAn...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>The objective of the Rodent Research-5 (RR-5) ...</td>\n",
       "      <td>14559633</td>\n",
       "      <td>https://sciencediscoveryengine.nasa.gov#/previ...</td>\n",
       "      <td>How many days after launch were the mice treat...</td>\n",
       "      <td>Six days after launch</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>The objective of the Rodent Research-5 (RR-5) ...</td>\n",
       "      <td>14559633</td>\n",
       "      <td>https://sciencediscoveryengine.nasa.gov#/previ...</td>\n",
       "      <td>How was the Earth recovery period for the subj...</td>\n",
       "      <td>Following their return to Earth, the experimen...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>The objective of the Rodent Research-5 (RR-5) ...</td>\n",
       "      <td>14559633</td>\n",
       "      <td>https://sciencediscoveryengine.nasa.gov#/previ...</td>\n",
       "      <td>What specific type of samples and from which c...</td>\n",
       "      <td>GeneLab received RNA later preserved dorsal sk...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                             context  context_hash  \\\n",
       "0  Genome-wide transcriptional profiling showed t...      27905003   \n",
       "1  Genome-wide transcriptional profiling showed t...      27905003   \n",
       "2  Genome-wide transcriptional profiling showed t...      27905003   \n",
       "3  Genome-wide transcriptional profiling showed t...      27905003   \n",
       "4  Genome-wide transcriptional profiling showed t...      27905003   \n",
       "5  The objective of the Rodent Research-5 (RR-5) ...      14559633   \n",
       "6  The objective of the Rodent Research-5 (RR-5) ...      14559633   \n",
       "7  The objective of the Rodent Research-5 (RR-5) ...      14559633   \n",
       "8  The objective of the Rodent Research-5 (RR-5) ...      14559633   \n",
       "9  The objective of the Rodent Research-5 (RR-5) ...      14559633   \n",
       "\n",
       "                                                 url  \\\n",
       "0  https://sciencediscoveryengine.nasa.gov#/previ...   \n",
       "1  https://sciencediscoveryengine.nasa.gov#/previ...   \n",
       "2  https://sciencediscoveryengine.nasa.gov#/previ...   \n",
       "3  https://sciencediscoveryengine.nasa.gov#/previ...   \n",
       "4  https://sciencediscoveryengine.nasa.gov#/previ...   \n",
       "5  https://sciencediscoveryengine.nasa.gov#/previ...   \n",
       "6  https://sciencediscoveryengine.nasa.gov#/previ...   \n",
       "7  https://sciencediscoveryengine.nasa.gov#/previ...   \n",
       "8  https://sciencediscoveryengine.nasa.gov#/previ...   \n",
       "9  https://sciencediscoveryengine.nasa.gov#/previ...   \n",
       "\n",
       "                                            question  \\\n",
       "0  Where were the gene expression alterations in ...   \n",
       "1  What kind of experiments showed differential r...   \n",
       "2  What is overrepresented in the different exper...   \n",
       "3  what analytical methodology provided integrati...   \n",
       "4  What are the implications of the findings in t...   \n",
       "5  What was the main goal of the Rodent Research-...   \n",
       "6  What type of mice and how old were they when f...   \n",
       "7  How many days after launch were the mice treat...   \n",
       "8  How was the Earth recovery period for the subj...   \n",
       "9  What specific type of samples and from which c...   \n",
       "\n",
       "                                              answer  \n",
       "0  reducing gravity levels in the International S...  \n",
       "1  simulation experiments on ground testing space...  \n",
       "2  the same GO groups including a large multigene...  \n",
       "3  Self-organizing maps of the Gene Expression Dy...  \n",
       "4  The findings indicate that the finely adjusted...  \n",
       "5  to evaluate bone loss in mice during spaceflig...  \n",
       "6  a cohort of forty 30-weeks-old female BALB/cAn...  \n",
       "7                              Six days after launch  \n",
       "8  Following their return to Earth, the experimen...  \n",
       "9  GeneLab received RNA later preserved dorsal sk...  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = simplify_columns(data)\n",
    "data.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "78a5cc46-2659-4b89-9cf0-f3fa7118fbbe",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(295, 5)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "8b3b805c-a5b2-46a1-b6e9-18acbcca0939",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Row = 0\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>context</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Genome-wide transcriptional profiling showed that reducing gravity levels in the International Space Station (ISS) causes important alterations in Drosophila gene expression intimately linked to imposed spaceflight-related environmental constrains during Drosophila metamorphosis. However, simulation experiments on ground testing space-related environmental constraints, show differential responses. Curiously, although particular genes are not common in the different experiments, the same GO groups including a large multigene family related with behavior, stress response and organogenesis are over represented in them. A global and integrative analysis using the gene expression dynamics inspector (GEDI) self-organizing maps, reveals different degrees in the responses of the transcriptome when using different environmental conditions or microgravity/hypergravity simulation devices. These results suggest that the transcriptome is finely tuned to normal gravity. In regular environmental conditions the alteration of this constant parameter on Earth can have mild effects on gene expression but when environmental conditions are far from optimal, the gene expression is much more intense and consistent effects.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>context_hash</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "27905003\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>url</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "https://sciencediscoveryengine.nasa.gov#/preview?id=%2FSDE%2Fgenelab_lsda%2F%7Chttps:%2F%2Fosdr.nasa.gov%2Fbio%2Frepo%2Fdata%2Fstudies%2FOSD-70%2F&query=%7B%22name%22:%22query-smd-primary%22,%22scope%22:%22All%22,%22text%22:%22osd-70%22%7D\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>question</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Where were the gene expression alterations in Drosophila observed during their metamorphosis?\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>answer</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "reducing gravity levels in the International Space Station (ISS)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "peek_data(data, row_index=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "16d87194-9768-4015-8103-6bfc2d5e89e6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Row = 1\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>context</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Genome-wide transcriptional profiling showed that reducing gravity levels in the International Space Station (ISS) causes important alterations in Drosophila gene expression intimately linked to imposed spaceflight-related environmental constrains during Drosophila metamorphosis. However, simulation experiments on ground testing space-related environmental constraints, show differential responses. Curiously, although particular genes are not common in the different experiments, the same GO groups including a large multigene family related with behavior, stress response and organogenesis are over represented in them. A global and integrative analysis using the gene expression dynamics inspector (GEDI) self-organizing maps, reveals different degrees in the responses of the transcriptome when using different environmental conditions or microgravity/hypergravity simulation devices. These results suggest that the transcriptome is finely tuned to normal gravity. In regular environmental conditions the alteration of this constant parameter on Earth can have mild effects on gene expression but when environmental conditions are far from optimal, the gene expression is much more intense and consistent effects.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>context_hash</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "27905003\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>url</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "https://sciencediscoveryengine.nasa.gov#/preview?id=%2FSDE%2Fgenelab_lsda%2F%7Chttps:%2F%2Fosdr.nasa.gov%2Fbio%2Frepo%2Fdata%2Fstudies%2FOSD-70%2F&query=%7B%22name%22:%22query-smd-primary%22,%22scope%22:%22All%22,%22text%22:%22osd-70%22%7D\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>question</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "What kind of experiments showed differential responses in gene expression?\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>answer</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "simulation experiments on ground testing space-related environmental constraints\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "peek_data(data, row_index=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bee52cc4-5257-47d5-9141-a82ac02ae9e2",
   "metadata": {},
   "source": [
    "# promptlab \n",
    "\n",
    "\n",
    "**promptlab** provides a programmatic approach to interact with the bot via its REST API.\n",
    "Specificially, the `/process/` endpoint can process any query and generate the respond accordingly.\n",
    "https://api.promptlab.nasa-impact.net/api/v1/process/\n",
    "\n",
    "We can send a request to the API with the query/question and get the response.\n",
    "\n",
    "Here, the `FlowRunner` class abstracts everything. Its `.run(...)` method processes the query which:\n",
    "- Takes in a query string\n",
    "- Returns a result json/dictionary\n",
    "\n",
    "The result from the API tentatively gives us following information:\n",
    "\n",
    "- *question* : input question fed into the bot\n",
    "- *answer* : generated answer based on retrieved contexts\n",
    "- *retrieved contexts* : List of **top-k** contexts retrieved by the RAG system"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "783bcbf4-b439-4045-b12f-80c4f5f545e0",
   "metadata": {},
   "source": [
    "## setup flow processors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "9b91d48d-1c8b-4572-a5f8-b1f04cca610b",
   "metadata": {},
   "outputs": [],
   "source": [
    "import requests\n",
    "from typing import Optional, Union\n",
    "import urllib.parse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "bd72e375-e9e6-438a-b5b0-ceb4d53d8eab",
   "metadata": {},
   "outputs": [],
   "source": [
    "from loguru import logger"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "3822fbeb-a9f7-4550-9411-85d1b24f33b8",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tenacity import retry, retry_if_result, stop_after_attempt, wait_fixed, retry_if_exception, retry_if_exception_type"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "b630a0a7-9552-483b-aaa2-ed1f5ac194fd",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Setup the base URL to point to the production\n",
    "BASE_API_URL = \"https://api.promptlab.nasa-impact.net/api/v1/process/\"\n",
    "\n",
    "# Setup api key to authenticate\n",
    "\n",
    "PROMPTLAB_API_KEY = os.environ[\"PROMPTLAB_API_KEY\"]\n",
    "assert PROMPTLAB_API_KEY"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "4027554b-6321-4c7f-bd63-edfca390c7e1",
   "metadata": {},
   "outputs": [],
   "source": [
    "# tweaks for the OSDR bot flow\n",
    "TWEAKS = {\n",
    "  \"ConversationalRetrievalChain-83vns\": {},\n",
    "  \"ChatOpenAI-gsfUD\": {\"model\": \"gpt-3.5-turbo\"},\n",
    "  \"ConversationBufferMemory-z2ZqE\": {},\n",
    "  \"Sinequa-Z8S73\": {}\n",
    "}\n",
    "\n",
    "# flow id of the OSDR bot in promptlab that is used for QA programmatically\n",
    "FLOW_ID = \"1edbe0f9-d7ff-4c2a-a371-5f6312964f05\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "cdc7b670-02a4-4974-ab57-3b7d0ac9baf0",
   "metadata": {},
   "outputs": [],
   "source": [
    "class FlowRunner:\n",
    "    def __init__(self, base_url: str, api_key: str, debug:bool = False) -> None:\n",
    "        self.base_url = base_url\n",
    "        self.api_key = api_key\n",
    "        self.debug=bool(debug)\n",
    "\n",
    "    def run_flow(\n",
    "        self,\n",
    "        inputs: dict,\n",
    "        flow_id: str,\n",
    "        tweaks: Optional[dict] = None,\n",
    "    ) -> dict:\n",
    "        \"\"\"\n",
    "        Run a flow with a given message and optional tweaks.\n",
    "    \n",
    "        :param message: The message to send to the flow\n",
    "        :param flow_id: The ID of the flow to run\n",
    "        :param tweaks: Optional tweaks to customize the flow\n",
    "        :return: The JSON response from the flow\n",
    "        \"\"\"\n",
    "        api_url = urllib.parse.urljoin(self.base_url, flow_id)\n",
    "        if self.debug:\n",
    "            logger.debug(f\"URL = {api_url}\")\n",
    "\n",
    "        api_key = self.api_key\n",
    "        payload = {\"inputs\": inputs}\n",
    "        headers = None\n",
    "        if tweaks:\n",
    "            payload[\"tweaks\"] = tweaks\n",
    "        if api_key:\n",
    "            headers = {\"x-api-key\": api_key}\n",
    "        response = requests.post(api_url, json=payload, headers=headers)\n",
    "        if self.debug:\n",
    "            logger.debug(f\"response = {response}\")\n",
    "            logger.debug(f\"status code = {response.status_code}\")\n",
    "        if response.status_code != 200:\n",
    "            raise requests.exceptions.HTTPError(f\"Error: {response.status_code}\")\n",
    "        return response.json()\n",
    "\n",
    "    @staticmethod\n",
    "    def if_not_valid_result(result: dict) -> bool:\n",
    "        return not (isinstance(result, dict) and \"result\" in result and \"answer\" in result.get(\"result\", {}))\n",
    "    \n",
    "    @retry(\n",
    "        retry=(retry_if_exception_type(requests.exceptions.RequestException) | retry_if_result(if_not_valid_result)),\n",
    "        stop=stop_after_attempt(3),\n",
    "        wait=wait_fixed(2)\n",
    "    )\n",
    "    def run(\n",
    "        self,\n",
    "        input: Union[dict, str],\n",
    "        flow_id: str,\n",
    "        tweaks: Optional[dict] = None,\n",
    "    ) -> dict:\n",
    "        if isinstance(input, str):\n",
    "            input = dict(question=input)\n",
    "        result = self.run_flow(\n",
    "            inputs=input,\n",
    "            flow_id=flow_id,\n",
    "            tweaks=tweaks,\n",
    "        )\n",
    "        result[\"question\"] = input[\"question\"] if isinstance(input, dict) else input\n",
    "\n",
    "        if \"result\" in result:\n",
    "            source_documents = result.get(\"result\", {}).get(\"source_documents\", \"[]\")\n",
    "            if isinstance(source_documents, str):\n",
    "                result[\"result\"][\"source_documents\"] = json.loads(source_documents)\n",
    "        return result\n",
    "\n",
    "    @staticmethod\n",
    "    def display_result(result: dict, question: Optional[str]=None, reference_answer=None):\n",
    "        header_tag = \"h4\"\n",
    "        question = result.get(\"question\", question)\n",
    "        if question:\n",
    "            display(HTML(f\"<{header_tag}>Question</{header_tag}>\"))\n",
    "            print(question)\n",
    "            display(HTML(\"<hr/>\"))\n",
    "\n",
    "        if \"result\" in result:\n",
    "            result = result[\"result\"]\n",
    "        display(HTML(f\"<{header_tag}>Generated Answer</{header_tag}>\"))\n",
    "        print(result[\"answer\"])\n",
    "        display(HTML(\"<hr/>\"))\n",
    "\n",
    "        if reference_answer:\n",
    "            display(HTML(f\"<{header_tag}>Exepcted Answer (Reference/Ground Truth)</{header_tag}>\"))\n",
    "            print(reference_answer)\n",
    "            display(HTML(\"<hr/>\"))\n",
    "        \n",
    "        # display(HTML(f\"<{header_tag}>Retrieved Contexts</{header_tag}>\"))\n",
    "        for idx, doc in enumerate(result[\"source_documents\"], start=1):\n",
    "            display(HTML(f\"<{header_tag}>Retrieved Context #{idx}</{header_tag}>\"))\n",
    "            print(doc[\"page_content\"])\n",
    "            print(doc[\"metadata\"][\"source\"])\n",
    "            display(HTML(\"<hr/>\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a0a6cd33-17a9-4f76-9b30-aac4b726ddcd",
   "metadata": {},
   "source": [
    "## Test Run\n",
    "\n",
    "Now, let's check what a sample test run looks like when we call the API for QA."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "bef70819-ac3d-43a0-a3a4-b6af8f6ba2f9",
   "metadata": {},
   "outputs": [],
   "source": [
    "flow_runner = FlowRunner(\n",
    "    base_url=BASE_API_URL,\n",
    "    api_key=PROMPTLAB_API_KEY\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "ee541856-c87c-4dc7-85ce-26ef6e3cdd73",
   "metadata": {},
   "outputs": [],
   "source": [
    "query = \"What parasites were sent to ISS to study host immunity?\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "197ff92a-0b01-4a79-9355-02af414e6d6c",
   "metadata": {},
   "outputs": [],
   "source": [
    "result = flow_runner.run(\n",
    "    input=query,\n",
    "    flow_id=FLOW_ID,\n",
    "    tweaks=TWEAKS,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "a0f6565d-cbfa-430f-9067-f429001e797f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'What parasites were sent to ISS to study host immunity?'"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result[\"question\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "340f1781-096f-485e-a8f4-ea48365f741d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<h4>Question</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "What parasites were sent to ISS to study host immunity?\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Generated Answer</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "In the study mentioned, fruit flies and their parasitic wasps were sent to the ISS to examine changes in host immunity and parasite virulence. The specific parasites mentioned in the study are Leptopilina boulardi 17 and Leptopilina heterotoma 14.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #1</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "In this <b>study</b>, fruit flies and their parasitic wasps were <b>sent</b> to the <b>ISS</b> to examine changes in <b>host</b> <b>immunity</b> and <b>parasite</b> virulence. Key findings from this work are as follows: (a) Spaceflight spiked the expression of inflammation genes in fruit flies; (b) Spaceflight enhanced tumor growth in flies already suffering from chronic inflammation; (c) Spaceflight did not significantly affect parasitic wasp development, emergence, or the expression of virulence genes; (d) <b>Parasite</b> mutants with visible wing color and shape alterations were obtained. This <b>study</b> contains Drosophila melanogaster data.\n",
      "https://osdr.nasa.gov/bio/repo/data/studies/OSD-588/\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #2</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "In this <b>study</b>, fruit flies and their parasitic wasps were <b>sent</b> to the <b>ISS</b> to examine changes in <b>host</b> <b>immunity</b> and <b>parasite</b> virulence. Key findings from this work are as follows: (a) Spaceflight spiked the expression of inflammation genes in fruit flies; (b) Spaceflight enhanced tumor growth in flies already suffering from chronic inflammation; (c) Spaceflight did not significantly affect parasitic wasp development, emergence, or the expression of virulence genes; (d) <b>Parasite</b> mutants with visible wing color and shape alterations were obtained. This <b>study</b> contains Leptopilina boulardi 17 data.\n",
      "https://osdr.nasa.gov/bio/repo/data/studies/OSD-610/\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #3</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "In this <b>study</b>, fruit flies and their parasitic wasps were <b>sent</b> to the <b>ISS</b> to examine changes in <b>host</b> <b>immunity</b> and <b>parasite</b> virulence. Key findings from this work are as follows: (a) Spaceflight spiked the expression of inflammation genes in fruit flies; (b) Spaceflight enhanced tumor growth in flies already suffering from chronic inflammation; (c) Spaceflight did not significantly affect parasitic wasp development, emergence, or the expression of virulence genes; (d) <b>Parasite</b> mutants with visible wing color and shape alterations were obtained. This <b>study</b> contains Leptopilina heterotoma14 data.\n",
      "https://osdr.nasa.gov/bio/repo/data/studies/OSD-609/\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #4</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "About  Charter  How to Join  Current AWG Members  Citizen Science  SOLSTICE  SOLSTICE Members  FAQ  Terms & Conditions  Genelab Tutorials  Contact Us  account_circle expand_more  1.26 TB  OSD-588 Version 1  Drosophila parasitoids go to space: Unexpected effects of spaceflight on <b>hosts</b> and their parasitoids - Drosophila data <b>Study</b>  Submitted Date: 02-Feb-2023  Initial Release Date: 02-Jan-2024  notifications share  GeneLab ID : GLDS-583 DOI : 10.26030/v9rh-5a70  Cite this <b>Study</b>  info Description  Description  In this <b>study</b>, fruit flies and their parasitic wasps were <b>sent</b> to the <b>ISS</b> to examine changes in <b>host</b> <b>immunity</b> and <b>parasite</b> virulence.\n",
      "https://osdr.nasa.gov/bio/repo/data/studies/OSD-588/\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #5</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "About  Charter  How to Join  Current AWG Members  Citizen Science  SOLSTICE  SOLSTICE Members  FAQ  Terms & Conditions  Genelab Tutorials  Contact Us  account_circle expand_more  148.92 GB  OSD-610 Version 1  Drosophila parasitoids go to space: Unexpected and differential effects of spaceflight on <b>hosts</b> and their parasitoids - Wasp Leptopilina boulardi 17 data <b>Study</b>  Submitted Date: 28-Apr-2023  Initial Release Date: 02-Jan-2024  notifications share  GeneLab ID : GLDS-587 DOI : 10.26030/9ee4-6s36  Cite this <b>Study</b>  info Description  Description  In this <b>study</b>, fruit flies and their parasitic wasps were <b>sent</b> to the <b>ISS</b> to examine changes in <b>host</b> <b>immunity</b> and <b>parasite</b> virulence.\n",
      "https://osdr.nasa.gov/bio/repo/data/studies/OSD-610/\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #6</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "About  Charter  How to Join  Current AWG Members  Citizen Science  SOLSTICE  SOLSTICE Members  FAQ  Terms & Conditions  Genelab Tutorials  Contact Us  account_circle expand_more  176.26 GB  OSD-609 Version 1  Drosophila parasitoids go to space: Unexpected and differential effects of spaceflight on <b>hosts</b> and their parasitoids - Wasp Leptopilina heterotoma 14 data <b>Study</b>  Submitted Date: 28-Apr-2023  Initial Release Date: 02-Jan-2024  notifications share  GeneLab ID : GLDS-586 DOI : 10.26030/5rjq-a347  Cite this <b>Study</b>  info Description  Description  In this <b>study</b>, fruit flies and their parasitic wasps were <b>sent</b> to the <b>ISS</b> to examine changes in <b>host</b> <b>immunity</b> and <b>parasite</b> virulence.\n",
      "https://osdr.nasa.gov/bio/repo/data/studies/OSD-609/\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #7</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The biology chosen for this experiment was Arabidopsis thaliana, ecotype Col-0, because of the plant's history of spaceflight experimentation within KFTs and wealth of comparative data from orbital experiments. KFTs were deployed as a wearable device, a leg pouch attached to the astronaut, which proved to be operationally effective during the course of the flight. Data from the inflight samples indicated that the microgravity period of the flight elicited the strongest transcriptomic responses as measured by the number of genes showing differential expression.\n",
      "https://osdr.nasa.gov/bio/repo/data/studies/OSD-624/\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #8</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fungal Pathogenesis , Tumorigenesis , and Effects of <b>Host</b> <b>Immunity</b> in Space  Project Type  Spaceflight <b>Study</b>  Flight Program  Space Transportation System (STS)  Experiment Platform  Shuttle Life Sciences Research (Middeck)  Sponsoring Agency  NASA  NASA Center  Ames Research Center (ARC)  Funding Source  This <b>study</b> work was supported by a NASA grant to SB for Fungal Pathogenesis, Tumorigenesis, and Effects of <b>Host</b> <b>Immunity</b> in the Space Environment (FSB-FIT-2005). The funder had a role in <b>study</b> design and in supporting data collection.  Acknowledgments  We thank David Schneider for helping us with the design of this experiment.\n",
      "https://osdr.nasa.gov/bio/repo/data/studies/OSD-3/\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #9</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Spaceflight uniquely alters the physiology of both human cells and microbial pathogens, stimulating cellular and molecular changes directly relevant to infectious disease. However, the influence of this environment on host-pathogen interactions remains poorly understood. Here we report our results from the STL-IMMUNE study flown aboard Space Shuttle mission STS-131, which investigated multi-omic responses (transcriptomic, proteomic) of human intestinal epithelial cells to infection with Salmonella Typhimurium when both host and pathogen were simultaneously exposed to spaceflight. To our knowledge, this was the first in-flight infection and dual RNA-seq analysis using human cells.\n",
      "https://osdr.nasa.gov/bio/repo/data/studies/OSD-323/\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #10</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Immune dysregulation is among the main adverse outcomes of spaceflight. Despite its crucial role in host protection, effects on the human antibody repertoire are unknown. Consequently, we examined the IgM repertoire of five cosmonauts 25 days before launch, after 64 and 129 days spent on the International Space Station (ISS), and after 1, 7 and 30 days of Earth reambulation using high-throughput sequencing. This is the first study of this kind.\n",
      "https://osdr.nasa.gov/bio/repo/data/studies/OSD-316/\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "flow_runner.display_result(result)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1145101f-e93c-44ef-a493-1d224e1dbf90",
   "metadata": {},
   "source": [
    "# Run OSDR Validation set\n",
    "\n",
    "Instead of manually running each question through the bot to generate the answer, we run all the questions from the validation data in bulk.\n",
    "\n",
    "We have some functions for housekeeping and the details for them are hidden away for brevity.\n",
    "\n",
    "> Note: Since it takes some time to run through all the questions, we'll pre-load the responses from a csv."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "a481fd4b-e50f-491e-8346-ebea7d26d5e0",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tqdm import tqdm\n",
    "from joblib import Parallel, delayed\n",
    "import ast"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "2c504385-1609-48a2-aad9-d02a48e71f4c",
   "metadata": {},
   "outputs": [],
   "source": [
    "def process_row(row, flow_runner, flow_id, tweaks):\n",
    "    try:\n",
    "        result = flow_runner.run(input=row.question, flow_id=flow_id, tweaks=tweaks)\n",
    "    except:\n",
    "        logger.error(f\"Error for question={row.question}. Skipping\")\n",
    "        result = {}\n",
    "    print(result.keys())\n",
    "    result[\"reference_context\"] = row.context\n",
    "    result[\"reference_answer\"] = row.answer\n",
    "    result[\"context_hash\"] = row.context_hash\n",
    "    return result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "1ba8708a-2bc7-4c96-9178-10a7280680cc",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_osdr_responses(\n",
    "    data: pd.DataFrame,\n",
    "    flow_runner: FlowRunner,\n",
    "    flow_id,\n",
    "    tweaks\n",
    ") -> pd.DataFrame:\n",
    "    results = []\n",
    "    for row in tqdm(data.itertuples()):\n",
    "        result = process_row(row, flow_runner, flow_id, tweaks)\n",
    "        results.append(result)\n",
    "    return pd.DataFrame(results)\n",
    "\n",
    "def get_osdr_responses_parallel(\n",
    "    data: pd.DataFrame,\n",
    "    flow_runner,\n",
    "    flow_id, \n",
    "    tweaks\n",
    ") -> pd.DataFrame:\n",
    "    results = Parallel(n_jobs=-1, backend=\"threading\")(\n",
    "        delayed(process_row)\\\n",
    "        (row, flow_runner, flow_id, tweaks) for row in tqdm(data.itertuples(index=False))\n",
    "    )\n",
    "    results_df = pd.DataFrame(results)\n",
    "    return results_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "55acaec1-1d4d-4845-bc00-9459dbab6a88",
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_responses(path: str) -> pd.DataFrame:\n",
    "    df = pd.read_csv(path)\n",
    "    df[\"result\"] = df[\"result\"].apply(ast.literal_eval)\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "1fedaca6-2564-4d89-aaa5-d666e9fce707",
   "metadata": {},
   "outputs": [],
   "source": [
    "# # get all the responses\n",
    "# responses = get_osdr_responses_parallel(\n",
    "#     data=simplify_columns(load_data(FILE_PATH)).iloc[:],\n",
    "#     flow_runner=flow_runner,\n",
    "#     flow_id=FLOW_ID,\n",
    "#     tweaks=TWEAKS\n",
    "# )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "8854fe51-db69-4821-9871-fced425035a8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# load already dumped responses\n",
    "responses = load_responses(\"../data/osdr-eval/responses-gpt-3.5-turbo.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "8a5746d6-c365-48e4-9323-02514e65b77d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>result</th>\n",
       "      <th>status</th>\n",
       "      <th>task</th>\n",
       "      <th>session_id</th>\n",
       "      <th>backend</th>\n",
       "      <th>question</th>\n",
       "      <th>reference_context</th>\n",
       "      <th>reference_answer</th>\n",
       "      <th>context_hash</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>{'answer': 'The gene expression alterations in...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'id': '139756224681744', 'href': 'api/v1/task...</td>\n",
       "      <td>RG8XKS:6b5b553ba5e766ccbc1429e4540911c1e7ac597...</td>\n",
       "      <td>anyio</td>\n",
       "      <td>Where were the gene expression alterations in ...</td>\n",
       "      <td>Genome-wide transcriptional profiling showed t...</td>\n",
       "      <td>reducing gravity levels in the International S...</td>\n",
       "      <td>30225908</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>{'answer': 'Experiments involving reducing gra...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'id': '139756227007232', 'href': 'api/v1/task...</td>\n",
       "      <td>JHTOXO:6b5b553ba5e766ccbc1429e4540911c1e7ac597...</td>\n",
       "      <td>anyio</td>\n",
       "      <td>What kind of experiments showed differential r...</td>\n",
       "      <td>Genome-wide transcriptional profiling showed t...</td>\n",
       "      <td>simulation experiments on ground testing space...</td>\n",
       "      <td>30225908</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>{'answer': 'The same GO groups, including a la...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'id': '139756223614704', 'href': 'api/v1/task...</td>\n",
       "      <td>VWW7ZL:6b5b553ba5e766ccbc1429e4540911c1e7ac597...</td>\n",
       "      <td>anyio</td>\n",
       "      <td>What is overrepresented in the different exper...</td>\n",
       "      <td>Genome-wide transcriptional profiling showed t...</td>\n",
       "      <td>the same GO groups including a large multigene...</td>\n",
       "      <td>30225908</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              result  status  \\\n",
       "0  {'answer': 'The gene expression alterations in...     NaN   \n",
       "1  {'answer': 'Experiments involving reducing gra...     NaN   \n",
       "2  {'answer': 'The same GO groups, including a la...     NaN   \n",
       "\n",
       "                                                task  \\\n",
       "0  {'id': '139756224681744', 'href': 'api/v1/task...   \n",
       "1  {'id': '139756227007232', 'href': 'api/v1/task...   \n",
       "2  {'id': '139756223614704', 'href': 'api/v1/task...   \n",
       "\n",
       "                                          session_id backend  \\\n",
       "0  RG8XKS:6b5b553ba5e766ccbc1429e4540911c1e7ac597...   anyio   \n",
       "1  JHTOXO:6b5b553ba5e766ccbc1429e4540911c1e7ac597...   anyio   \n",
       "2  VWW7ZL:6b5b553ba5e766ccbc1429e4540911c1e7ac597...   anyio   \n",
       "\n",
       "                                            question  \\\n",
       "0  Where were the gene expression alterations in ...   \n",
       "1  What kind of experiments showed differential r...   \n",
       "2  What is overrepresented in the different exper...   \n",
       "\n",
       "                                   reference_context  \\\n",
       "0  Genome-wide transcriptional profiling showed t...   \n",
       "1  Genome-wide transcriptional profiling showed t...   \n",
       "2  Genome-wide transcriptional profiling showed t...   \n",
       "\n",
       "                                    reference_answer  context_hash  \n",
       "0  reducing gravity levels in the International S...      30225908  \n",
       "1  simulation experiments on ground testing space...      30225908  \n",
       "2  the same GO groups including a large multigene...      30225908  "
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "responses.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "117d7ed2-cc9e-441a-9dcf-baa909843a4a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# responses.to_csv(\"data/responses-gpt-3.5-turbo.csv\", index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "db8843f3-6983-42be-8644-b6b719f4eeaa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<h4>Question</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Were male and female mice used?\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Generated Answer</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Yes, both male and female mice were used in the studies mentioned.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Exepcted Answer (Reference/Ground Truth)</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Male mice were used in this study.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #1</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Twenty BALB/c, 12-weeks old <b>female</b> <b>mice</b> (ten controls and ten treated) were flown to the ISS and housed in the Rodent Habitat for 39-42 days. Twenty <b>mice</b> of similar age, and matching sex and strain were <b>used</b> for ground controls housed in identical hardware and matching ISS environmental conditions. Basal controls were housed in standard vivarium cages. Spaceflight, ground controls and basal groups had blood collected, then were euthanized, had one hind limb removed, and finally whole carcasses were stored at -80 C until dissection. All <b>mice</b> in this data set received only the control/sham injection.\n",
      "https://osdr.nasa.gov/bio/repo/data/studies/OSD-352/\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #2</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Twenty BALB/c, 12-weeks old <b>female</b> <b>mice</b> (ten controls and ten treated) were flown to the ISS and housed in the Rodent Habitat for 39-42 days. Twenty <b>mice</b> of similar age, sex and strain were <b>used</b> for ground controls housed in identical hardware and matching ISS environmental conditions. Basal controls were housed in standard vivarium cages. Spaceflight, ground controls and basal groups had blood collected, then were euthanized, had one hind limb removed, and finally whole carcasses were stored at -80 C until dissection. All <b>mice</b> in this data set received only the control/sham injection.  Factor(s)  Factor\n",
      "https://osdr.nasa.gov/bio/repo/data/studies/OSD-162/\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #3</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Twenty BALB/c, 12-weeks old <b>female</b> <b>mice</b> (ten controls and ten treated) were flown to the ISS and housed in the Rodent Habitat for 39-42 days. Twenty <b>mice</b> of similar age, sex and strain were <b>used</b> for ground controls housed in identical hardware and matching ISS environmental conditions. Basal controls were housed in standard vivarium cages. Spaceflight, ground controls and basal groups had blood collected, then were euthanized, had one hind limb removed, and finally whole carcasses were stored at -80 C until dissection. All <b>mice</b> in this data set received only the control/sham injection.\n",
      "https://osdr.nasa.gov/bio/repo/data/studies/OSD-270/\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #4</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Previous studies of rodents exposed to a single such stressor document significant deficits, but our study is the first to investigate possible cumulative and synergistic impacts of simultaneous ionizing radiation, confinement, and altered gravity on behavior and cognition. Our cohort was divided between 6-month-old <b>female</b> and <b>male</b> <b>mice</b> in group, social isolation, or hindlimb unloading housing, exposed to 0 or 50 cGy of 5 ion simplified simulated galactic cosmic radiation (GCRsim). We report interactions and independent effects of GCRsim exposure and housing conditions on behavioral and cognitive performance.\n",
      "https://osdr.nasa.gov/bio/repo/data/studies/OSD-640/\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #5</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cosmic radiation (GCRsim). We report interactions and independent effects of GCRsim exposure and housing conditions on behavioral and cognitive performance. Exposure to GCRsim drove changes in immune cell populations in peripheral blood collected early after irradiation, while housing conditions drove changes in blood collected at a later point. <b>Female</b> <b>mice</b> were largely resilient to deficits observed in <b>male</b> <b>mice</b>. Finally, we <b>used</b> principal component analysis to represent total deficits as principal component scores, which were predicted by general linear models <b>using</b> GCR exposure, housing condition, and early blood biomarkers.\n",
      "https://osdr.nasa.gov/bio/repo/data/studies/OSD-640/\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sample_row = responses.sample(1)\n",
    "FlowRunner.display_result(\n",
    "    sample_row.result.iloc[0],\n",
    "    sample_row.question.iloc[0],\n",
    "    reference_answer=sample_row.reference_answer.iloc[0]\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9cf2f1ae-33da-4a54-9019-27bcf2c2931b",
   "metadata": {},
   "source": [
    "# Evaluation\n",
    "\n",
    "Here, we evaluate the generated responses. We primarily follow two modes of evaluation:\n",
    "\n",
    "- I. Evaluation of **retrieved contexts** (recall)\n",
    "\n",
    "    How good/accurate are the retrieved contexts based on the input query?\n",
    "\n",
    "- II. Evaluation of **generated answers** (different evaluation metrics)\n",
    "\n",
    "    How good/accurate are the generated answers?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "bc002b54-2f36-479a-ac20-bf88aa718472",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-03-25 18:43:42.335048: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
      "To enable the following instructions: SSE4.1 SSE4.2 AVX AVX2 AVX512F FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n"
     ]
    }
   ],
   "source": [
    "from evalem.nlp.metrics import (\n",
    "    BertScore,\n",
    "    ExactMatchMetric,\n",
    "    MeteorMetric,\n",
    "    RougeMetric,\n",
    ")\n",
    "from evalem import Evaluator\n",
    "from evalem._base.structures import MetricResult"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "0eca9b67-d8cf-4bff-a387-23301253d279",
   "metadata": {},
   "outputs": [],
   "source": [
    "from nltk.corpus import stopwords\n",
    "from nltk.tokenize import word_tokenize\n",
    "from rake_nltk import Rake"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "76b8d635-d6b1-4ee7-8254-166f0639396a",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package stopwords to /home/sagemaker-\n",
      "[nltk_data]     user/nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import nltk\n",
    "nltk.download(\"stopwords\")"
   ]
  },
  {
   "attachments": {
    "258c7d85-21c9-48ee-a018-a4d6d760ea9c.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "id": "8670d2d4-06c6-441d-8195-22ad35ccebd6",
   "metadata": {},
   "source": [
    "## Retrieval Evaluation\n",
    "\n",
    "In this mode of evaluation, we compare whether the actual context (which is the ground truth context) lies in any of the **top-k** retrieved context chunks.\n",
    "\n",
    "$$\\text{Recall} = \\begin{cases} \n",
    "1 & \\text{if actual reference context is in top } k \\text{ retrieved contexts}, \\\\\n",
    "0 & \\text{otherwise}.\n",
    "\\end{cases}$$\n",
    "\n",
    "![recall.png](attachment:258c7d85-21c9-48ee-a018-a4d6d760ea9c.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "69cf023c-9bc4-4575-846e-994f4de625e4",
   "metadata": {},
   "source": [
    "### Keyword-based evaluation\n",
    "\n",
    "We compute the recall of retrieved contexts based on the overlapping of keywords between the\n",
    "actual context (ground truth context) and the retrieved context."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "f78219bc-28aa-4ece-83b1-b72d110c52a0",
   "metadata": {},
   "outputs": [],
   "source": [
    "def extract_keywords(text, top_k: int = 10):\n",
    "    text = preprocess_text(text, top_k)\n",
    "    text = text.lower()\n",
    "    stop_words = set(stopwords.words('english'))\n",
    "    words = word_tokenize(text)\n",
    "    keywords = [word.lower() for word in words if word.isalpha() and word.lower() not in stop_words]\n",
    "    return set(keywords)\n",
    "\n",
    "def preprocess_text(text, top_k=10):\n",
    "    text = text.lower()\n",
    "    r = Rake()\n",
    "    r.extract_keywords_from_text(text)\n",
    "    return \" \".join(set(r.get_ranked_phrases()[:top_k]))\n",
    "\n",
    "def compute_recall_kws(\n",
    "    reference: str,\n",
    "    retrieved_values: List[str],\n",
    "    top_k: int = 1,\n",
    "    top_k_kws: int = 10,\n",
    "    threshold:float = 0.5,\n",
    "    debug: bool = False,\n",
    ") -> float:\n",
    "    ref_keywords = extract_keywords(reference, top_k_kws)\n",
    "    relevant_items = 0\n",
    "\n",
    "    recall = 0.0\n",
    "\n",
    "    if debug:\n",
    "        print(f\">>>>> reference = {reference}\")\n",
    "        print(f\">>>>> reference keywords = {ref_keywords} | {len(ref_keywords)}\")\n",
    "    for idx, value in enumerate(retrieved_values, start=1):\n",
    "        keywords = extract_keywords(value, top_k=top_k_kws)\n",
    "        overlap = len(ref_keywords.intersection(keywords))\n",
    "        score = overlap/len(ref_keywords)\n",
    "        if debug:\n",
    "            print(f\">>>>> retrieved = {value}\")\n",
    "            print(f\">>>>> retrieved kws = {keywords}\")\n",
    "            print(f\">>>>> overlap = {overlap}\")\n",
    "            print(f\">>>> score = {score}\")\n",
    "\n",
    "        if score >= threshold:\n",
    "            recall = 1.0\n",
    "        if idx >= top_k:\n",
    "            break\n",
    "    return recall\n",
    "\n",
    "def evaluate_retrieval_keywords(\n",
    "    response: pd.DataFrame,\n",
    "    threshold=0.5,\n",
    "    top_k:int = 1,\n",
    "    top_k_kws: int = 10,\n",
    "    debug:bool=False\n",
    ") -> pd.DataFrame:\n",
    "    if \"question\" not in response and \"prediction\" not in response and \"reference\" not in response:\n",
    "        response = simplify_response(response)\n",
    "\n",
    "    response = response[response.retrieved_contexts.str.len() != 0]\n",
    "\n",
    "    results = []\n",
    "    for row in response.itertuples():\n",
    "        context = row.context\n",
    "        retrieved_contexts = row.retrieved_contexts\n",
    "        retrieved_contexts = list(map(lambda x: x[\"page_content\"], retrieved_contexts))\n",
    "        recall = compute_recall_kws(\n",
    "            context,\n",
    "            retrieved_contexts,\n",
    "            threshold=threshold,\n",
    "            top_k=top_k,\n",
    "            top_k_kws=top_k_kws,\n",
    "            debug=debug\n",
    "        )\n",
    "        results.append(dict(\n",
    "            context=context,\n",
    "            question=row.question,\n",
    "            answer=row.reference,\n",
    "            prediction=row.prediction,\n",
    "            retrieved_contexts=retrieved_contexts,\n",
    "            recall=recall\n",
    "        ))\n",
    "    return pd.DataFrame(results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "f9cf8089-27ce-4bf9-a1e4-90acca7bdfe4",
   "metadata": {},
   "outputs": [],
   "source": [
    "def simplify_response(response: pd.DataFrame) -> pd.DataFrame:\n",
    "    response = response.copy()\n",
    "    try:\n",
    "        response[\"result\"] = response[\"result\"].apply(ast.literal_eval)\n",
    "    except:\n",
    "        pass\n",
    "    res = []\n",
    "    for row in response.itertuples():\n",
    "        res.append(dict(\n",
    "            context=row.reference_context,\n",
    "            context_hash=row.context_hash,\n",
    "            question=row.question,\n",
    "            reference=row.reference_answer,\n",
    "            ground_truth=row.reference_answer,\n",
    "            prediction=row.result[\"answer\"],\n",
    "            retrieved_contexts=row.result[\"source_documents\"]\n",
    "        ))\n",
    "    return pd.DataFrame(res)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "a1e534ed-1420-4fc1-a351-6b0bc97df72e",
   "metadata": {},
   "outputs": [],
   "source": [
    "def inspect_retrieval_eval(row, func, threshold: float = 0.5, top_k:int = 5, top_k_kws: int = 25):\n",
    "    print(f\"top_k = {top_k}\")\n",
    "    recall = func(\n",
    "            row.context,\n",
    "            list(map(lambda x: x[\"page_content\"], row.retrieved_contexts)),\n",
    "            threshold=threshold,\n",
    "            top_k=top_k,\n",
    "            top_k_kws=top_k_kws,\n",
    "            debug=True\n",
    "        )\n",
    "    print(recall)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "7536ad44-3eaa-4e0b-9141-4c1e9320ece1",
   "metadata": {},
   "outputs": [],
   "source": [
    "def batch_evaluate_retrieval_keywords(\n",
    "    response: pd.DataFrame,\n",
    "):\n",
    "    final_df = []\n",
    "    for threshold in [0.25, 0.3, 0.35, 0.4, 0.45, 0.5]:\n",
    "        for top_k in range(1, 10):\n",
    "            for top_k_kws in [25, 50]:\n",
    "                result = evaluate_retrieval_keywords(\n",
    "                    response,\n",
    "                    threshold=threshold,\n",
    "                    top_k=top_k,\n",
    "                    top_k_kws=top_k_kws,\n",
    "                    debug=False\n",
    "                )\n",
    "                fname = f\"data/evals/retrieval/keywords/eval-context-keywords-{threshold}-{top_k}-{top_k_kws}.csv\"\n",
    "                logger.debug(f\"Dumping eval to {fname}\")\n",
    "                result.to_csv(fname, index=False)\n",
    "\n",
    "def load_retrieval_eval_keywords(\n",
    "    directory: str,\n",
    "    pattern: Optional[str] = None\n",
    ") -> pd.DataFrame:\n",
    "    \n",
    "    if pattern is None:\n",
    "        pattern = \"eval-context-keywords-([0-9.]+)-([0-9]+)-([0-9.]+).csv\"\n",
    "    \n",
    "    dfs = []\n",
    "    search_path = f\"{directory}/eval-context-keywords*.csv\"\n",
    "    for path in glob.glob(search_path):\n",
    "        df = pd.read_csv(path)\n",
    "        match = re.search(pattern, path)\n",
    "        if match:\n",
    "            df[\"threshold\"] = float(match.group(1))\n",
    "            df[\"top_k\"] = int(match.group(2))\n",
    "            df[\"top_k_kws\"] = int(match.group(3))\n",
    "        dfs.append(df)\n",
    "    df = pd.concat(dfs, axis=0)\n",
    "    df[\"retrieved_contexts\"] = df[\"retrieved_contexts\"].apply(ast.literal_eval)\n",
    "    return df\n",
    "\n",
    "def load_retrieval_eval(path: str) -> pd.DataFrame:\n",
    "    df = pd.read_csv(path)\n",
    "    df[\"retrieved_contexts\"] = df[\"retrieved_contexts\"].apply(ast.literal_eval)\n",
    "    return df\n",
    "\n",
    "def peek_evaluation_result(df: pd.DataFrame, peek_type:str =\"max\"):\n",
    "    df = df.copy()\n",
    "    peek_type = peek_type.lower().strip()\n",
    "    if peek_type in [\"max\", \"good\"]:\n",
    "        df = df[df['recall'] == df['recall'].max()]\n",
    "    else:\n",
    "        df = df[df['recall'] == df['recall'].min()]\n",
    "        \n",
    "    row = df.sample(n=1).iloc[0]\n",
    "\n",
    "    header_tag = \"h4\"\n",
    "    display(HTML(f\"<h3>{'Good' if peek_type in ['max', 'good'] else 'Bad'} Recall (Passage Overlap Threshold={row['threshold']}, top_k={row['top_k']})</h3\"))\n",
    "    display(HTML(f\"<{header_tag}>Question</{header_tag}>\"))\n",
    "    print(row[\"question\"])\n",
    "\n",
    "    display(HTML(f\"<{header_tag}>Generated Answer</{header_tag}>\"))\n",
    "    print(row[\"prediction\"])\n",
    "\n",
    "    display(HTML(f\"<{header_tag}>Actual Answer</{header_tag}>\"))\n",
    "    print(row[\"answer\"])\n",
    "\n",
    "    display(HTML(f\"<{header_tag}>Actual Context</{header_tag}>\"))\n",
    "    print(row[\"context\"])\n",
    "\n",
    "    display(HTML(\"<hr/>\"))\n",
    "    for idx, ctx in enumerate(row[\"retrieved_contexts\"], start=1):\n",
    "        display(HTML(f\"<{header_tag}>Retrieved Context #{idx}</{header_tag}>\"))\n",
    "        print(ctx)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "26ba0221-f7f5-412e-ad77-4023dbcd2f49",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_mean_recall_kws(df, figsize=(12, 6)):\n",
    "    df = df.copy()\n",
    "    df['top_k'] = pd.to_numeric(df['top_k'], errors='coerce')\n",
    "    df['threshold'] = pd.to_numeric(df['threshold'], errors='coerce')\n",
    "    df['recall'] = pd.to_numeric(df['recall'], errors='coerce')\n",
    "    df['top_k_kws'] = pd.to_numeric(df['top_k_kws'], errors='coerce')\n",
    "\n",
    "    # df = df.groupby(['threshold', 'top_k', 'top_k_kws'])['recall'].mean().reset_index()\n",
    "\n",
    "    # df_mean_recall = df.copy()\n",
    "\n",
    "    # mean recall vs. top_k with lines for each threshold\n",
    "    plt.figure(figsize=figsize)\n",
    "    ax = sns.lineplot(data=df, x='top_k', y='recall', hue='threshold', marker='o', palette='hls')\n",
    "    plt.title('Recall by Top K Retrieved Contexts (colored by Passage Overlap Threshold)')\n",
    "    plt.xlabel('Top K')\n",
    "    plt.ylabel('Recall')\n",
    "    plt.legend(title='Passage Overlap Threshold')\n",
    "\n",
    "    lines, labels = ax.get_legend_handles_labels()\n",
    "    for line, label in zip(ax.lines, labels[:]):  # Skip the first label as it is the legend title\n",
    "        x, y = line.get_data()\n",
    "        plt.text(x[-1], y[-1], f' {label}', verticalalignment='center')  # Annotate the end of the line\n",
    "    \n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "1d743982-f528-438d-b2a5-59df3ea2b9db",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "top_k = 5\n",
      ">>>>> reference = Genome-wide transcriptional profiling showed that reducing gravity levels in the International Space Station (ISS) causes important alterations in Drosophila gene expression intimately linked to imposed spaceflight-related environmental constrains during Drosophila metamorphosis. However, simulation experiments on ground testing space-related environmental constraints, show differential responses. Curiously, although particular genes are not common in the different experiments, the same GO groups including a large multigene family related with behavior, stress response and organogenesis are over represented in them. A global and integrative analysis using the gene expression dynamics inspector (GEDI) self-organizing maps, reveals different degrees in the responses of the transcriptome when using different environmental conditions or microgravity/hypergravity simulation devices. These results suggest that the transcriptome is finely tuned to normal gravity. In regular environmental conditions the alteration of this constant parameter on Earth can have mild effects on gene expression but when environmental conditions are far from optimal, the gene expression is much more intense and consistent effects.\n",
      ">>>>> reference keywords = {'hypergravity', 'environmental', 'differential', 'reducing', 'linked', 'using', 'reveals', 'gene', 'normal', 'constraints', 'responses', 'large', 'show', 'international', 'levels', 'genes', 'multigene', 'family', 'go', 'although', 'groups', 'causes', 'alterations', 'intimately', 'experiments', 'conditions', 'important', 'profiling', 'wide', 'metamorphosis', 'expression', 'analysis', 'ground', 'dynamics', 'including', 'station', 'particular', 'drosophila', 'transcriptional', 'gravity', 'related', 'devices', 'integrative', 'degrees', 'simulation', 'different', 'constrains', 'space', 'testing', 'showed', 'regular', 'inspector'} | 52\n",
      ">>>>> retrieved = 26-Jun-2013  notifications share  GeneLab ID : GLDS-70 DOI : 10.26030/syws-9962 Source Accession : GSE33779  Cite this Study  info Description  Description  Genome-wide transcriptional profiling showed that reducing gravity levels in the International Space Station (ISS) causes important <b>alterations</b> in <b>Drosophila</b> <b>gene</b> <b>expression</b> intimately linked to imposed spaceflight-related environmental constrains <b>during</b> <b>Drosophila</b> <b>metamorphosis</b>. However, simulation experiments on ground testing space-related environmental constraints, show differential responses. Curiously, although particular <b>genes</b> are not common in the different experiments, the same GO groups including a large multigene family related with behavior, stress response and organogenesis are over represented in them.\n",
      ">>>>> retrieved kws = {'environmental', 'stress', 'differential', 'accession', 'reducing', 'linked', 'id', 'gene', 'cite', 'constraints', 'responses', 'large', 'show', 'international', 'genes', 'info', 'levels', 'multigene', 'genome', 'family', 'go', 'groups', 'intimately', 'alterations', 'share', 'experiments', 'study', 'profiling', 'b', 'wide', 'notifications', 'metamorphosis', 'expression', 'spaceflight', 'ground', 'source', 'genelab', 'response', 'including', 'station', 'description', 'transcriptional', 'drosophila', 'imposed', 'related', 'gravity', 'simulation', 'constrains', 'space', 'testing', 'showed'}\n",
      ">>>>> overlap = 35\n",
      ">>>> score = 0.6730769230769231\n",
      ">>>>> retrieved = <b>Gene</b> <b>expression</b> levels were determined in 3rd instar and adult <b>Drosophila</b> melanogaster reared <b>during</b> spaceflight, to elucidate the genetic and molecular mechanisms underpinning the effects of microgravity on the immune system. The goal was to validate the <b>Drosophila</b> model for understanding <b>alterations</b> of innate immune responses in humans due to spaceflight. Five containers of flies, with ten female and five male fruit flies in each container, were housed and bred on the space shuttle (average orbit altitude of 330.35 km) for 12 days and 18.5 hours, with a new generation reared in microgravity.\n",
      ">>>>> retrieved kws = {'reared', 'gene', 'fruit', 'responses', 'instar', 'containers', 'km', 'flies', 'mechanisms', 'ten', 'female', 'alterations', 'molecular', 'underpinning', 'innate', 'b', 'expression', 'melanogaster', 'due', 'days', 'humans', 'immune', 'hours', 'average', 'drosophila', 'orbit', 'altitude', 'male', 'five', 'system', 'space', 'generation', 'new', 'shuttle'}\n",
      ">>>>> overlap = 6\n",
      ">>>> score = 0.11538461538461539\n",
      ">>>>> retrieved = pathways and gene products. Robust transcriptional response was evident among the pupae developed in a hypergravity environment compared to control. 1,513 genes showed significantly (p < 0.05) altered gene expression in the 3 g samples. These findings were supported with qRT-PCR data. Major biological processes affected include ion transport, redox homeostasis, immune and humoral stress response, proteolysis, and cuticle development.  Factor(s)  Factor  Ontology: Concept  Hypergravity  Hypergravity  Organism(s)  Drosophila melanogaster  Assay(s)  Measurement  Technology  Device Platform  transcription profiling  RNA Sequencing (RNA-Seq)  Illumina HiSeq 2000  Project  Project Title  Gravity and oxidative stress responses in Drosophila melanogaster  Project Type\n",
      ">>>>> retrieved kws = {'rna', 'hypergravity', 'products', 'stress', 'samples', 'processes', 'qrt', 'ontology', 'development', 'gene', 'supported', 'cuticle', 'responses', 'pupae', 'oxidative', 'genes', 'illumina', 'affected', 'title', 'concept', 'sequencing', 'type', 'project', 'transcription', 'ion', 'environment', 'developed', 'profiling', 'technology', 'transport', 'altered', 'expression', 'melanogaster', 'data', 'include', 'humoral', 'measurement', 'compared', 'response', 'organism', 'transcriptional', 'drosophila', 'pcr', 'gravity', 'major', 'device', 'redox', 'significantly', 'platform', 'biological', 'seq', 'among', 'assay', 'g', 'robust', 'hiseq', 'showed', 'factor', 'evident', 'homeostasis'}\n",
      ">>>>> overlap = 10\n",
      ">>>> score = 0.19230769230769232\n",
      ">>>>> retrieved = To assess brain <b>gene</b> profile changes associated with microgravity and fluid shift, a large-scale analysis of mRNA <b>expression</b> levels was performed in the brains of 2 weeks control and hindlimb-unloaded (HU) mice using cDNA microarrays. Although to different extent, all functional categories displayed significantly regulated <b>genes</b> indicating that considerable transcriptomic <b>alterations</b> are induced by HU.\n",
      ">>>>> retrieved kws = {'transcriptomic', 'unloaded', 'brain', 'using', 'gene', 'weeks', 'changes', 'large', 'regulated', 'genes', 'levels', 'indicating', 'microarrays', 'associated', 'assess', 'alterations', 'b', 'mrna', 'expression', 'mice', 'categories', 'analysis', 'extent', 'considerable', 'shift', 'significantly', 'induced', 'control', 'cdna', 'different', 'functional', 'fluid', 'scale', 'displayed', 'performed', 'microgravity', 'profile'}\n",
      ">>>>> overlap = 9\n",
      ">>>> score = 0.17307692307692307\n",
      ">>>>> retrieved = Gene expression that is modified under these parameters were compared between Tgfb1 wild type and heterozygote littermates in order to determine which genes induced or repressed by radiation were mediated via Tgfb1 status. Differential gene expression was analyzed in Tgfb1 heterozygote and wild type littermate 4th mammary glands, after whole body exposure to an acute dose of 10cGy ionizing radiation. Estrus cycle was normalized in all mice two days prior to irradiation by injection with an estrogen and progesterone mixture.\n",
      ">>>>> retrieved kws = {'differential', 'determine', 'two', 'injection', 'gene', 'prior', 'parameters', 'whole', 'order', 'modified', 'mammary', 'estrogen', 'progesterone', 'genes', 'estrus', 'cycle', 'exposure', 'type', 'littermates', 'acute', 'wild', 'via', 'body', 'expression', 'dose', 'mice', 'days', 'compared', 'littermate', 'mediated', 'heterozygote', 'repressed', 'mixture', 'glands', 'induced', 'radiation', 'irradiation', 'normalized', 'status', 'ionizing'}\n",
      ">>>>> overlap = 4\n",
      ">>>> score = 0.07692307692307693\n",
      "1.0\n"
     ]
    }
   ],
   "source": [
    "inspect_retrieval_eval(\n",
    "    row=simplify_response(load_responses(\"../data/osdr-eval/responses-gpt-3.5-turbo.csv\")).iloc[0],\n",
    "    func=compute_recall_kws,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "a6221d61-b051-441a-8bf1-95476cc48377",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(30996, 9)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>context</th>\n",
       "      <th>question</th>\n",
       "      <th>answer</th>\n",
       "      <th>prediction</th>\n",
       "      <th>retrieved_contexts</th>\n",
       "      <th>recall</th>\n",
       "      <th>threshold</th>\n",
       "      <th>top_k</th>\n",
       "      <th>top_k_kws</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Genome-wide transcriptional profiling showed t...</td>\n",
       "      <td>Where were the gene expression alterations in ...</td>\n",
       "      <td>reducing gravity levels in the International S...</td>\n",
       "      <td>The gene expression alterations in Drosophila ...</td>\n",
       "      <td>[26-Jun-2013  notifications share  GeneLab ID ...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.25</td>\n",
       "      <td>6</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Genome-wide transcriptional profiling showed t...</td>\n",
       "      <td>What kind of experiments showed differential r...</td>\n",
       "      <td>simulation experiments on ground testing space...</td>\n",
       "      <td>Experiments involving reducing gravity levels ...</td>\n",
       "      <td>[26-Jun-2013  notifications share  GeneLab ID ...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.25</td>\n",
       "      <td>6</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Genome-wide transcriptional profiling showed t...</td>\n",
       "      <td>What is overrepresented in the different exper...</td>\n",
       "      <td>the same GO groups including a large multigene...</td>\n",
       "      <td>The same GO groups, including a large multigen...</td>\n",
       "      <td>[Curiously, although &lt;b&gt;particular&lt;/b&gt; &lt;b&gt;gene...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.25</td>\n",
       "      <td>6</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Genome-wide transcriptional profiling showed t...</td>\n",
       "      <td>What are the implications of the findings in t...</td>\n",
       "      <td>The findings indicate that the finely adjusted...</td>\n",
       "      <td>The findings suggest that the transcriptome is...</td>\n",
       "      <td>[These results suggest that, in addition to be...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.25</td>\n",
       "      <td>6</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>The objective of the Rodent Research-5 (RR-5) ...</td>\n",
       "      <td>What was the main goal of the Rodent Research-...</td>\n",
       "      <td>to evaluate bone loss in mice during spaceflig...</td>\n",
       "      <td>The main goal of the Rodent Research-5 (RR-5) ...</td>\n",
       "      <td>[12-Aug-2019  notifications share  GeneLab ID ...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.25</td>\n",
       "      <td>6</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                             context  \\\n",
       "0  Genome-wide transcriptional profiling showed t...   \n",
       "1  Genome-wide transcriptional profiling showed t...   \n",
       "2  Genome-wide transcriptional profiling showed t...   \n",
       "3  Genome-wide transcriptional profiling showed t...   \n",
       "4  The objective of the Rodent Research-5 (RR-5) ...   \n",
       "\n",
       "                                            question  \\\n",
       "0  Where were the gene expression alterations in ...   \n",
       "1  What kind of experiments showed differential r...   \n",
       "2  What is overrepresented in the different exper...   \n",
       "3  What are the implications of the findings in t...   \n",
       "4  What was the main goal of the Rodent Research-...   \n",
       "\n",
       "                                              answer  \\\n",
       "0  reducing gravity levels in the International S...   \n",
       "1  simulation experiments on ground testing space...   \n",
       "2  the same GO groups including a large multigene...   \n",
       "3  The findings indicate that the finely adjusted...   \n",
       "4  to evaluate bone loss in mice during spaceflig...   \n",
       "\n",
       "                                          prediction  \\\n",
       "0  The gene expression alterations in Drosophila ...   \n",
       "1  Experiments involving reducing gravity levels ...   \n",
       "2  The same GO groups, including a large multigen...   \n",
       "3  The findings suggest that the transcriptome is...   \n",
       "4  The main goal of the Rodent Research-5 (RR-5) ...   \n",
       "\n",
       "                                  retrieved_contexts  recall  threshold  \\\n",
       "0  [26-Jun-2013  notifications share  GeneLab ID ...     1.0       0.25   \n",
       "1  [26-Jun-2013  notifications share  GeneLab ID ...     1.0       0.25   \n",
       "2  [Curiously, although <b>particular</b> <b>gene...     1.0       0.25   \n",
       "3  [These results suggest that, in addition to be...     1.0       0.25   \n",
       "4  [12-Aug-2019  notifications share  GeneLab ID ...     1.0       0.25   \n",
       "\n",
       "   top_k  top_k_kws  \n",
       "0      6         25  \n",
       "1      6         25  \n",
       "2      6         25  \n",
       "3      6         25  \n",
       "4      6         25  "
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# eval_retrieval_kws = load_retrieval_eval_keywords(directory=\"../data/osdr-eval/evals/retrieval/keywords/\")\n",
    "eval_retrieval_kws = load_retrieval_eval(\"../data/osdr-eval/osdr-eval-retrieval-kws.csv\")\n",
    "print(eval_retrieval_kws.shape)\n",
    "eval_retrieval_kws.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "27dd072d-7ee6-4334-8998-1b3ec5a68254",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>recall</th>\n",
       "      <th>threshold</th>\n",
       "      <th>top_k</th>\n",
       "      <th>top_k_kws</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>30996.000000</td>\n",
       "      <td>30996.000000</td>\n",
       "      <td>30996.000000</td>\n",
       "      <td>30996.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.359595</td>\n",
       "      <td>0.375000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>37.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.479889</td>\n",
       "      <td>0.085393</td>\n",
       "      <td>2.582031</td>\n",
       "      <td>12.500202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>25.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.300000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>25.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.375000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>37.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.450000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>50.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>50.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             recall     threshold         top_k     top_k_kws\n",
       "count  30996.000000  30996.000000  30996.000000  30996.000000\n",
       "mean       0.359595      0.375000      5.000000     37.500000\n",
       "std        0.479889      0.085393      2.582031     12.500202\n",
       "min        0.000000      0.250000      1.000000     25.000000\n",
       "25%        0.000000      0.300000      3.000000     25.000000\n",
       "50%        0.000000      0.375000      5.000000     37.500000\n",
       "75%        1.000000      0.450000      7.000000     50.000000\n",
       "max        1.000000      0.500000      9.000000     50.000000"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "eval_retrieval_kws.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "f259f1c8-c958-4200-8874-50abd18ed7da",
   "metadata": {},
   "outputs": [],
   "source": [
    "# eval_retrieval_kws.to_csv(\"data/evals/osdr-eval-retrieval-kws.csv\", index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "6142066a-3fa1-4a42-a1a1-15c2cb5a8c84",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<h3>Good Recall (Passage Overlap Threshold=0.3, top_k=9)</h3"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Question</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "What role does HsfA2 play in physiological adaptation to the spaceflight environment and what cellular processes are involved?\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Generated Answer</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "HsfA2 plays a role in the physiological adaptation to the spaceflight environment by influencing cell wall remodeling, signal perception and transduction, and starch biosynthesis. It also has a universal stress response role. Additionally, HsfA2 is involved in the chaperone-mediated protein folding machinery, endoplasmic reticulum stress, and unfolded protein response, which are crucial cellular processes for adaptation to the spaceflight environment.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Actual Answer</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "HsfA2 plays a role in the universal stress response, as well as specific functions in physiological adaptation to spaceflight, involving processes like cell wall remodeling and starch biosynthesis.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Actual Context</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Heat Shock Factor A2 (HsfA2), as a part of the HSF network, is essential to the plant's response to almost any environmental stress and to the cellular homeostatic control mechanisms. Plant cell cultures disabled in HsfA2 function were grown aboard the International Space Station (ISS) in order to ascertain whether or not they use the same terrestrially effective systems to adapt to the novel environment of spaceflight. Cultured lines of Arabidopsis thaliana derived from wild type (WT) cultivar Col-0 and from a knock-out line deficient in the gene encoding HSFA2 (HSFA2 KO) were launched to the ISS on SpaceX-2 as part of the Cellular Expression Logic (CEL) experiment of the BRIC17 spaceflight mission and were fixed in-flight after 10 days on orbit. Microarray gene expression data were analyzed using a two-part comparative approach. First, differentially expressed genes were identified between the environments (spaceflight to ground) within cells of the same genotype, which represented physiological adaptation to the spaceflight environment. Second, gene expression profiles were identified between the genotypes (HSFA2 KO to WT) within the same environment, defining genes uniquely required by the two genotypes in the ground and spaceflight adapted states. The physiological state of the cells as a result of disabling a gene has tremendous control over the mechanisms induced to adapt to the environment of spaceflight. The HsfA2 demonstrated a role in the physiological adaptation to the spaceflight environment since the cells disabled in the HsfA2 gene used substantially different genes to achieve the spaceflight adapted state than the WT cells.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #1</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The <b>HsfA2</b> demonstrated a <b>role</b> in the <b>physiological</b> <b>adaptation</b> to the <b>spaceflight</b> <b>environment</b> since the cells disabled in the <b>HsfA2</b> gene used substantially different genes to achieve the <b>spaceflight</b> adapted state than the WT cells. The endoplasmic reticulum (ER) stress and unfolded protein response (UPR) define the <b>HSFA2</b> KO cells' <b>physiological</b> state regardless of the <b>environment</b> and likely result from the deficiency in the chaperone-mediated protein folding machinery. <b>HsfA2</b> seems to have a universal stress response <b>role</b> but also specific <b>roles</b> in the <b>physiological</b> <b>adaptation</b> to <b>spaceflight</b> through cell wall remodeling, signal perception and transduction and starch biosynthesis.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #2</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<b>HsfA2</b> seems to have a universal stress response <b>role</b> but also specific <b>roles</b> in the <b>physiological</b> <b>adaptation</b> to <b>spaceflight</b> through cell wall remodeling, signal perception and transduction and starch biosynthesis. Implementation of knock-out cells identified a set of genes with a required expression level in order for a cell to achieve a <b>spaceflight</b>-adapted state. The <b>HSFA2</b> KO cells helped to unravel the <b>HsfA2</b>-dependent genes of the adaption of wild type cells to the <b>environment</b> of <b>spaceflight</b>.  Factor(s)  Factor  Ontology: Concept  <b>Spaceflight</b>  Space Flight  Genotype  Genotype  Organism(s)  Arabidopsis thaliana  Assay(s)  Measurement  Technology  Device Platform  transcription profiling\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #3</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Second, gene expression profiles were identified between the genotypes (<b>HSFA2</b> KO to WT) within the same <b>environment</b>, defining genes uniquely required by the two genotypes in the ground and <b>spaceflight</b> adapted states. The <b>physiological</b> state of the cells as a result of disabling a gene has tremendous control over the mechanisms induced to adapt to the <b>environment</b> of <b>spaceflight</b>. The <b>HsfA2</b> demonstrated a <b>role</b> in the <b>physiological</b> <b>adaptation</b> to the <b>spaceflight</b> <b>environment</b> since the cells disabled in the <b>HsfA2</b> gene used substantially different genes to achieve the <b>spaceflight</b> adapted state than the WT cells.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #4</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MAS5  GLDS-205_microarray_GSM2509687-tbl-1.txt  GeneLab microarray data processing protocol  GLDS-205_array_raw_intensities_probe.csv  GLDS-205_array_normalized_expression_probeset.csv , GLDS-205_array_normalized_intensities_probe.csv  GLDS-205_array_SampleTable.csv , GLDS-205_array_contrasts.csv , GLDS-205_array_differential_expression.csv  Items per page:  25  1 - 16 of 16  menu_book Publications  <b>HSFA2</b> Functions in the <b>Physiological</b> <b>Adaptation</b> of Undifferentiated Plant Cells to <b>Spaceflight</b>  Authors : Zupanska, A.K., Collin L., Ferl, R.J, and Paul, A-L.  PubMed ID : 30658467  DOI : 10.3390/ijms20020390  sim_card_download Files  Study Files Selected: 0  Search Files  search  OSD-205  GeneLab <b>Processed</b> Microarray Data Files  Microarray Data Files  Study Metadata Files  view_timeline Version History  Selected Version  Version 3  Updated Date:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #5</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<b>Cellular</b> Expression Logic (CEL) experiment of the BRIC17 <b>spaceflight</b> mission and were fixed in-flight after 10 days on orbit. Microarray gene expression data were analyzed using a two-part comparative approach. First, differentially expressed genes were identified between the <b>environments</b> (<b>spaceflight</b> to ground) within cells of the same genotype, which represented <b>physiological</b> <b>adaptation</b> to the <b>spaceflight</b> <b>environment</b>. Second, gene expression profiles were identified between the genotypes (<b>HSFA2</b> KO to WT) within the same <b>environment</b>, defining genes uniquely required by the two genotypes in the ground and <b>spaceflight</b> adapted states.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #6</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Zupanska AK1, Schultz ER1, Yao J2,3, Sng NJ1, Zhou M1, Callaham JB1, Ferl RJ1,2, Paul AL1.  PubMed ID : 29088549  DOI : 10.1089/ast.2016.1538  <b>HSFA2</b> Functions in the <b>Physiological</b> <b>Adaptation</b> of Undifferentiated Plant Cells to <b>Spaceflight</b>  Authors : Zupanska AK, LeFrois C, Ferl RJ, Paul AL  PubMed ID : 30658467  sim_card_download Files  Study Files Selected: 0  Search Files  search  OSD-147  GeneLab <b>Processed</b> Microarray Data Files  Microarray Data Files  Study Metadata Files  view_timeline Version History  Selected Version  Version 7  Updated Date: 03-Oct-2023  Changes: <b>Processed</b> data release  Files Added (19)  OSD-147_metadata_GLDS-147-ISA.zip  GLDS-147_array_processed_md5sum.tsv\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #7</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "About  Charter  How to Join  Current AWG Members  Citizen Science  SOLSTICE  SOLSTICE Members  FAQ  Terms & Conditions  Genelab Tutorials  Contact Us  account_circle expand_more  399.26 MB  OSD-205 Version 3  <b>HSFA2</b> functions in the <b>physiological</b> <b>adaptation</b> of undifferentiated plant cells to <b>spaceflight</b> microgravity <b>environment</b> Study  Submitted Date: 27-Feb-2017  Initial Release Date: 01-Dec-2018  notifications share  GeneLab ID : GLDS-205 DOI : 10.26030/bjyc-pj66 Source Accession : GSE95388  Cite this Study  info Description  Description\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #8</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "01-Dec-2018  notifications share  GeneLab ID : GLDS-205 DOI : 10.26030/bjyc-pj66 Source Accession : GSE95388  Cite this Study  info Description  Description  The Heat Shock Factor A2 (<b>HsfA2</b>), as a part of the HSF network, is essential to the plant's response to almost any environmental stress and to the <b>cellular</b> homeostatic control mechanisms. Plant cell cultures disabled in <b>HsfA2</b> function were grown aboard the International Space Station (ISS) in order to ascertain whether or not they use the same terrestrially effective systems to adapt to the novel <b>environment</b> of <b>spaceflight</b>.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #9</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Plant cell cultures disabled in <b>HsfA2</b> function were grown aboard the International Space Station (ISS) in order to ascertain whether or not they use the same terrestrially effective systems to adapt to the novel <b>environment</b> of <b>spaceflight</b>. Cultured lines of Arabidopsis thaliana derived from wild type (WT) cultivar Col-0 and from a knock-out line deficient in the gene encoding <b>HSFA2</b> (<b>HSFA2</b> KO) were launched to the ISS on SpaceX-2 as part of the <b>Cellular</b> Expression Logic (CEL) experiment of the BRIC17 <b>spaceflight</b> mission and were fixed in-flight after 10 days on orbit.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #10</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The cultured cell lines were grown within 60mm Petri plates in Petri Dish Fixation Units (PDFUs) that were housed within the Biological Research <b>In</b> Canisters (BRIC) hardware. <b>Spaceflight</b> samples were fixed on orbit. Differentially expressed genes were identified between the two <b>environments</b> (<b>spaceflight</b> and comparable ground controls) and the two genotypes (WT and ARG1 KO). Each genotype engaged unique genes during <b>physiological</b> <b>adaptation</b> to the <b>spaceflight</b> <b>environment</b>, with little overlap.\n"
     ]
    }
   ],
   "source": [
    "peek_evaluation_result(eval_retrieval_kws, peek_type=\"good\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "id": "a4d86613-a98c-438a-939b-6fc420faeed8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<h3>Bad Recall (Passage Overlap Threshold=0.3, top_k=9)</h3"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Question</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "What kind of tissues did this study use?\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Generated Answer</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The study mentioned in the context used various tissues such as lumbar spine, tibialis anterior, tibia, soleus, femur, and extensor digitorum longus from mice for different assays like DXA Scanning and Western blot.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Actual Answer</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Hair\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Actual Context</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "In each mission, five strands of hair were sampled six times from each astronaut within the period of July 2009 to February 2013. The sampling timings were as follows: 6 sampling time points (Fig 1); first preflight (Launch (L) -180 to -90 days: 6 to 3 months before launch), second preflight (L-60 to -14: 2 months to 2 weeks before launch), first inflight (L+20 to 37: 20 to 37 days after launch), second inflight (Return (R) -20 to -7: 20 to 7 days before return), first postflight (R+2 to 7: 2 to 7 days after return), and second postflight (R+30 to 90: from 1 to 3 months after return). In sampling day, two astronauts were paired and plucked another hair sample using tweezer in the direction of hair growth without damaging the hair roots. These plucked samples were stored at -80C until analysis.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #1</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "In response to these and other observations, the Rapid Freeze hardware was built for <b>use</b> on the International Space Station. The Rapid Freeze hardware freezes mouse <b>tissues</b> (Glovebox freezer) and whole carcasses (Cryochiller) at rates closely mimicking those attained with immersion in liquid nitrogen. Because this hardware will be <b>used</b> extensively on future rodent research missions, it is crucial to understand whether or not it preserves signals in gene expression data in order to maximize the value of these rare and expensive spaceflight experiments. Therefore, this <b>study</b> was designed with three goals:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #2</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This <b>study</b> derives results from the DXA Scanning assay <b>using</b> lumbar spine <b>tissue</b> from HLS mice. The lumbar spine data in this <b>study</b> are related to other <b>studies</b> <b>using</b> <b>tissues</b> from the same experiment; OSD-654 (tibia), OSD-660 (tibialis anterior), OSD-662 (soleus), OSD-663 (femur), and OSD-664 (extensor digitorum longus).  Factor(s)  Factor  Ontology: Concept  Hindlimb Unloading  Hindlimb Suspension  Genotype  Genotype  Organism(s)  Mus musculus  Assay(s)  Measurement  Technology  Device Platform  Bone Microstructure  Dual-Energy X-Ray Absorptiometry  osteosys insight  Project  Payload Identifier  GROUND-2  Project Type  Ground <b>Study</b>  Experiment Platform  Hindlimb Unloading  Sponsoring Agency  NASA, CSA\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #3</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "01-Jul-2011  notifications share  GeneLab ID : GLDS-92 DOI : 10.26030/2k69-6j41 Source Accession : E-GEOD-16935  Cite this <b>Study</b>  info Description  Description  Accumulating data suggest that the biological responses to high and low doses of radiation are qualitatively different, necessitating the direct <b>study</b> of low dose responses. Most such <b>studies</b> have utilized 2-dimensional culture systems, which may not fully represent responses in 3-dimensional <b>tissues</b>.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #4</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This <b>study</b> derives results from the Western blot assay <b>using</b> tibialis anterior <b>tissue</b> from the RR9 mission. The tibialis anterior data in this <b>study</b> are related to other <b>studies</b> <b>using</b> <b>tissues</b> from the same experiment; OSD-654 (tibia), OSD-661 (lumbar spine), OSD-662 (soleus), OSD-663 (femur), and OSD-664 (extensor digitorum longus).  Factor(s)  Factor  Ontology: Concept  Spaceflight  Space Flight  Organism(s)  Mus musculus  Assay(s)  Measurement  Technology  Device Platform  protein quantification  Western Blot  BioRad ChemiDoc  Project  Payload Identifier  RR-9  Project Type  Spaceflight <b>Study</b>  Flight Program  ISS  Experiment Platform  Rodent Habitat  Sponsoring Agency  NASA  NASA Center\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #5</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "strength, and promoted the oxidative fiber type with Earth-based hindlimb unloading. In bone, GSK3 activation was enhanced after spaceflight; and strikingly, muscle-specific Gsk3 deletion increased bone mineral density in response to hindlimb unloading. Thus, future <b>studies</b> should test the effects of GSK3 inhibition during spaceflight. This <b>study</b> derives results from the Western blot assay <b>using</b> tibialis anterior <b>tissue</b> from the RR9 mission. The tibialis anterior data in this <b>study</b> are related to other <b>studies</b> <b>using</b> <b>tissues</b> from the same experiment;\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #6</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "different molecular mechanisms of tumorigenesis. Also seen in this <b>study</b> was novel unmapped transcripts that were significantly affected by HZE. To investigate the biological functions of these transcripts, we <b>used</b> a machine learning technique known as self-organizing maps (SOMs) to characterize the transcriptome expression profiles of 60 samples (45 HZE-irradiated, 15 non-irradiated control) from liver <b>tissues</b>. A handful of localized modules in the maps emerged as groups of co-regulated and co-expressed transcripts. The functional context of these modules was discovered <b>using</b> overrepresentation analysis.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #7</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "order to maximize the value of these rare and expensive spaceflight experiments. Therefore, this <b>study</b> was designed with three goals: 1) To evaluate the temperature profile of the Cryochiller and Glovebox freezer cartridges (Rapid Freeze hardware) over time during mock on-orbit procedures; 2) To determine the freezing profiles of <b>tissues</b> and carcasses <b>using</b> Rapid Freeze hardware at both optimal and sub-optimal temperatures (to mimic on-orbit operations), compared with those frozen in liquid nitrogen (the laboratory gold standard) or frozen at -80 C (the current standard method); 3) To identify gene expression changes in a) <b>tissues</b> that were\n"
     ]
    }
   ],
   "source": [
    "peek_evaluation_result(eval_retrieval_kws, peek_type=\"bad\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "d39fd2a8-be95-490e-9475-a99cfa696fd3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_mean_recall_kws(eval_retrieval_kws)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f3bad7a6-b60b-43f9-afe8-fd754f370be7",
   "metadata": {},
   "source": [
    "### Semantic Retrieval Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "e0bdca06-0871-454b-9244-35c2ec736c72",
   "metadata": {},
   "outputs": [],
   "source": [
    "BERT_SCORE = BertScore(per_instance_score=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "525a9ba4-d02e-4e92-8cd3-cea549b03d3c",
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_recall_semantic(\n",
    "    reference: str,\n",
    "    retrieved_values: List[str],\n",
    "    top_k: int = 1,\n",
    "    top_k_kws: int = 10,\n",
    "    threshold:float = 0.5,\n",
    "    debug: bool = False,\n",
    ") -> float:\n",
    "    \n",
    "    recall = 0.0\n",
    "    \n",
    "    retrieved_values = retrieved_values[:top_k]\n",
    "        \n",
    "    references = [reference] * len(retrieved_values)\n",
    "    predictions = retrieved_values.copy()\n",
    "    result = BERT_SCORE(references=references, predictions=predictions)\n",
    "    if debug:\n",
    "        print(f\">>>>> reference = {reference}\")\n",
    "        print(\">>>> Retrieved Contexts\")\n",
    "        for idx, (ctx, score) in enumerate(zip(retrieved_values, result.extra[\"bertscore\"][\"f1\"]), start=1):\n",
    "            print(f\"#{idx} : {ctx} | {score}\")\n",
    "            \n",
    "    max_score = max(result.extra[\"bertscore\"][\"recall\"])\n",
    "    return float(max_score >= threshold), max_score\n",
    "\n",
    "def evaluate_retrieval_semantic(\n",
    "    response: pd.DataFrame,\n",
    "    threshold=0.5,\n",
    "    top_k:int = 1,\n",
    "    top_k_kws: int = 10,\n",
    "    debug:bool=False\n",
    ") -> pd.DataFrame:\n",
    "    if \"question\" not in response and \"prediction\" not in response and \"reference\" not in response:\n",
    "        response = simplify_response(response)\n",
    "\n",
    "    response = response[response.retrieved_contexts.str.len() != 0]\n",
    "\n",
    "    results = []\n",
    "    for row in response.itertuples():\n",
    "        context = row.context\n",
    "        retrieved_contexts = row.retrieved_contexts\n",
    "        retrieved_contexts = list(map(lambda x: x[\"page_content\"], retrieved_contexts))\n",
    "        recall, score = compute_recall_semantic(\n",
    "            context,\n",
    "            retrieved_contexts,\n",
    "            threshold=threshold,\n",
    "            top_k=top_k,\n",
    "            top_k_kws=top_k_kws,\n",
    "            debug=debug\n",
    "        )\n",
    "        if debug:\n",
    "            print(recall, score)\n",
    "        results.append(dict(\n",
    "            context=context,\n",
    "            question=row.question,\n",
    "            answer=row.reference,\n",
    "            prediction=row.prediction,\n",
    "            retrieved_contexts=retrieved_contexts,\n",
    "            recall=recall,\n",
    "            score=score,\n",
    "        ))\n",
    "    return pd.DataFrame(results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "d550373d-7d7f-4637-aac1-f21aa33b82d5",
   "metadata": {},
   "outputs": [],
   "source": [
    "def batch_eval_semantic(response):\n",
    "    thresholds = [0.5, 0.65]\n",
    "    top_k_vals = [1, 5, 10]\n",
    "    for threshold in thresholds:\n",
    "        for top_k in top_k_vals:\n",
    "            logger.debug(f\"Evaluating for threshold={threshold}, top_k={top_k}\")\n",
    "            result = evaluate_retrieval_semantic_v2(\n",
    "                response,\n",
    "                threshold=threshold,\n",
    "                top_k=top_k, \n",
    "                top_k_kws=50,\n",
    "                debug=False)\n",
    "            fname = f\"data/evals/retrieval/bertscore/eval-context-bertscore-{threshold}-{top_k}.csv\"\n",
    "            logger.debug(f\"Dumping eval to {fname}\")\n",
    "            result.to_csv(fname, index=False)\n",
    "\n",
    "def load_retrieval_eval_semantic(\n",
    "    directory: str,\n",
    "    pattern: Optional[str] = None\n",
    ") -> pd.DataFrame:\n",
    "    \n",
    "    if pattern is None:\n",
    "        pattern = \"eval-context-bertscore-([0-9.]+)-([0-9]+).csv\"\n",
    "    \n",
    "    dfs = []\n",
    "    search_path = f\"{directory}/eval-context-bertscore*.csv\"\n",
    "    for path in glob.glob(search_path):\n",
    "        df = pd.read_csv(path)\n",
    "        match = re.search(pattern, path)\n",
    "        if match:\n",
    "            df[\"threshold\"] = float(match.group(1))\n",
    "            df[\"top_k\"] = int(match.group(2))\n",
    "        dfs.append(df)\n",
    "    df = pd.concat(dfs, axis=0)\n",
    "    df[\"retrieved_contexts\"] = df[\"retrieved_contexts\"].apply(ast.literal_eval)\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "8f24d738-5d55-41c3-abee-d7a9b3ead510",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_mean_recall_semantic(df, figsize=(12, 6)):\n",
    "    df = df.copy()\n",
    "    df['top_k'] = pd.to_numeric(df['top_k'], errors='coerce')\n",
    "    df['threshold'] = pd.to_numeric(df['threshold'], errors='coerce')\n",
    "    df['recall'] = pd.to_numeric(df['recall'], errors='coerce')\n",
    "\n",
    "    # recall vs. top_k with lines for each threshold\n",
    "    plt.figure(figsize=figsize)\n",
    "    ax = sns.lineplot(data=df, x='top_k', y='recall', hue='threshold', marker='o', palette='dark')\n",
    "    plt.title('Recall by Top K Retrieved Contexts (colored by Passage Overlap Threshold)')\n",
    "    plt.xlabel('Top K')\n",
    "    plt.ylabel('Recall')\n",
    "    plt.legend(title='Passage Overlap Threshold')\n",
    "\n",
    "    lines, labels = ax.get_legend_handles_labels()\n",
    "    for line, label in zip(ax.lines, labels[:]):  # Skip the first label as it is the legend title\n",
    "        x, y = line.get_data()\n",
    "        plt.text(x[-1], y[-1], f' {label}', verticalalignment='center')  # Annotate the end of the line\n",
    "    \n",
    "    plt.show()\n",
    "\n",
    "    # recall vs. threshold with lines for each top_k_kws\n",
    "    # plt.figure(figsize=figsize)\n",
    "    # sns.lineplot(data=df, x='threshold', y='recall', marker='o', palette='vlag')\n",
    "    # plt.title('Recall by Passage Overlap Threshold')\n",
    "    # plt.xlabel('Passage Overlap Threshold')\n",
    "    # plt.ylabel('Recall')\n",
    "    # plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "c1a202d1-84e9-4bc8-aafb-955e8302ce6d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "top_k = 5\n",
      ">>>>> reference = Genome-wide transcriptional profiling showed that reducing gravity levels in the International Space Station (ISS) causes important alterations in Drosophila gene expression intimately linked to imposed spaceflight-related environmental constrains during Drosophila metamorphosis. However, simulation experiments on ground testing space-related environmental constraints, show differential responses. Curiously, although particular genes are not common in the different experiments, the same GO groups including a large multigene family related with behavior, stress response and organogenesis are over represented in them. A global and integrative analysis using the gene expression dynamics inspector (GEDI) self-organizing maps, reveals different degrees in the responses of the transcriptome when using different environmental conditions or microgravity/hypergravity simulation devices. These results suggest that the transcriptome is finely tuned to normal gravity. In regular environmental conditions the alteration of this constant parameter on Earth can have mild effects on gene expression but when environmental conditions are far from optimal, the gene expression is much more intense and consistent effects.\n",
      ">>>> Retrieved Contexts\n",
      "#1 : 26-Jun-2013  notifications share  GeneLab ID : GLDS-70 DOI : 10.26030/syws-9962 Source Accession : GSE33779  Cite this Study  info Description  Description  Genome-wide transcriptional profiling showed that reducing gravity levels in the International Space Station (ISS) causes important <b>alterations</b> in <b>Drosophila</b> <b>gene</b> <b>expression</b> intimately linked to imposed spaceflight-related environmental constrains <b>during</b> <b>Drosophila</b> <b>metamorphosis</b>. However, simulation experiments on ground testing space-related environmental constraints, show differential responses. Curiously, although particular <b>genes</b> are not common in the different experiments, the same GO groups including a large multigene family related with behavior, stress response and organogenesis are over represented in them. | 0.6429465413093567\n",
      "#2 : <b>Gene</b> <b>expression</b> levels were determined in 3rd instar and adult <b>Drosophila</b> melanogaster reared <b>during</b> spaceflight, to elucidate the genetic and molecular mechanisms underpinning the effects of microgravity on the immune system. The goal was to validate the <b>Drosophila</b> model for understanding <b>alterations</b> of innate immune responses in humans due to spaceflight. Five containers of flies, with ten female and five male fruit flies in each container, were housed and bred on the space shuttle (average orbit altitude of 330.35 km) for 12 days and 18.5 hours, with a new generation reared in microgravity. | 0.49236226081848145\n",
      "#3 : pathways and gene products. Robust transcriptional response was evident among the pupae developed in a hypergravity environment compared to control. 1,513 genes showed significantly (p < 0.05) altered gene expression in the 3 g samples. These findings were supported with qRT-PCR data. Major biological processes affected include ion transport, redox homeostasis, immune and humoral stress response, proteolysis, and cuticle development.  Factor(s)  Factor  Ontology: Concept  Hypergravity  Hypergravity  Organism(s)  Drosophila melanogaster  Assay(s)  Measurement  Technology  Device Platform  transcription profiling  RNA Sequencing (RNA-Seq)  Illumina HiSeq 2000  Project  Project Title  Gravity and oxidative stress responses in Drosophila melanogaster  Project Type | 0.5513865947723389\n",
      "#4 : To assess brain <b>gene</b> profile changes associated with microgravity and fluid shift, a large-scale analysis of mRNA <b>expression</b> levels was performed in the brains of 2 weeks control and hindlimb-unloaded (HU) mice using cDNA microarrays. Although to different extent, all functional categories displayed significantly regulated <b>genes</b> indicating that considerable transcriptomic <b>alterations</b> are induced by HU. | 0.5268535017967224\n",
      "#5 : Gene expression that is modified under these parameters were compared between Tgfb1 wild type and heterozygote littermates in order to determine which genes induced or repressed by radiation were mediated via Tgfb1 status. Differential gene expression was analyzed in Tgfb1 heterozygote and wild type littermate 4th mammary glands, after whole body exposure to an acute dose of 10cGy ionizing radiation. Estrus cycle was normalized in all mice two days prior to irradiation by injection with an estrogen and progesterone mixture. | 0.48797276616096497\n",
      "(1.0, 0.7042104005813599)\n"
     ]
    }
   ],
   "source": [
    "inspect_retrieval_eval(\n",
    "    row=simplify_response(load_responses(\"../data/osdr-eval/responses-gpt-3.5-turbo.csv\")).iloc[0],\n",
    "    func=compute_recall_semantic,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "e2d5d680-5ce7-4909-8026-46087fbc711a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>context</th>\n",
       "      <th>question</th>\n",
       "      <th>answer</th>\n",
       "      <th>prediction</th>\n",
       "      <th>retrieved_contexts</th>\n",
       "      <th>recall</th>\n",
       "      <th>score</th>\n",
       "      <th>threshold</th>\n",
       "      <th>top_k</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Genome-wide transcriptional profiling showed t...</td>\n",
       "      <td>Where were the gene expression alterations in ...</td>\n",
       "      <td>reducing gravity levels in the International S...</td>\n",
       "      <td>The gene expression alterations in Drosophila ...</td>\n",
       "      <td>[26-Jun-2013  notifications share  GeneLab ID ...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.709230</td>\n",
       "      <td>0.5</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Genome-wide transcriptional profiling showed t...</td>\n",
       "      <td>What kind of experiments showed differential r...</td>\n",
       "      <td>simulation experiments on ground testing space...</td>\n",
       "      <td>Experiments involving reducing gravity levels ...</td>\n",
       "      <td>[26-Jun-2013  notifications share  GeneLab ID ...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.701952</td>\n",
       "      <td>0.5</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Genome-wide transcriptional profiling showed t...</td>\n",
       "      <td>What is overrepresented in the different exper...</td>\n",
       "      <td>the same GO groups including a large multigene...</td>\n",
       "      <td>The same GO groups, including a large multigen...</td>\n",
       "      <td>[Curiously, although &lt;b&gt;particular&lt;/b&gt; &lt;b&gt;gene...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.706340</td>\n",
       "      <td>0.5</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Genome-wide transcriptional profiling showed t...</td>\n",
       "      <td>What are the implications of the findings in t...</td>\n",
       "      <td>The findings indicate that the finely adjusted...</td>\n",
       "      <td>The findings suggest that the transcriptome is...</td>\n",
       "      <td>[These results suggest that, in addition to be...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.722527</td>\n",
       "      <td>0.5</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>The objective of the Rodent Research-5 (RR-5) ...</td>\n",
       "      <td>What was the main goal of the Rodent Research-...</td>\n",
       "      <td>to evaluate bone loss in mice during spaceflig...</td>\n",
       "      <td>The main goal of the Rodent Research-5 (RR-5) ...</td>\n",
       "      <td>[12-Aug-2019  notifications share  GeneLab ID ...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.587473</td>\n",
       "      <td>0.5</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                             context  \\\n",
       "0  Genome-wide transcriptional profiling showed t...   \n",
       "1  Genome-wide transcriptional profiling showed t...   \n",
       "2  Genome-wide transcriptional profiling showed t...   \n",
       "3  Genome-wide transcriptional profiling showed t...   \n",
       "4  The objective of the Rodent Research-5 (RR-5) ...   \n",
       "\n",
       "                                            question  \\\n",
       "0  Where were the gene expression alterations in ...   \n",
       "1  What kind of experiments showed differential r...   \n",
       "2  What is overrepresented in the different exper...   \n",
       "3  What are the implications of the findings in t...   \n",
       "4  What was the main goal of the Rodent Research-...   \n",
       "\n",
       "                                              answer  \\\n",
       "0  reducing gravity levels in the International S...   \n",
       "1  simulation experiments on ground testing space...   \n",
       "2  the same GO groups including a large multigene...   \n",
       "3  The findings indicate that the finely adjusted...   \n",
       "4  to evaluate bone loss in mice during spaceflig...   \n",
       "\n",
       "                                          prediction  \\\n",
       "0  The gene expression alterations in Drosophila ...   \n",
       "1  Experiments involving reducing gravity levels ...   \n",
       "2  The same GO groups, including a large multigen...   \n",
       "3  The findings suggest that the transcriptome is...   \n",
       "4  The main goal of the Rodent Research-5 (RR-5) ...   \n",
       "\n",
       "                                  retrieved_contexts  recall     score  \\\n",
       "0  [26-Jun-2013  notifications share  GeneLab ID ...     1.0  0.709230   \n",
       "1  [26-Jun-2013  notifications share  GeneLab ID ...     1.0  0.701952   \n",
       "2  [Curiously, although <b>particular</b> <b>gene...     1.0  0.706340   \n",
       "3  [These results suggest that, in addition to be...     1.0  0.722527   \n",
       "4  [12-Aug-2019  notifications share  GeneLab ID ...     1.0  0.587473   \n",
       "\n",
       "   threshold  top_k  \n",
       "0        0.5     10  \n",
       "1        0.5     10  \n",
       "2        0.5     10  \n",
       "3        0.5     10  \n",
       "4        0.5     10  "
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# eval_retrieval_semantic = load_retrieval_eval_semantic(\"../data/osdr-eval/evals/retrieval/bertscore/\")\n",
    "eval_retrieval_semantic = load_retrieval_eval(\"../data/osdr-eval/osdr-eval-retrieval-semantic.csv\")\n",
    "eval_retrieval_semantic.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "0b6283c0-02da-4a29-a45a-8b081564f50d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>recall</th>\n",
       "      <th>score</th>\n",
       "      <th>threshold</th>\n",
       "      <th>top_k</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3444.000000</td>\n",
       "      <td>3444.000000</td>\n",
       "      <td>3444.000000</td>\n",
       "      <td>3444.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.527294</td>\n",
       "      <td>0.581817</td>\n",
       "      <td>0.575000</td>\n",
       "      <td>5.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.499327</td>\n",
       "      <td>0.107731</td>\n",
       "      <td>0.075011</td>\n",
       "      <td>3.682322</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.357976</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.490861</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.572280</td>\n",
       "      <td>0.575000</td>\n",
       "      <td>5.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.675793</td>\n",
       "      <td>0.650000</td>\n",
       "      <td>10.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.936264</td>\n",
       "      <td>0.650000</td>\n",
       "      <td>10.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            recall        score    threshold        top_k\n",
       "count  3444.000000  3444.000000  3444.000000  3444.000000\n",
       "mean      0.527294     0.581817     0.575000     5.333333\n",
       "std       0.499327     0.107731     0.075011     3.682322\n",
       "min       0.000000     0.357976     0.500000     1.000000\n",
       "25%       0.000000     0.490861     0.500000     1.000000\n",
       "50%       1.000000     0.572280     0.575000     5.000000\n",
       "75%       1.000000     0.675793     0.650000    10.000000\n",
       "max       1.000000     0.936264     0.650000    10.000000"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "eval_retrieval_semantic.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "f622112c-f771-4ce7-b36a-19a1ff8fbbf2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# eval_retrieval_semantic.to_csv(\"../data/osdr-eval/evals/osdr-eval-retrieval-semantic.csv\", index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "id": "dedf00e4-f377-42a7-9402-8c12fa8837b2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<h3>Good Recall (Passage Overlap Threshold=0.65, top_k=1)</h3"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Question</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Why was in vivo pQCT used?\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Generated Answer</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "In vivo pQCT was used to measure trabecular bone density at the proximal tibia and assess the structural and functional skeletal alterations in rats exposed to varying degrees of partial weight bearing over different time periods.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Actual Answer</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "To find trabecular bone density\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Actual Context</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reduced skeletal loading leads to marked bone loss. Animal models of hindlimb suspension are widely used to assess alterations in skeleton during the course of complete unloading. More recently, the effects of partial unloading on the musculoskeletal system have been interrogated in mice and rats, revealing dose-dependent effects of partial weight bearing (PWB) on the skeleton and skeletal muscle. Here, we extended these studies to determine the structural and functional skeletal alterations in 14-week-old male Wister rats exposed to 20%, 40%, 70%, or 100% of body weight for 1, 2, or 4 weeks (n  equals 11–12/group). Using in vivo pQCT, we found that trabecular bone density at the proximal tibia declined in proportion to the degree of unloading and continued progressively with time, without evidence of a plateau by 4 weeks. Ex vivo measurements of trabecular microarchitecture in the distal femur by microcomputed tomography revealed deficits in bone volume fraction, 2 and 4 weeks after unloading. Histologic analyses of trabecular bone in the distal femur revealed the decreased osteoblast number and mineralizing surface in unloaded rats. Three-point bending of the femoral diaphysis indicated modest or no reductions in femoral stiffness and estimated modulus due to PWB. Our results suggest that this rat model of PWB leads to trabecular bone deterioration that is progressive and generally proportional to the degree of PWB, with minimal effects on cortical bone. This study derives results from the Mechanical Testing, Bone Microstructure (microCT), and bone histomorphometry assays using femur tissue.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #1</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<b>Using</b> <b>in</b> <b>vivo</b> <b>pQCT</b>, we found that trabecular bone density at the proximal tibia declined in proportion to the degree of unloading and continued progressively with time, without evidence of a plateau by 4 weeks. Ex vivo measurements of trabecular microarchitecture in the distal femur by microcomputed tomography revealed deficits in bone volume fraction, 2 and 4 weeks after unloading. Histologic analyses of trabecular bone in the distal femur revealed the decreased osteoblast number and mineralizing surface in unloaded rats. Three-point bending of the femoral diaphysis indicated modest or no reductions in femoral stiffness and estimated modulus due to PWB.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #2</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "More recently, the effects of partial unloading on the musculoskeletal system have been interrogated in mice and rats, revealing dose-dependent effects of partial weight bearing (PWB) on the skeleton and skeletal muscle. Here, we extended these studies to determine the structural and functional skeletal alterations in 14-week-old male Wister rats exposed to 20%, 40%, 70%, or 100% of body weight for 1, 2, or 4 weeks (n equals 11-12/group). <b>Using</b> <b>in</b> <b>vivo</b> <b>pQCT</b>, we found that trabecular bone density at the proximal tibia declined in proportion to the degree of unloading and continued progressively with time, without\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #3</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5.0, 5.5, 6.0 millimeter  Proximal Tibia Plateau  20.0, 20.5 millimeter  100 micrometer  Not Available  LSDS-41_peripheral_quantitative_computed_tomography_pQCT_LSDS-41_<b>pQCT_KoSUBMITTED.xlsx</b>  ALSDA Transformed Data  LSDS-41_peripheral_quantitative_computed_tomography_pQCT_LSDS-41_<b>pQCT_KoTRANSFORMED.csv</b>  Items per page:  25  1 - 25 of 120  menu_book Publications  Dose-dependent skeletal deficits due to varied reductions in mechanical loading in rats  Authors : Ko FC, Mortreux M, Riveros D, Nagy JA, Rutkove SB, Bouxsein ML.  DOI : 10.1038/s41526-020-0105-0  sim_card_download Files  Study Files Selected: 0  Search Files  search  OSD-608  Study Metadata Files  <b>peripheral_quantitative_computedtomography_pQCT</b>  view_timeline Version History  Selected Version  Version 1  Updated Date: 17-May-2023  Changes: New data release\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #4</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The data were analyzed using BRB-Array Tools (NIH), and further gene ontology analysis and network analysis was performed with Panther (Applied Biosystems) and IPA (Ingenuity), accordingly. Significantly responding genes were identified at all distances and included sets common to both direct and bystander responses. False discovery rate in bystander samples did not exceed 20% (p=0.001) and was sufficiently low in the samples obtained after the whole tissue exposure (0.06-1.16%). Analysis of the fragments cut at the same distance revealed 52, 54 and 88 differentially expressed genes.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #5</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Gamma irradiations were completed at 8:40 am, while HZE irradiations were conducted in two runs (due to space limitations) which were completed at 1:09 and 1:28pm respectively. Gamma treated seedlings were sampled at 10:10 am, 11:40 am, 2:55 pm, 8:40 pm, and 8:40 am. HZE treated seedlings were sampled at 2:39 pm, 4:09 pm, 7:24 pm, 1:09 am, and 1:09 pm. Un-irradiated WT and atm-1 control seedlings were sampled at 10:45 am on Day #1 and 9:15 am on Day #2.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #6</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PCR) methods after propidium monoazide (PMA) treatment. The 16S-tag and metagenome analyses were used to elucidate viable microbial diversity. The cultivable bacterial population yield from the polyester wipes was very high (5 to 7-logs) when compared with the contact slides (10^2 to 10^3 CFU/m2). The PMA-qPCR analysis showed considerable variation of viable bacterial population (10^5 to 10^9 16S rDNA gene copies/m2) among locations sampled.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #7</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "We performed comparative RNA-Seq transcriptomic analysis to assess the carcinogenic effects of 600 MeV/n 56Fe (0.2 Gy), 1 GeV/n 16O (0.2 Gy), and 350 MeV/n 28Si (0.2 Gy) ions in a mouse model for radiation-induced hepatocellular carcinoma. C3H/HeNCrl mice were subjected to total body irradiation to simulate space HZE-irradiation environment and liver tissues were extracted at five different time points post-irradiation to investigate the time-dependent gradual carcinogenic response at the transcriptomic level.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #8</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "In order to identify changes in gene expression that may be informative for detecting such exposure, and to begin examining the molecular responses involved, we have profiled global gene expression in mice injected with 137CsCl. We extracted RNA from the blood of control or 137CsCl-injected mice at 2, 3, 5, 20 or 30 days after exposure. Gene expression was measured using Agilent Whole Mouse Genome Microarrays, and the data was analyzed using BRB-ArrayTools.\n"
     ]
    }
   ],
   "source": [
    "peek_evaluation_result(eval_retrieval_semantic, peek_type=\"good\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "id": "ad358d5c-77e5-43e9-bd0f-f13e24e7e704",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<h3>Bad Recall (Passage Overlap Threshold=0.5, top_k=1)</h3"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Question</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "What does WS stand for?\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Generated Answer</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "In the context provided, WS likely stands for a specific genotype or ecotype of Arabidopsis thaliana, a type of plant used in the study.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Actual Answer</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wassilewskija.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Actual Context</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The increasing availability of flights on suborbital rockets creates new avenues for the study of spaceflight effects on biological systems, in particular the transitions between hypergravity and microgravity. This paper presents an initial comparison of the responses of Arabidopsis thaliana to suborbital and atmospheric parabolic flights as an important step toward characterizing these emerging suborbital platforms and their effects on biology. Transcriptomic profiling of the response of the Arabidopsis ecotype Wassilewskija (WS) to the aggregate spaceflight experiences in the Blue Origin New Shepard and Virgin Galactic SpaceShipTwo rockets revealed that the transcriptomic load induced by flight differed greatly between the two flights, yet was biologically related to traditional parabolic flight responses. The sku5 skewing mutant and 14-3-3κ:GFP regulatory protein overexpression lines each showed altered intra-platform responses compared to WS in the Blue Origin and parabolic flights, respectively. An additional parabolic flight using the F-104 Starfighter showed that the response of 14-3-3κ:GFP to flight was modulated in a similar manner to the WS line. Despite the differing genotypes, experimental workflows, flight profiles and platforms, alteration of gene expression remodeling central metabolic processes was commonly observed as a response to the flights. The processes included carbon and nitrogen metabolism, branched-chain amino acid degradation, and hypoxic responses. The timing and directionality of differentially-expressed genes involved in the conserved pathways differed among the platforms. The data presented herein highlight the potential for various suborbital platforms to contribute insights into biological responses to spaceflight, and further suggest that in-flight fixation during suborbital experiments will provide insights into responses to each phase of flight.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #1</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The EXPRESS Rack  60% <b>WS</b>, 40% Col-0  ABRS  22 -24 degree Celsius  12 day  below -40 degree Celsius  Space Flight  hypocotyl  RNALater  Plants  <b>Atha_WS</b>-0_Col-0_Hypocotyl_FLT_Rep4  F4H2  P-MTAB-28101  Arabidopsis thaliana  solid 0.5x MS media plates; Advanced Biological Research System (ABRS); The EXPRESS Rack  60% <b>WS</b>, 40% Col-0  ABRS  22 -24 degree Celsius  12 day  below -40 degree Celsius  Space Flight  hypocotyl  RNALater  Plants  <b>Atha_WS</b>-0_Col-0_Hypocotyl_FLT_Rep5  F6H2  P-MTAB-28101  Arabidopsis thaliana  solid 0.5x MS media plates; Advanced Biological Research System (ABRS); The EXPRESS Rack  60% <b>WS</b>, 40% Col-0  ABRS  22 -24 degree Celsius  12 day\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #2</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "These differential genotypic responses suggest that genetic manipulation could further reduce, or perhaps eliminate the metabolic cost of spaceflight adaptation. When plants were germinated and then left in the dark on the ISS, the <b>WS</b> genotype actually mounted a larger transcriptome response than Col-0, suggesting that the in-space light environment affects physiological adaptation, which implies that manipulating the local habitat can also substantially impact the metabolic cost of spaceflight adaptation.  Factor(s)  Factor  Ontology: Concept  Ecotype  Ecotype  Spaceflight  Space Flight  Treatment  treatment  Organism(s)  Arabidopsis thaliana  Assay(s)  Measurement  Technology  Device Platform  transcription profiling  RNA Sequencing (RNA-Seq)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #3</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<b>Atha_WS</b>-0_Col-0_Shoot_GC_Rep5  G6S2  P-MTAB-28101  Arabidopsis thaliana  solid 0.5x MS media plates; Advanced solid media plates; Advanced Biological Research System (ABRS); GC Unit, OES chamber at KSC  60% <b>WS</b>, 40% Col-0  ABRS  22 -24 degree Celsius  12 day  below -40 degree Celsius  Ground Control  shoot  RNALater  Plants  <b>Atha_WS</b>-0_Col-0_Hypocotyl_FLT_Rep1  F2H1  P-MTAB-28101  Arabidopsis thaliana  solid 0.5x MS media plates; Advanced Biological Research System (ABRS); The EXPRESS Rack  60% <b>WS</b>, 40% Col-0  ABRS  22 -24 degree Celsius  12 day  below -40 degree Celsius  Space Flight  hypocotyl  RNALater  Plants  <b>Atha_WS</b>-0_Col-0_Hypocotyl_FLT_Rep2  F4H1\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #4</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The EXPRESS Rack  60% <b>WS</b>, 40% Col-0  ABRS  22 -24 degree Celsius  12 day  below -40 degree Celsius  Space Flight  shoot  RNALater  Plants  <b>Atha_WS</b>-0_Col-0_Shoot_FLT_Rep4  F4S2  P-MTAB-28101  Arabidopsis thaliana  solid 0.5x MS media plates; Advanced Biological Research System (ABRS); The EXPRESS Rack  60% <b>WS</b>, 40% Col-0  ABRS  22 -24 degree Celsius  12 day  below -40 degree Celsius  Space Flight  shoot  RNALater  Plants  <b>Atha_WS</b>-0_Col-0_Shoot_FLT_Rep5  F6S2  P-MTAB-28101  Arabidopsis thaliana  solid 0.5x MS media plates; Advanced Biological Research System (ABRS); The EXPRESS Rack  60% <b>WS</b>, 40% Col-0  ABRS  22 -24 degree Celsius  12 day\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #5</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ABRS  22 -24 degree Celsius  12 day  below -40 degree Celsius  Space Flight  hypocotyl  RNALater  Plants  <b>Atha_WS</b>-0_Col-0_Hypocotyl_FLT_Rep2  F4H1  P-MTAB-28101  Arabidopsis thaliana  solid 0.5x MS media plates; Advanced Biological Research System (ABRS); The EXPRESS Rack  60% <b>WS</b>, 40% Col-0  ABRS  22 -24 degree Celsius  12 day  below -40 degree Celsius  Space Flight  hypocotyl  RNALater  Plants  <b>Atha_WS</b>-0_Col-0_Hypocotyl_FLT_Rep3  F6H1  P-MTAB-28101  Arabidopsis thaliana  solid 0.5x MS media plates; Advanced Biological Research System (ABRS); The EXPRESS Rack  60% <b>WS</b>, 40% Col-0  ABRS  22 -24 degree Celsius  12 day  below -40 degree Celsius  Space Flight  hypocotyl  RNALater  Plants\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #6</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Plants  <b>Atha_WS</b>-0_Whole-Plant_GC_Rep3  GCTAGES6  P-MTAB-28101  Arabidopsis thaliana  solid 0.5x MS media plates; Advanced solid media plates; Advanced Biological Research System (ABRS); GC Unit, OES chamber at KSC  <b>WS</b>  ABRS  22 -24 degree Celsius  12 day  below -40 degree Celsius  Ground Control  whole plant  RNALater  Plants  <b>Atha_WS</b>-0_Col-0_Shoot_FLT_Rep1  F2S1  P-MTAB-28101  Arabidopsis thaliana  solid 0.5x MS media plates; Advanced Biological Research System (ABRS); The EXPRESS Rack  60% <b>WS</b>, 40% Col-0  ABRS  22 -24 degree Celsius  12 day  below -40 degree Celsius  Space Flight  shoot  RNALater  Plants  <b>Atha_WS</b>-0_Col-0_Shoot_FLT_Rep2  F4S1  P-MTAB-28101\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #7</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Advanced Biological Research System (ABRS); The EXPRESS Rack  <b>WS</b>  ABRS  22 -24 degree Celsius  12 day  below -40 degree Celsius  Space Flight  whole plant  RNALater  Plants  <b>Atha_WS</b>-0_Whole-Plant_FLT_Rep3  FLTAGES6  P-MTAB-28101  Arabidopsis thaliana  solid 0.5x MS media plates; Advanced Biological Research System (ABRS); The EXPRESS Rack  <b>WS</b>  ABRS  22 -24 degree Celsius  12 day  below -40 degree Celsius  Space Flight  whole plant  RNALater  Plants  <b>Atha_WS</b>-0_Whole-Plant_GC_Rep1  GCTAGES2  P-MTAB-28101  Arabidopsis thaliana  solid 0.5x MS media plates; Advanced solid media plates; Advanced Biological Research System (ABRS); GC Unit, OES chamber at KSC  <b>WS</b>  ABRS  22 -24 degree Celsius  12 day\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #8</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "solid 0.5x MS media plates; Advanced Biological Research System (ABRS); The EXPRESS Rack  60% <b>WS</b>, 40% Col-0  ABRS  22 -24 degree Celsius  12 day  below -40 degree Celsius  Space Flight  shoot  RNALater  Plants  <b>Atha_WS</b>-0_Col-0_Shoot_GC_Rep1  G2S1  P-MTAB-28101  Arabidopsis thaliana  solid 0.5x MS media plates; Advanced solid media plates; Advanced Biological Research System (ABRS); GC Unit, OES chamber at KSC  60% <b>WS</b>, 40% Col-0  ABRS  22 -24 degree Celsius  12 day  below -40 degree Celsius  Ground Control  shoot  RNALater  Plants  <b>Atha_WS</b>-0_Col-0_Shoot_GC_Rep2  G4S1  P-MTAB-28101  Arabidopsis thaliana  solid 0.5x MS media plates;\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #9</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "22 -24 degree Celsius  12 day  below -40 degree Celsius  Space Flight  shoot  RNALater  Plants  <b>Atha_WS</b>-0_Col-0_Shoot_FLT_Rep2  F4S1  P-MTAB-28101  Arabidopsis thaliana  solid 0.5x MS media plates; Advanced Biological Research System (ABRS); The EXPRESS Rack  60% <b>WS</b>, 40% Col-0  ABRS  22 -24 degree Celsius  12 day  below -40 degree Celsius  Space Flight  shoot  RNALater  Plants  <b>Atha_WS</b>-0_Col-0_Shoot_FLT_Rep3  F6S1  P-MTAB-28101  Arabidopsis thaliana  solid 0.5x MS media plates; Advanced Biological Research System (ABRS); The EXPRESS Rack  60% <b>WS</b>, 40% Col-0  ABRS  22 -24 degree Celsius  12 day  below -40 degree Celsius  Space Flight  shoot  RNALater  Plants\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h4>Retrieved Context #10</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The whole-genome sequences (WGS) of three pathogenic bacterial strains collected from the International Space Station (ISS) were generated and identified as being part of Class Alphaproteobacteria. The WGS studies would enable further genomic/biochemical characterization of taxa for these bacteria that have been subjected to the microgravity conditions of space.  Factor(s)  Factor  Ontology: Concept  Spaceflight  Space Flight  Strain  strain  Organism(s)  Agrobacterium genomosp. 3  Assay(s)  Measurement  Technology  Device Platform  genome sequencing  nucleotide sequencing  Illumina NovaSeq 6000  Project  Payload Identifier  ISS-MO (MT-1B)  Project Title  ISS Microbial Observatory (Microbial Tracking-1B)  Project Type  Spaceflight Study  Flight Program\n"
     ]
    }
   ],
   "source": [
    "peek_evaluation_result(eval_retrieval_semantic, peek_type=\"bad\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "1e1196b9-39d3-43e3-bc4e-0f2b29145b0d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_mean_recall_semantic(eval_retrieval_semantic)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "47a2e922-52a8-4e08-b449-e69873fc5276",
   "metadata": {},
   "source": [
    "## QA Evaluation\n",
    "\n",
    "Here we focus on semantic evaluation of generated answers which suggest the correctness of the answer w.r.t reference answer.\n",
    "\n",
    "We use following metrics:\n",
    "\n",
    "\n",
    "1. **Exact Match**\n",
    "2. **METEOR** (Metric for Evaluation of Translation with Explicit Ordering)\n",
    "    - more aligned with human judgment\n",
    "    - accounts for exact word matches, *synonyms*, and *paraphrases*\n",
    "    - useful for evaluating correctness across varied wordings\n",
    "3. **ROUGE** (Recall-Oriented Understudy for Gisting Evaluation)\n",
    "   - focuses on recall, measuring *word overlaps* with reference\n",
    "   - indicates content overlap rather than linguistic quality\n",
    "   - lower score may suggest fewer exact word matches\n",
    "4. **BERT Score**\n",
    "    - *semantic* similarity using contextual embeddings\n",
    "    - measures how well the generated answers match with references in meaning\n",
    "    - generalized comparison for a variety of downstream tasks (QA, Translation, Summarization, etc.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "d6df79dc-7f47-423f-8544-9080a61b6865",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<h2>Inspecting the metrics</h2>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h3>Question</h3>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "How did radiation alter indicators of apoptosis?\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h3>Expected Answer</h3>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Each of the radiation exposures caused only small increases in cleaved caspase3 and numbers of apoptotic nuclei.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h3>Predicted Answer</h3>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Radiation altered indicators of apoptosis by causing qualitative and quantitative differences in the down-regulation of miRNA levels, including miR-92b, 137, 660, and 656. Additionally, microRNAs miR-579, 608, 548-3p, and 585 were noted for targeting genes involved in radioresponsive mechanisms, such as cell cycle checkpoint and apoptosis. These changes in miRNA levels and targeting genes suggest that radiation exposure can impact the regulation of apoptosis.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.10/site-packages/jury/metrics/_core/auto.py:106: UserWarning: Metric exact_match is not available on jury, falling back to evaluate metric. You may not fully utilize this metric for different input types, e.g multiple predictions or multiple references.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h3>ExactMatchMetric</h3>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h3>BertScore</h3>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.523653507232666\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h3>MeteorMetric</h3>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.24073812217194573\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h3>RougeMetric</h3>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.09411764705882353\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h2>Inspecting the metrics</h2>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h3>Question</h3>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "What is overrepresented in the different experiments despite particular genes not being common?\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h3>Expected Answer</h3>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the same GO groups including a large multigene family related with behavior, stress response and organogenesis\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h3>Predicted Answer</h3>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The same GO groups, including a large multigene family related to behavior, stress response, and organogenesis, are overrepresented in the different experiments despite particular genes not being common.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h3>ExactMatchMetric</h3>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h3>BertScore</h3>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.7794027328491211\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h3>MeteorMetric</h3>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8686046511627907\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<h3>RougeMetric</h3>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.666125541125541\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<hr/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# what the scores mean\n",
    "def inspect_metrics(df):\n",
    "    df = df.copy()\n",
    "    references = df.answer.to_list()\n",
    "    predictions = df.prediction.to_list()\n",
    "    questions = df.question.to_list()\n",
    "\n",
    "    display(HTML(\"<h2>Inspecting the metrics</h2>\"))\n",
    "    display(HTML(f\"<h3>Question</h3>\"))\n",
    "    print(questions[0])\n",
    "    display(HTML(f\"<h3>Expected Answer</h3>\"))\n",
    "    print(references[0])\n",
    "    display(HTML(f\"<h3>Predicted Answer</h3>\"))\n",
    "    print(predictions[0])\n",
    "\n",
    "    evaluator = Evaluator(metrics=[\n",
    "        ExactMatchMetric(),\n",
    "        BertScore(),\n",
    "        MeteorMetric(),\n",
    "        RougeMetric(),\n",
    "    ])\n",
    "\n",
    "    results = evaluator(references=references, predictions=predictions)\n",
    "\n",
    "    for result in results:\n",
    "        score = result.score\n",
    "        name = result.metric_name\n",
    "        display(HTML(f\"<h3>{name}</h3>\"))\n",
    "        print(score)\n",
    "    display(HTML(\"<hr/>\"))\n",
    "    \n",
    "\n",
    "inspect_metrics(eval_retrieval_semantic.iloc[100:101])\n",
    "inspect_metrics(eval_retrieval_semantic.iloc[2:3])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "186a8cfa-56f0-4939-ad9f-3d70a3de68c2",
   "metadata": {},
   "outputs": [],
   "source": [
    "def evaluate_qa(response: pd.DataFrame) -> pd.DataFrame:\n",
    "    if \"question\" not in response and \"prediction\" not in response and \"reference\" not in response:\n",
    "        response = simplify_response(response)\n",
    "    evaluator = Evaluator(metrics=[\n",
    "        BertScore(),\n",
    "        ExactMatchMetric(),\n",
    "        MeteorMetric(),\n",
    "        RougeMetric(),\n",
    "    ])\n",
    "    results = evaluator(\n",
    "        references=response[\"reference\"].to_list(),\n",
    "        predictions=response[\"prediction\"].to_list()\n",
    "    )\n",
    "    results = pd.DataFrame(results)\n",
    "    return results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "9e8d3d4a-6318-49f1-8805-8fdaad112a16",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_eval_result(df: pd.DataFrame):\n",
    "    plt.figure(figsize=(12, 8))\n",
    "    barplot = sns.barplot(x='metric_name', y='score', data=df, palette='husl')\n",
    "    for p in barplot.patches:\n",
    "        barplot.annotate(format(p.get_height(), '.3f'), \n",
    "                         (p.get_x() + p.get_width() / 2., p.get_height()), \n",
    "                         ha = 'center', va = 'center', \n",
    "                         xytext = (0, 9), \n",
    "                         textcoords = 'offset points')\n",
    "    plt.title('Comparison of Metrics')\n",
    "    plt.xlabel('Metric')\n",
    "    plt.ylabel('Score')\n",
    "    plt.grid(True, which='both', axis='y', linestyle='--', linewidth=0.5)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "00040e3a-6d2c-4111-82d2-2a57a05aa5fc",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.10/site-packages/jury/metrics/_core/auto.py:106: UserWarning: Metric exact_match is not available on jury, falling back to evaluate metric. You may not fully utilize this metric for different input types, e.g multiple predictions or multiple references.\n",
      "  warnings.warn(\n"
     ]
    }
   ],
   "source": [
    "eval_qa = evaluate_qa(\n",
    "    response=pd.read_csv(\"../data/osdr-eval/responses-simplified-gpt-3.5-turbo-rephrased.csv\")\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "a8e93e54-57a1-44ae-b866-170efb206e07",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>score</th>\n",
       "      <th>total_items</th>\n",
       "      <th>metric_name</th>\n",
       "      <th>empty_items</th>\n",
       "      <th>extra</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.665739</td>\n",
       "      <td>295</td>\n",
       "      <td>BertScore</td>\n",
       "      <td>0</td>\n",
       "      <td>{'bertscore': {'score': 0.6657390932915574, 'p...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.037288</td>\n",
       "      <td>295</td>\n",
       "      <td>ExactMatchMetric</td>\n",
       "      <td>0</td>\n",
       "      <td>{'exact_match': 0.03728813559322034, 'flattene...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.510066</td>\n",
       "      <td>295</td>\n",
       "      <td>MeteorMetric</td>\n",
       "      <td>0</td>\n",
       "      <td>{'meteor': {'score': 0.5100657697400938}}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.372489</td>\n",
       "      <td>295</td>\n",
       "      <td>RougeMetric</td>\n",
       "      <td>0</td>\n",
       "      <td>{'rouge': {'rouge1': 0.42175915507801653, 'rou...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      score  total_items       metric_name  empty_items  \\\n",
       "0  0.665739          295         BertScore            0   \n",
       "1  0.037288          295  ExactMatchMetric            0   \n",
       "2  0.510066          295      MeteorMetric            0   \n",
       "3  0.372489          295       RougeMetric            0   \n",
       "\n",
       "                                               extra  \n",
       "0  {'bertscore': {'score': 0.6657390932915574, 'p...  \n",
       "1  {'exact_match': 0.03728813559322034, 'flattene...  \n",
       "2          {'meteor': {'score': 0.5100657697400938}}  \n",
       "3  {'rouge': {'rouge1': 0.42175915507801653, 'rou...  "
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "eval_qa"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "3b8cce33-18d2-4093-9a20-35a389baa323",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_5640/3910356460.py:3: FutureWarning: \n",
      "\n",
      "Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the same effect.\n",
      "\n",
      "  barplot = sns.barplot(x='metric_name', y='score', data=df, palette='husl')\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_eval_result(eval_qa)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b9cbd5e2-7ddd-4285-a987-c4edf15335f9",
   "metadata": {},
   "source": [
    "# Footnotes\n",
    "- [evalem](https://github.com/NASA-IMPACT/evalem)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}