Outlines v0.0.10

What's Changed

• Make sure that indices are available on all relevant devices by @brandonwillard in #328
• Clean repository by @rlouf in #329
• Add Issue and PR templates by @rlouf in #332
• Fix math error in sampling example by @mwesthelle in #334
• Remove the function interface by @rlouf in #341
• fixes BUG #353. Example scripts don't run as expected. by @HerrIvan in #354
• Fix Documentation Inconsistency in Joke Class Example by @MelodiousThonk in #356
• Use MkDocs for the documentation by @rlouf in #331
• Improve the coverage of the JSON Schema specification by @rlouf in #342
• Remove approximate guided generation methods for OpenAI APIs by @rlouf in #357

New Contributors

• @mwesthelle made their first contribution in #334
• @HerrIvan made their first contribution in #354
• @MelodiousThonk made their first contribution in #356

Full Changelog: 0.0.9...0.0.10

Outlines v0.0.9

What's Changed

• Fix ReAct example link path by @lukestanley in #220
• Fix typo in README.md by @eltociear in #235
• Make transformers dependency explicit by @rlouf in #240
• Explicitly list the packages in modules by @rlouf in #242
• Fix some typos and language by @jrysana in #254
• Fix non-matching function definition and call by @krasserm in #252
• Remove unused from typing import List in README.md by @alvarobartt in #256
• Fix some typos/language in regex and sequence by @jrysana in #257
• Allow users to choose devices for models by @BramVanroy in #264
• Escape curly braces in the regexes by @rlouf in #243
• Change default generate behavior from fixed seed to random seed by @mondaychen in #228
• Use FSMs for scanning during grammar-guided generation by @brandonwillard in #178
• Added a fix for mismatched logit and mask tensor shapes by @xaviruvpadhiyar98 in #236
• Misc. Refactoring and Link Updates by @brandonwillard in #279
• Fix missing spaces in Tokenizer.convert_token_to_string by @brandonwillard in #280
• More HF Llama tokenizer space fixes by @brandonwillard in #282
• Fix whitespace and control character handling in JSON guidance by @brandonwillard in #283
• Add py.typed and package data information by @brandonwillard in #294
• Example to generate dating app profiles- combines prompt templating with JSON generation by @veezbo in #287
• Refactor for Python 3.8-3.9 compatibility by @brandonwillard in #295
• Introduce Numba-based FSM utilities by @brandonwillard in #272
• Use device_map value when device is unspecified by @brandonwillard in #300
• Fix typos in model usage examples by @Barahlush in #307
• Modify JSON schema $ref resolution strategy to support resolving values internal to the schema by @mattkindy in #312
• Fix JSON schema generation to accept schemas without 'title' property by @mattkindy in #311
• Update resize padding multiple in test_transformers_logits_vocab_size by @brandonwillard in #313
• Avoid unhashable additional_special_tokens values in special_tokens_map by @brandonwillard in #314
• Remove unnecessary array allocations in generation process and enable caching by @brandonwillard in #308
• Allow configurable sampling steps by @brandonwillard in #318
• Fix some Sequence interface issues by @brandonwillard in #319
• Fix potential REDUCE issues in partial parse states by @brandonwillard in #320
• Add missing sample tests by @brandonwillard in #321
• Support recursive arrays in JSON when an item is an array by @AL-377 in #316
• Fix and simplify walk_fsm by @brandonwillard in #322
• Disable eager Numba FSM info construction step in parser by @brandonwillard in #323
• Add another missing walk_fsm condition by @brandonwillard in #324

New Contributors

• @lukestanley made their first contribution in #220
• @eltociear made their first contribution in #235
• @jrysana made their first contribution in #254
• @krasserm made their first contribution in #252
• @alvarobartt made their first contribution in #256
• @BramVanroy made their first contribution in #264
• @xaviruvpadhiyar98 made their first contribution in #236
• @veezbo made their first contribution in #287
• @Barahlush made their first contribution in #307
• @mattkindy made their first contribution in #312
• @AL-377 made their first contribution in #316

Full Changelog: 0.0.8...0.0.9

Outlines v0.0.8

What's Changed

• Update the README by @rlouf in #202
• Fix error when using type="float" on OpenAI model by @mondaychen in #212
• Add JSON generation method by @rlouf in #140

New Contributors

• @mondaychen made their first contribution in #212

Full Changelog: 0.0.7...0.0.8

Outlines v0.0.7

What's Changed

• Convert tokens to strings before partial matching by @rlouf in #194
• Generate a choice between different strings by @rlouf in #188

Full Changelog: 0.0.6...0.0.7

Outlines v0.0.6

What's Changed

• Add missing requirements by @rlouf in #200

Full Changelog: 0.0.5...0.0.6

Outlines v0.0.5

What's Changed

• Create masks from regex by @rlouf in #124
• Add mps backend for Pytorch in HF models by @arunpatro in #138
• Update development statement by @dgerlanc in #146
• Add coverage checks to CI by @brandonwillard in #149
• Add retry mechanism for openai API calls by @harshkumarchourasia in #129
• Add Transformers model and Completion sequence generation by @rlouf in #139
• Updates for Pydantic 2.0 by @brandonwillard in #168
• Add basic parser-driven masking utilities by @brandonwillard in #131
• Parser-driven masking updates for basic "off-line" vocabulary pre-parsing by @brandonwillard in #171
• Use PyTorch instead of NumPy by @rlouf in #167
• Add Integer sequence generator by @rlouf in #166
• Add Regex generation method by @rlouf in #175
• Fix overview document and a type hint by @mlelarge in #183
• Stop generation with Continuation when a specific string was generated by @rlouf in #187
• Make sure Continuation and Regex use the correct device by @brandonwillard in #195
• Generalize parse state cloning by @brandonwillard in #172
• Clone and patch parse state in example by @brandonwillard in #199

New Contributors

• @arunpatro made their first contribution in #138
• @dgerlanc made their first contribution in #146
• @mlelarge made their first contribution in #183

Full Changelog: 0.0.4...0.0.5

Outlines v0.0.4

What's Changed

• Update README with outlines.vectorize by @rlouf in #133
• Fix vectorization of transformers models by @rlouf in #134

Full Changelog: 0.0.3...0.0.4

Outlines v0.0.3

What's Changed

• Fix bug in https://github.com/normal-computing/outlines#prompting example by @tiendung in #121
• Fix typo in OpenAICompletion.generate_choice docstring by @harshkumarchourasia in #128
• Vectorize function calls by @rlouf in #120

New Contributors

• @tiendung made their first contribution in #121
• @harshkumarchourasia made their first contribution in #128

Full Changelog: 0.0.2...0.0.3

Outlines v0.0.2

Add README to pyproject

Outlines v0.0.1

What's Changed

• Parse scripts and transpile to Outlines graphs by @rlouf in #10
• Add LanguageModel Op and integration with HuggingFace's GPT2 implementation by @rlouf in #30
• Add ImageVariable Op and integrations with HF's StableDiffusion implementation by @rlouf in #33
• Add an interactive execution mode by @rlouf in #34
• Integrate all causal LM models available with HuggingFace's transformers library by @rlouf in #37
• Integration with the OpenAI API by @rlouf in #38
• Simplify and clean the codebase by @rlouf in #40
• Add decorator to turn python functions into an Op by @rlouf in #41
• Add eval method and chain function to compile outlines graphs by @rlouf in #42
• Refactor the language model interface by @rlouf in #44
• Add max_tokens and temperature kwargs for the text.completion operator by @rlouf in #45
• Image generation by @rlouf in #46
• Remove the symbolic structure by @rlouf in #48
• Update the BabyAGI example by @rlouf in #57
• Add integration with the OpenAI embeddings endpoint by @rlouf in #59
• Add integration for OpenAI's image generation API by @rlouf in #60
• Remove whitespaces introduced by the use of \ by @rlouf in #63
• Remove chat completion and refactor the model router by @rlouf in #65
• Add tool decorator to automatically document tools by @rlouf in #64
• Add a utility to build "natural language functions" by @rlouf in #66
• Add an in-memory vector store by @rlouf in #67
• Use Jinja filters instead of a tool decorator by @rlouf in #69
• Use tempfile to generate test cache directories by @brandonwillard in #74
• Add "Pick the odd one out" example by @rlouf in #84
• Update the BabyAGI demo by @brandonwillard in #73
• Add react example by @rlouf in #83
• Remove the completion decorator by @rlouf in #86
• Add documentation draft by @rlouf in #92
• Update the documentation with installation instructions and overview by @rlouf in #94
• Takes several samples from the generative models by @rlouf in #95
• Add type constraint to generation by @rlouf in #91
• Allow disabling caching by @rlouf in #102
• Pretty print dictionary when passed to prompt by @rlouf in #103
• Add tests for the HuggingFace integrations by @rlouf in #105
• Move stop_at parameter to completion call by @brosand in #110
• Add is_in and stop_at generation constraints to transformers integration by @rlouf in #108
• Simplify the OpenAI text completion interface by @rlouf in #112
• Add a Conda dev environment file by @brandonwillard in #114
• Add LICENSE by @rlouf in #111
• Update the README by @rlouf in #113
• Update the documentation before release by @rlouf in #104

New Contributors

• @brandonwillard made their first contribution in #74

Full Changelog: https://github.com/normal-computing/outlines/commits/0.0.1