Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Refactor checkpoint conversion script for improved readability and efficiency #633

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
56 changes: 34 additions & 22 deletions inference/convert.py
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,7 @@
import shutil
from argparse import ArgumentParser
from glob import glob
from pathlib import Path
from tqdm import tqdm, trange

import torch
Expand Down Expand Up @@ -43,46 +44,55 @@ def main(hf_ckpt_path, save_path, n_experts, mp):
Returns:
None
"""
assert mp > 0, "Model parallelism (mp) must be greater than 0"

torch.set_num_threads(8)
n_local_experts = n_experts // mp
state_dicts = [{} for _ in range(mp)]

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Add a comment to define state_dicts variable


hf_ckpt_path = Path(hf_ckpt_path)
save_path = Path(save_path)
save_path.mkdir(parents=True, exist_ok=True)

for file_path in tqdm(glob(os.path.join(hf_ckpt_path, "*.safetensors"))):
for file_path in tqdm(hf_ckpt_path.glob("*.safetensors")):
with safe_open(file_path, framework="pt", device="cpu") as f:
for name in f.keys():
if "model.layers.61" in name:
continue

param: torch.Tensor = f.get_tensor(name)

if name.startswith("model."):
name = name[len("model."):]
name = name.replace("self_attn", "attn")
name = name.replace("mlp", "ffn")
name = name.replace("weight_scale_inv", "scale")
name = name.replace("e_score_correction_bias", "bias")

name = (
name.replace("self_attn", "attn")
.replace("mlp", "ffn")
.replace("weight_scale_inv", "scale")
.replace("e_score_correction_bias", "bias")
)

key = name.split(".")[-2]
assert key in mapping, f"Key {key} not found in mapping"
new_key, dim = mapping[key]
name = name.replace(key, new_key)
for i in range(mp):
new_param = param
if "experts" in name and "shared_experts" not in name:
idx = int(name.split(".")[-3])
if idx < i * n_local_experts or idx >= (i + 1) * n_local_experts:
continue
elif dim is not None:
assert param.size(dim) % mp == 0, f"Dimension {dim} must be divisible by {mp}"
shard_size = param.size(dim) // mp
new_param = param.narrow(dim, i * shard_size, shard_size).contiguous()
state_dicts[i][name] = new_param

os.makedirs(save_path, exist_ok=True)

if "experts" in name and "shared_experts" not in name:
idx = int(name.split(".")[-3])
target_index = idx // n_local_experts
if target_index < mp:
state_dicts[target_index][name] = param
elif dim is not None:
assert param.size(dim) % mp == 0, f"Dimension {dim} must be divisible by {mp}"
shard_size = param.size(dim) // mp
for i in range(mp):
state_dicts[i][name] = param[:, i * shard_size : (i + 1) * shard_size] if dim == 1 else param[i * shard_size : (i + 1) * shard_size]
Comment on lines +80 to +89

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We can add comments here for better clarity.


for i in trange(mp):
save_file(state_dicts[i], os.path.join(save_path, f"model{i}-mp{mp}.safetensors"))
save_file(state_dicts[i], save_path / f"model{i}-mp{mp}.safetensors")

for file_path in glob(os.path.join(hf_ckpt_path, "*token*")):
new_file_path = os.path.join(save_path, os.path.basename(file_path))
shutil.copyfile(file_path, new_file_path)
for file_path in hf_ckpt_path.glob("*token*"):
shutil.copyfile(file_path, save_path / file_path.name)


if __name__ == "__main__":
Expand All @@ -92,5 +102,7 @@ def main(hf_ckpt_path, save_path, n_experts, mp):
parser.add_argument("--n-experts", type=int, required=True)
parser.add_argument("--model-parallel", type=int, required=True)
args = parser.parse_args()

assert args.n_experts % args.model_parallel == 0, "Number of experts must be divisible by model parallelism"

main(args.hf_ckpt_path, args.save_path, args.n_experts, args.model_parallel)