mirror of
https://github.com/huggingface/text-generation-inference.git
synced 2025-04-19 22:02:06 +00:00
abd58ff82c
# What does this PR do? Reworked the loading logic. Idea is to use cleaner loading code: - Remove need for `no_init_weights` - Remove all weird `bnb_linear` and `load_weights` and `post_load_weights`. New code layout: - New class `Weights` in charge of handling loading the weights from multiple files into appropiate tensors (potentially sharded) - TP layers now are "shells", they contain the code to know what kind of sharding we need + eventual `all_reduce`. They do not inherit from linear, but they contain some kind of Linear instead - the contained linear can be either FastLinear, BnbLinear or GPTq Linear next. - All modeling code is explictly made for sharding, process group is just no-ops for non sharded code (removes a lot of test cases)  --------- Co-authored-by: Ubuntu <ubuntu@ip-172-31-41-161.taildb5d.ts.net> Co-authored-by: Ubuntu <ubuntu@ip-172-31-41-161.ec2.internal> Co-authored-by: OlivierDehaene <olivier@huggingface.co> Co-authored-by: OlivierDehaene <23298448+OlivierDehaene@users.noreply.github.com>
43 lines
3.2 KiB
Plaintext
43 lines
3.2 KiB
Plaintext
backoff==2.2.1 ; python_version >= "3.9" and python_version < "4.0"
|
|
bitsandbytes==0.38.1 ; python_version >= "3.9" and python_version < "4.0"
|
|
certifi==2023.5.7 ; python_version >= "3.9" and python_version < "4.0"
|
|
charset-normalizer==3.1.0 ; python_version >= "3.9" and python_version < "4.0"
|
|
click==8.1.3 ; python_version >= "3.9" and python_version < "4.0"
|
|
colorama==0.4.6 ; python_version >= "3.9" and python_version < "4.0" and sys_platform == "win32" or python_version >= "3.9" and python_version < "4.0" and platform_system == "Windows"
|
|
deprecated==1.2.13 ; python_version >= "3.9" and python_version < "4.0"
|
|
filelock==3.12.0 ; python_version >= "3.9" and python_version < "4.0"
|
|
fsspec==2023.5.0 ; python_version >= "3.9" and python_version < "4.0"
|
|
googleapis-common-protos==1.59.0 ; python_version >= "3.9" and python_version < "4.0"
|
|
grpc-interceptor==0.15.2 ; python_version >= "3.9" and python_version < "4.0"
|
|
grpcio-reflection==1.55.0 ; python_version >= "3.9" and python_version < "4.0"
|
|
grpcio-status==1.55.0 ; python_version >= "3.9" and python_version < "4.0"
|
|
grpcio==1.55.0 ; python_version >= "3.9" and python_version < "4.0"
|
|
hf-transfer==0.1.3 ; python_version >= "3.9" and python_version < "4.0"
|
|
huggingface-hub==0.14.1 ; python_version >= "3.9" and python_version < "4.0"
|
|
idna==3.4 ; python_version >= "3.9" and python_version < "4"
|
|
loguru==0.6.0 ; python_version >= "3.9" and python_version < "4.0"
|
|
opentelemetry-api==1.15.0 ; python_version >= "3.9" and python_version < "4.0"
|
|
opentelemetry-exporter-otlp-proto-grpc==1.15.0 ; python_version >= "3.9" and python_version < "4.0"
|
|
opentelemetry-exporter-otlp-proto-http==1.15.0 ; python_version >= "3.9" and python_version < "4.0"
|
|
opentelemetry-exporter-otlp==1.15.0 ; python_version >= "3.9" and python_version < "4.0"
|
|
opentelemetry-instrumentation-grpc==0.36b0 ; python_version >= "3.9" and python_version < "4.0"
|
|
opentelemetry-instrumentation==0.36b0 ; python_version >= "3.9" and python_version < "4.0"
|
|
opentelemetry-proto==1.15.0 ; python_version >= "3.9" and python_version < "4.0"
|
|
opentelemetry-sdk==1.15.0 ; python_version >= "3.9" and python_version < "4.0"
|
|
opentelemetry-semantic-conventions==0.36b0 ; python_version >= "3.9" and python_version < "4.0"
|
|
packaging==23.1 ; python_version >= "3.9" and python_version < "4.0"
|
|
protobuf==4.23.1 ; python_version >= "3.9" and python_version < "4.0"
|
|
pyyaml==6.0 ; python_version >= "3.9" and python_version < "4.0"
|
|
requests==2.31.0 ; python_version >= "3.9" and python_version < "4.0"
|
|
safetensors==0.3.1 ; python_version >= "3.9" and python_version < "4.0"
|
|
sentencepiece==0.1.99 ; python_version >= "3.9" and python_version < "4.0"
|
|
setuptools==67.8.0 ; python_version >= "3.9" and python_version < "4.0"
|
|
tokenizers==0.13.3 ; python_version >= "3.9" and python_version < "4.0"
|
|
transformers==4.29.2 ; python_version >= "3.9" and python_version < "4.0"
|
|
tqdm==4.65.0 ; python_version >= "3.9" and python_version < "4.0"
|
|
typer==0.6.1 ; python_version >= "3.9" and python_version < "4.0"
|
|
typing-extensions==4.6.0 ; python_version >= "3.9" and python_version < "4.0"
|
|
urllib3==2.0.2 ; python_version >= "3.9" and python_version < "4.0"
|
|
win32-setctime==1.1.0 ; python_version >= "3.9" and python_version < "4.0" and sys_platform == "win32"
|
|
wrapt==1.15.0 ; python_version >= "3.9" and python_version < "4.0"
|