ColossalAI

mirror of https://github.com/hpcaitech/ColossalAI.git synced 2026-01-14 12:10:55 +00:00

Files

Wenhao Chen bb0a668fee [hotfix] set return_outputs=False in examples and polish code (#5404 )

* fix: simplify merge_batch

* fix: use return_outputs=False to eliminate extra memory consumption

* feat: add return_outputs warning

* style: remove `return_outputs=False` as it is the default value

2024-03-25 12:31:09 +08:00

benchmark_utils.py

[misc] update pre-commit and run all files (#4752 )

2023-09-19 14:20:26 +08:00

benchmark.py

[misc] update pre-commit and run all files (#4752 )

2023-09-19 14:20:26 +08:00

benchmark.sh

[booster] update bert example, using booster api (#3885 )

2023-06-07 15:51:00 +08:00

data.py

[pipeline]: support arbitrary batch size in forward_only mode (#5201 )

2024-01-02 23:41:12 +08:00

finetune.py

[hotfix] set return_outputs=False in examples and polish code (#5404 )

2024-03-25 12:31:09 +08:00

README.md

[shardformer] update shardformer readme (#4617 )

2023-09-05 13:14:41 +08:00

requirements.txt

[booster] update bert example, using booster api (#3885 )

2023-06-07 15:51:00 +08:00

test_ci.sh

[pipeline]: support arbitrary batch size in forward_only mode (#5201 )

2024-01-02 23:41:12 +08:00

README.md

Overview

This directory includes two parts: Using the Booster API finetune Huggingface Bert and AlBert models and benchmarking Bert and AlBert models with different Booster Plugin.

Finetune

bash test_ci.sh

Bert-Finetune Results

Plugin	Accuracy	F1-score	GPU number
torch_ddp	84.4%	88.6%	2
torch_ddp_fp16	84.7%	88.8%	2
gemini	84.0%	88.4%	2
hybrid_parallel	84.5%	88.6%	4

Benchmark

bash benchmark.sh

Now include these metrics in benchmark: CUDA mem occupy, throughput and the number of model parameters. If you have custom metrics, you can add them to benchmark_util.

Results

Bert

	max cuda mem	throughput(sample/s)	params
ddp	21.44 GB	3.0	82M
ddp_fp16	16.26 GB	11.3	82M
gemini	11.0 GB	12.9	82M
low_level_zero	11.29 G	14.7	82M

AlBert

	max cuda mem	throughput(sample/s)	params
ddp	OOM
ddp_fp16	OOM
gemini	69.39 G	1.3	208M
low_level_zero	56.89 G	1.4	208M