Benchmark for Tuning Accuracy and Efficiency
Overview
The benchmark includes our efforts in using Colossal-AI to train different tasks to achieve SOTA results. We are interested in both validataion accuracy and training speed, and prefer larger batch size to take advantage of more GPU devices. For example, we trained vision transformer with batch size 512 on CIFAR10 and 4096 on ImageNet1k, which are basically not used in existing works. Some of the results in the benchmark trained with 8x A100 are shown below.
| Task | Model | Training Time | Top-1 Accuracy |
|---|---|---|---|
| CIFAR10 | ViT-Lite-7/4 | ~ 16 min | ~ 90.5% |
| ImageNet1k | ViT-S/16 | ~ 16.5 h | ~ 74.5% |
The train.py script in each task runs training with the specific configuration script in configs/ for different parallelisms. Supported parallelisms include data parallel only (ends with vanilla), 1D (ends with 1d), 2D (ends with 2d), 2.5D (ends with 2p5d), 3D (ends with 3d).
Each configuration scripts basically includes the following elements, taking ImageNet1k task as example:
TOTAL_BATCH_SIZE = 4096
LEARNING_RATE = 3e-3
WEIGHT_DECAY = 0.3
NUM_EPOCHS = 300
WARMUP_EPOCHS = 32
# data parallel only
TENSOR_PARALLEL_SIZE = 1
TENSOR_PARALLEL_MODE = None
# parallelism setting
parallel = dict(
pipeline=1,
tensor=dict(mode=TENSOR_PARALLEL_MODE, size=TENSOR_PARALLEL_SIZE),
)
fp16 = dict(mode=AMP_TYPE.TORCH, ) # amp setting
gradient_accumulation = 2 # accumulate 2 steps for gradient update
BATCH_SIZE = TOTAL_BATCH_SIZE // gradient_accumulation # actual batch size for dataloader
clip_grad_norm = 1.0 # clip gradient with norm 1.0
Upper case elements are basically what train.py needs, and lower case elements are what Colossal-AI needs to initialize the training.
Usage
To start training, use the following command to run each worker:
$ DATA=/path/to/dataset python train.py --world_size=WORLD_SIZE \
--rank=RANK \
--local_rank=LOCAL_RANK \
--host=MASTER_IP_ADDRESS \
--port=MASTER_PORT \
--config=CONFIG_FILE
It is also recommended to start training with torchrun as:
$ DATA=/path/to/dataset torchrun --nproc_per_node=NUM_GPUS_PER_NODE \
--nnodes=NUM_NODES \
--node_rank=NODE_RANK \
--master_addr=MASTER_IP_ADDRESS \
--master_port=MASTER_PORT \
train.py --config=CONFIG_FILE
3 Mar 31, 2022
729 Jan 02, 2023
110 Dec 27, 2022
110 Jan 02, 2023
2.6k Jan 08, 2023
2 Oct 12, 2022
16 Nov 18, 2022
153 Nov 30, 2022
9 Nov 06, 2022
51 Nov 22, 2022
23 Dec 21, 2022
49 Dec 28, 2022
21 Dec 12, 2022
127 Dec 20, 2022
20 Dec 06, 2022
718 Jan 01, 2023
10 Dec 06, 2022
2 May 11, 2022
29 Oct 19, 2022
139 Jan 01, 2023