GroupMixFormer-Pytorch

This is a warehouse for GroupMixFormer-pytorch-model, can be used to train your image datasets for classification tasks.

Project Structure

├── datasets: Load datasets
    ├── my_dataset.py: Customize reading data sets and define transforms data enhancement methods
    ├── split_data.py: Define the function to read the image dataset and divide the training-set and test-set
    ├── threeaugment.py: Additional data augmentation methods
├── models: GroupMixFormer Model
    ├── build_model.py: Construct "GroupMixFormer" model
├── util:
    ├── engine.py: Function code for a training/validation process
    ├── losses.py: Knowledge distillation loss, combined with teacher model (if any)
    ├── optimizer.py: Define Sophia optimizer
    ├── samplers.py: Define the parameter of "sampler" in DataLoader
    ├── utils.py: Record various indicator information and output and distributed environment
├── estimate_model.py: Visualized evaluation indicators ROC curve, confusion matrix, classification report, etc.
└── train_gpu.py: Training model startup file

Precautions

Before you use the code to train your own data set, please first enter the train_gpu.py file and modify the data_root, batch_size and nb_classes parameters. If you want to draw the confusion matrix and ROC curve, you only need to remove the comments of Plot_ROC and Predictor at the end of the code. For the third parameter, you should change it to the path of your own model weights file(.pth). Taking the model(groupmixformer_tiny) as an example, inputting a 3-channel image with a height and width of 224, the number of model parameters that need to be trained is as follows:

===================================================================================================================
Total params: 10,709,357
Trainable params: 10,709,357
Non-trainable params: 0
Total mult-adds (M): 466.37
===================================================================================================================
Input size (MB): 0.60
Forward/backward pass size (MB): 342.21
Params size (MB): 42.84
Estimated Total Size (MB): 385.65
===================================================================================================================

Use Sophia Optimizer (in util/optimizer.py)

You can use anther optimizer sophia, just need to change the optimizer in train_gpu.py, for this training sample, can achieve better results

# optimizer = create_optimizer(args, model_without_ddp)
optimizer = SophiaG(model.parameters(), lr=2e-4, betas=(0.965, 0.99), rho=0.01, weight_decay=args.weight_decay)

Train this model

Parameters Meaning:

1. nproc_per_node: <The number of GPUs you want to use on each node (machine/server)>
2. CUDA_VISIBLE_DEVICES: <Specify the index of the GPU corresponding to a single node (machine/server) (starting from 0)>
3. nnodes: <number of nodes (machine/server)>
4. node_rank: <node (machine/server) serial number>
5. master_addr: <master node (machine/server) IP address>
6. master_port: <master node (machine/server) port number>

Note:

If you want to use multiple GPU for training, whether it is a single machine with multiple GPUs or multiple machines with multiple GPUs, each GPU will divide the batch_size equally. For example, batch_size=4 in my train_gpu.py. If I want to use 2 GPUs for training, each GPU will divide the batch_size. That means batch_size=2 on each GPU. Do not let batch_size=1 on each GPU, otherwise BN layer maybe report an error. If you recive an error like "ONE-PEACE training and evaluation script: error: unrecognized arguments: --local-rank=1" when you use distributed multi-GPUs training, just replace the command "torch.distributed.launch" to "torch.distributed.run".

train model with single-machine single-GPU：

python train_gpu.py

train model with single-machine multi-GPU：

python -m torch.distributed.launch --nproc_per_node=8 train_gpu.py

train model with single-machine multi-GPU:

(using a specified part of the GPUs: for example, I want to use the second and fourth GPUs)

CUDA_VISIBLE_DEVICES=1,3 python -m torch.distributed.launch --nproc_per_node=2 train_gpu.py

train model with multi-machine multi-GPU:

(For the specific number of GPUs on each machine, modify the value of --nproc_per_node. If you want to specify a certain GPU, just add CUDA_VISIBLE_DEVICES= to specify the index number of the GPU before each command. The principle is the same as single-machine multi-GPU training)

On the first machine: python -m torch.distributed.launch --nproc_per_node=1 --nnodes=2 --node_rank=0 --master_addr=<Master node IP address> --master_port=<Master node port number> train_gpu.py

On the second machine: python -m torch.distributed.launch --nproc_per_node=1 --nnodes=2 --node_rank=1 --master_addr=<Master node IP address> --master_port=<Master node port number> train_gpu.py

Citation

@inproceedings{Ge2023AdvancingVT,
  title={Advancing Vision Transformers with Group-Mix Attention},
  author={Chongjian Ge and Xiaohan Ding and Zhan Tong and Li Yuan and Jiangliu Wang and Yibing Song and Ping Luo},
  year={2023},
  url={https://api.semanticscholar.org/CorpusID:265456206}
}

ChrisChen1023/GroupMixFormer-Pytorch