This is a PyTorch implementation of the MMMAL - Net based on the paper Fine-Grained Visual Classification using Modified Multi-Branch and Multi-Scale Attention Network. Much of this code is reused from the original repository.
- python 3.7
- pytorch 1.3.1
- numpy 1.17.3
- scikit-image 0.16.2
- Tensorboard 1.15.0
- TensorboardX 2.0
- tqdm 4.41.1
- imageio 2.6.1
- pillow 6.1.0
- self-attention-cv 1.2.3
Download the FGVC-Aircraft datasets and copy the contents of the extracted data/images folder into datasets/FGVC_Aircraft/data/images)
You can also try other fine-grained datasets.
If you want to train the MMMAL-Net, please download the pretrained model of ResNet-50 and move it to models/pretrained before run python train.py. You may need to change the configurations in config.py if your GPU memory is not enough. The parameter N_list is N1, N2, N3 in the original paper and you can adjust them according to GPU memory. During training, the log file and checkpoint file will be saved in checkpoint directory.
If you want to test the MMMAL-Net, just run python test.py. You need to specify the model_path in test.py to choose the checkpoint model for testing.
We also provide the checkpoint model trained by ourselves for 60 epochs on FGVC-Aircraft daataset, you can see it at checkpoint/aircraft/epoch58.pth with 94.21% test accuracy.