Skip to content

Adding args for names of train and val directories #1544

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Nov 26, 2019
Merged

Adding args for names of train and val directories #1544

Changes from 2 commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
74482f0
Generalised for custom dataset
rsomani95 Oct 31, 2019
85e3f8d
Typo, redundant code, sensible default
rsomani95 Nov 4, 2019
fdcbb52
Args for name of train and val dir
rsomani95 Nov 26, 2019
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 6 additions & 2 deletions references/video_classification/train.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -116,8 +116,8 @@ def main(args):

# Data loading code
print("Loading data")
traindir = os.path.join(args.data_path, 'train_avi-480p')
valdir = os.path.join(args.data_path, 'val_avi-480p')
traindir = os.path.join(args.data_path, args.train_dir)
valdir = os.path.join(args.data_path, args.val_dir)
normalize = T.Normalize(mean=[0.43216, 0.394666, 0.37645],
std=[0.22803, 0.22145, 0.216989])

Expand Down Expand Up @@ -203,6 +203,7 @@ def main(args):

print("Creating model")
model = torchvision.models.video.__dict__[args.model](pretrained=args.pretrained)
model.fc.out_features = args.output_classes
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is not quite what I meant.

Changing model.fc.out_features doesn't change the number of output planes, because the model weights have already been generated.
What I meant was to do something like

model = torchvision.models.video.__dict__[args.model](pretrained=args.pretrained, num_classes=args.num_classes)

But then, using pretrained=True and num-classes different than 400 will give an error, because the state-dicts are not compatible.

I'd rather not expose this option to the training scripts, as the user can modify the model as they want to perform their type of fine-tuning.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If fine-tuning on a custom dataset, one loads in a pretrained model with that model's num_class (400 here), and then changes the output classes to their custom dataset.

Though the output planes are different, the biggest benefit from a pretrained model aren't in the fc part of the model, but the convolutional filters that come in the previous parts of the model. The fc parts get retrained when fine-tuning this way.

I'd rather not expose this option to the training scripts, as the user can modify the model as they want to perform their type of fine-tuning.

This change in model.fc.out_features is mandatory when fine-tuning on a custom dataset, so I thought it would make sense to have it in the args. But of course, that's your call :)

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

What I mean is that this change doesn't actually change anything in the model. The weights are still the same as before, and they are not resized to a different size.

Putting it differently, you'll only predict at most 400 classes in this way.

model.to(device)
if args.distributed and args.sync_bn:
model = torch.nn.SyncBatchNorm.convert_sync_batchnorm(model)
Expand Down Expand Up @@ -274,7 +275,10 @@ def parse_args():
parser = argparse.ArgumentParser(description='PyTorch Classification Training')

parser.add_argument('--data-path', default='/datasets01_101/kinetics/070618/', help='dataset')
parser.add_argument('--train-dir', default='train_avi-480p', help='name of train dir')
parser.add_argument('--val-dir', default='val_avi-480p', help='name of val dir')
parser.add_argument('--model', default='r2plus1d_18', help='model')
parser.add_argument('--output-classes', default=400, help='no. of output classes (if finetuning)')
parser.add_argument('--device', default='cuda', help='device')
parser.add_argument('--clip-len', default=16, type=int, metavar='N',
help='number of frames per clip')
Expand Down