kenshohara / 3D-ResNets-PyTorch

3D ResNets for Action Recognition (CVPR 2018)
MIT License
3.91k stars 933 forks source link

Question About Batch_size and Epoch #260

Open EmirhanToprak opened 3 years ago

EmirhanToprak commented 3 years ago

Hi, i implemented the code to my dataset. I have 360 video in total for 2 class (270 training 90 for validation). I select the "sample duration" = 32

I didn't quite understand the batch size and epoch. When i set batch_size = 12, training dataset for per epoch = 270/12 = 23 video

Can anyone explain please? Thank you.

The opts.py i selected parser.add_argument('--sample_size', default=224, type=int, help='Height and width of inputs') parser.add_argument('--sample_duration', default=32, type=int, help='Temporal duration of inputs') parser.add_argument('--batch_size', default=12, type=int, help='Batch Size') parser.add_argument('--n_val_samples', default=12, type=int, help='Number of validation samples for each activity')