Question about feature dimension

[DungVo1507]

Great to find your repo, @antoine77340!

I have 3 questions for you to answer:

1. The shape of my .npy file is (66, 2048), 66 is calculated as the total number of frames in the video divided by the batch size right?
2. The output shape I want is (10, 66, 2048), 10 is 10 crops argumentation, can you tell me how to do it?
3. If I want to change the resnet101 model to the resnet50 model, what components do I need to change in your repo?

Thank you so much!

[h1manshus1ngh]

Great to find your repo, @antoine77340!

I have 3 questions for you to answer:

1. The shape of my .npy file is (66, 2048), 66 is calculated as the total number of frames in the video divided by the batch size right?

2. The output shape I want is (10, 66, 2048), 10 is 10 crops argumentation, can you tell me how to do it?

3. If I want to change the resnet101 model to the resnet50 model, what components do I need to change in your repo?

Thank you so much!

can you provide me csv file you have genrated

[antoine77340]

Very sorry for the late reply:

1. 66 is the number of feature per video. Since they are extracted at one feature per second, it should mean your video was 66 seconds right?
2. The is no implementation of crop augmentation here, so just one single center crop.
3. Do you mean changing the ResNext 101? you can do it here: https://github.com/antoine77340/video_feature_extractor/blob/master/model.py#L25