Closed wangyihuan123 closed 5 years ago
Hi, I notice that the code below, and am wondering whether the scale 720 is on purpose (not 512*1.25)
train/main.py: https://github.com/sacmehta/ESPNet/blob/afe71c38edaee3514ca44e0adcafdf36109bf437/train/main.py#L227
trainDataset_scale2 = myTransforms.Compose([ myTransforms.Normalize(mean=data['mean'], std=data['std']), myTransforms.Scale(1280, 720), # <<<<<<<<<< 512*1.25 == 640 ? myTransforms.RandomCropResize(100), myTransforms.RandomFlip(),
myTransforms.ToTensor(args.scaleIn), #
])
There is no specific reason behind this choice. We just selected it to have 720p resolution.
Hi, I notice that the code below, and am wondering whether the scale 720 is on purpose (not 512*1.25)
train/main.py: https://github.com/sacmehta/ESPNet/blob/afe71c38edaee3514ca44e0adcafdf36109bf437/train/main.py#L227
trainDataset_scale2 = myTransforms.Compose([ myTransforms.Normalize(mean=data['mean'], std=data['std']), myTransforms.Scale(1280, 720), # <<<<<<<<<< 512*1.25 == 640 ? myTransforms.RandomCropResize(100), myTransforms.RandomFlip(),
myTransforms.RandomCrop(64),
])