Open zzw1123 opened 5 years ago
@zzw1123 You are right, we train about 150 epochs to achieve the reported results. Now, please use the updated code.
@liutinglt Thank you so much! I will update my code.
@liutinglt I noticed that the size of input image is changed to '384,384', right?
@zzw1123 The size with '384,384' already can achieve the reported performance. The edge is generated automatically, now you can change it to any size.
Hi, I have noticed the class LIPParsingEdgeDateset in datasets.py is implemented by following method:
max_iters=batch_size*total_iters which means if there are 1000 images in the whole dataset, it's going to be max_iters images now, so if I change the batch_size, the total number of images to be trained will be different. Am I right?