Heliang-Zheng

@zdm123 @weisq2691 I got the same problem. Finally, I found it was because my data and label were not loaded properly. My data file is modified from pascal_voc.py, "cls =...

@Kurumi233 @wangyirui 128 is the batch_size, and I have replaced this number to self.batch_size. Such an item is for normalizing. 512 is the channel number of the last stage. Since...

Thanks for your interests. I will release a gluon version and a pytorch version before the end of Aug. @zimenglan-sysu-512 @RogerQi

@yogsin Sorry for my late reply, and thanks for your valuable comments. 1. I think you are right, and I will fix this bug. 2. I thought it is a...

Thanks for your interests. I will release a gluon version and a pytorch version before the end of Aug.

@whyou5945 1. The data for part-cnn are obtained by running the attention model on the training dataset and crop. 2. Fine tune every part individually, then extract the parts' 512-dim...

@JohnnieXDU For example, given an 300*500 image, first resize it to 448* 746 (keep ratio), then random crop (in training) 448*448 patches to be the input of the classification CNN,...

@FanQC protobuf-v120.2.6.1

@JaHangirYin You can try https://pan.baidu.com/s/1kWHOJxt

@hellobxx Here is the link for ''caffe'' folder: https://drive.google.com/open?id=1EIASCb2Bcc9W5wGDQTbvb7XHGq5ANmjW