asuradayuci/cnn-rnn2016

Reimplementation the paper of Recurrent Convolutional Network for Video-based Person Re-Identification in Pytorch

cnn-rnn2016's Issues

TypeError: object of type 'Image' has no len()

你好 訓練跑完準備評估之後出現如此問題 請問該如何解決(除路徑以外未改其他code)

File D:\CNN-RNN2016-master\prid_data.py:149 in
main(args)

File D:\CNN-RNN2016-master\prid_data.py:106 in main
top1 = evaluator.evaluate(query_loader, gallery_loader, dataset.queryinfo, dataset.galleryinfo)

File D:\CNN-RNN2016-master\eval\evaluator.py:85 in evaluate
pooled_probe = self.extract_feature(query_loader) # 1980 * 128

File D:\CNN-RNN2016-master\eval\evaluator.py:51 in extract_feature
for i, inputs in enumerate(data_loader):

File "D:\CNN-RNN2016-master\dataprocess\video_loader.py", line 140, in getitem
img = self.transform(img)
File "D:\CNN-RNN2016-master\dataprocess\seqtransform.py", line 32, in call
seqs = t(seqs)
File "D:\CNN-RNN2016-master\dataprocess\seqtransform.py", line 43, in call
modallen = len(seqs) # 1个list

TypeError: object of type 'Image' has no len()

您好，一直以来从您的github和博客上学到了很多，关于您这篇复现工作我有些疑惑的地方。造成复现的实验结果和原文结果有差距的原因是什么呢？是复现的代码没有完全复现到位，还是有什么别的原因？

splits_prid2011.json is not find

I downloaded the dataset from https://www.tugraz.at/institute/icg/research/team-bischof/lrs/downloads/PRID11/ , but there is no splits_prid2011.json file, How can I get this file?