Open wenhuchen opened 8 years ago
I found the split of test/val is different from what is given in karpathy/neuraltalk2. According to their script https://github.com/karpathy/neuraltalk2/blob/master/coco/coco_preprocess.ipynb
They tried to get first 5000 as val, 5000-10000 as test from this dataset http://msvocds.blob.core.windows.net/annotations-1-0-3/captions_train-val2014.zip. But when I output the filename, I found it's totally different. Did I miss something?
I found the split of test/val is different from what is given in karpathy/neuraltalk2. According to their script https://github.com/karpathy/neuraltalk2/blob/master/coco/coco_preprocess.ipynb
They tried to get first 5000 as val, 5000-10000 as test from this dataset http://msvocds.blob.core.windows.net/annotations-1-0-3/captions_train-val2014.zip. But when I output the filename, I found it's totally different. Did I miss something?