Error while using coco-caption to evaluate

[MozhdehDokhani]

Hi, I tried to use coco caption to evaluate my neuraltalk2 results. but this error occured:

Loading and preparing results... Traceback (most recent call last): File "myeval.py", line 29, in <module> cocoRes = coco.loadRes(resFile) File "/home/mozhdeh/Documents/neuraltalk2-master/coco-caption/pycocotools/coco.py", line 318, in loadRes if 'caption' in anns[0]: IndexError: list index out of range /home/mozhdeh/torch/install/bin/luajit: ./misc/utils.lua:17: attempt to index local 'file' (a nil value) stack traceback: ./misc/utils.lua:17: in function 'read_json' ./misc/net_utils.lua:202: in function 'language_eval' eval.lua:167: in function 'eval_split' eval.lua:173: in main chunk [C]: in function 'dofile' ...hdeh/torch/install/lib/luarocks/rocks/trepl/scm-1/bin/th:150: in main chunk [C]: at 0x55cbcb3e8570

Any solution?

[mymuli]

I had the same problem and tried to solve it with Python 3, but it was still wrong....

[mymuli]

我知道怎么解决了.... 在eval阶段，设置命令-language_eval 0....如果-language_eval 1, 会报错： if 'caption' in anns[0]: IndexError: list index out of range

[BetterZhouXu]

我知道怎么解决了.... 在eval阶段，设置命令-language_eval 0....如果-language_eval 1, 会报错： if 'caption' in anns[0]: IndexError: list index out of range

你好啊，你是在本地测评的还是在线测评的？没太看懂这个错误是怎么生成的，没有language_eval参数啊，并且我看我生成的结果是按照{'image_id': 123, 'caption': xxx xxx xxx}排成了列表， 结果和你一样的错误

[Sun-WeiZhen]

我知道怎么解决了.... 在eval阶段，设置命令-language_eval 0....如果-language_eval 1, 会报错： if 'caption' in anns[0]: IndexError: list index out of range

你好！在eval阶段，resFile一直是None，请问怎样解决这个问题？谢谢！ Loading and preparing results...
Traceback (most recent call last): File "G:/graph attention/SCA-image_captioning/evaluate.py", line 32, in cocoRes = coco.loadRes(resFile) File "G:\graph attention\SCA-image_captioning\pycocotools\coco.py", line 280, in loadRes anns = json.load(open(resFile)) TypeError: expected str, bytes or os.PathLike object, not NoneType