The vocabulary file to convert the tokens to human readable language is missing

[GaganaMD]

In the drive that is shared, the vocabulary file to convert the tokens to human readable language is missing. So my model output is a set of tokens that I am not able to read.

[jchenghu]

Hi, sorry for the late reply. I hope you found a solution, and if you didn't, here's a quick trick.

Execute train.py, with the following lines after the data loading

print(coco_dataset.caption_word2idx_dict)
print(coco_dataset.caption_idx2word_list)