gmryu
gmryu
@robotsp Sorry, I have no confidence in these since I have not tried it yet. I guess I would need to try myself later. However, I only have time in...
@robotsp Congratulations! It is so nice of you to tell me this works too. Hope it can solve your issue afterwards.
@suraj143rosy Are you comparing NLLB (multilingual model) to a only japanese-english model? While this is not an absolute reason, it definetly impacts: https://github.com/facebookresearch/fairseq/issues/4560 So I believe a lot of characters...
@dina-adel @AhmedEssam19 Have you tried this advice? https://github.com/facebookresearch/fairseq/issues/4597#issuecomment-1198377793 It feels like a wrong logic is implemented with w2v_path, resulting a forever loop. So you need delete it and declare the...
https://github.com/facebookresearch/fairseq/tree/main/examples/translation or any other projects you can found in `{fairseq_repository}/examples`.
If you want to do it yourself, I suggest you to copy and edit fairseq's generate.py https://github.com/facebookresearch/fairseq/blob/main/fairseq_cli/generate.py#L187-L211 I assume you have used `fairseq-generate`, then you can just use this `generate.py`...
It is probably the input tensors in `sample` are on cpu by default. You may try: ``` print(sample) # to see what is inside, # probably there is a "net_input"...
@martianmartina Hi. In short, you have to fix the tokenized data to contain your custom tokens correctly and add your custom tokens into dict.txt (the `--srcdict` if they appear in...
@martianmartina `fairseq-preprocess` is actually [{fairseq repository}/fairseq_cli/preprocess.py](https://github.com/facebookresearch/fairseq/blob/main/fairseq_cli/preprocess.py), if you search `encode_fn`, you find nothing. Yet if you search `encode_fn` `decode_fn` in the whole `fairseq_cli`, you find: - `fairseq-generate` has only `decode_fn`...
@martianmartina Yes, `fairseq-preprocess` creates a dict for you. There are `--thresholdtgt` `--thresholdsrc` `--nwordstgt` `--nwordssrc` `--padding-factor` that affect the created dict. See https://fairseq.readthedocs.io/en/latest/command_line_tools.html#Preprocessing A small disadvantage is that dict is not...