When downloading model using download.py, there are some trouble. To avoid this, I had to download directly using the URL at the beginning of the script and to decompress it myself in the pretrained-models folder.
Traceback (most recent call last):
File "download.py", line 105, in <module>
main()
File "download.py", line 102, in main
download_model(name=args.model)
File "download.py", line 79, in download_model
tar = tarfile.open(file_destination, "r:xz")
File "/users/troux/.conda/envs/character-bert/lib/python3.8/tarfile.py", line 1621, in open
return func(name, filemode, fileobj, **kwargs)
File "/users/troux/.conda/envs/character-bert/lib/python3.8/tarfile.py", line 1734, in xzopen
raise ReadError("not an lzma file")
tarfile.ReadError: not an lzma file
When downloading model using
download.py
, there are some trouble. To avoid this, I had to download directly using the URL at the beginning of the script and to decompress it myself in the pretrained-models folder.Regards,