CTranslate2
CTranslate2 copied to clipboard
Support for UMT5
New UMT5 models from Google are currently the most interesting variation of the original T5s.
However, trying to convert a UMT5 model using the transformers converter by running:
ct2-transformers-converter --model google/umt5-xl --output_dir ct2-umt5-3b --quantization int8
yields:
Downloading (…)lve/main/config.json: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 812/812 [00:00<00:00, 4.96MB/s]
Traceback (most recent call last):
File "/home/user/miniconda3/bin/ct2-transformers-converter", line 8, in <module>
sys.exit(main())
^^^^^^
File "/home/user/miniconda3/lib/python3.11/site-packages/ctranslate2/converters/transformers.py", line 1719, in main
converter.convert_from_args(args)
File "/home/user/miniconda3/lib/python3.11/site-packages/ctranslate2/converters/converter.py", line 50, in convert_from_args
return self.convert(
^^^^^^^^^^^^^
File "/home/user/miniconda3/lib/python3.11/site-packages/ctranslate2/converters/converter.py", line 89, in convert
model_spec = self._load()
^^^^^^^^^^^^
File "/home/user/miniconda3/lib/python3.11/site-packages/ctranslate2/converters/transformers.py", line 106, in _load
raise ValueError(
ValueError: No conversion is registered for the model configuration UMT5Config (supported configurations are: BartConfig, BertConfig, BloomConfig, CodeGenConfig, DistilBertConfig, FalconConfig, GPT2Config, GPTBigCodeConfig, GPTJConfig, GPTNeoXConfig, LlamaConfig, M2M100Config, MBartConfig, MPTConfig, MT5Config, MarianConfig, OPTConfig, PegasusConfig, RWConfig, T5Config, WhisperConfig, XLMRobertaConfig)
Is there an easy workaround ? Is this something that should be added in the package ?