I have docker pull huggingface/optimum-nvidia and
then started docker container by docker run -it --rm --gpus all -v /path/to/my_model:/my_model huggingface/optimum-nvidia
I have ran llama.py from example folder once and generated build.json and copied that to my model folder as well.
I have created one more file python file inside docker with 'Generate' section of 'Quickstart Guide' as contents, while running that im getting,
Traceback (most recent call last):
File "/opt/optimum-nvidia/examples/text-generation/tst.py", line 26, in <module>
tokenizer.batch_decode(generated_ids, skip_special_tokens=True)
File "/usr/local/lib/python3.10/dist-packages/transformers/tokenization_utils_base.py", line 3710, in batch_decode
return [
File "/usr/local/lib/python3.10/dist-packages/transformers/tokenization_utils_base.py", line 3711, in <listcomp>
self.decode(
File "/usr/local/lib/python3.10/dist-packages/transformers/tokenization_utils_base.py", line 3750, in decode
return self._decode(
File "/usr/local/lib/python3.10/dist-packages/transformers/tokenization_utils_fast.py", line 626, in _decode
text = self._tokenizer.decode(token_ids, skip_special_tokens=skip_special_tokens)
TypeError: argument 'ids': 'list' object cannot be interpreted as an integer
upon checking the 'token_ids' in the above line is a list(list(list(Int))) e.g [[[1,2]]]
while self._tokenizer.decode(token_ids[0][0], skip_special_tokens=skip_special_tokens) is working. Can you please check this or am i doing anything wrong?
Also getting errors when passing multiple prompts.
Hi,
I have
docker pull huggingface/optimum-nvidia
and then started docker container bydocker run -it --rm --gpus all -v /path/to/my_model:/my_model huggingface/optimum-nvidia
I have ran llama.py from example folder once and generated build.json and copied that to my model folder as well.I have created one more file python file inside docker with 'Generate' section of 'Quickstart Guide' as contents, while running that im getting,
upon checking the 'token_ids' in the above line is a list(list(list(Int))) e.g [[[1,2]]] while self._tokenizer.decode(token_ids[0][0], skip_special_tokens=skip_special_tokens) is working. Can you please check this or am i doing anything wrong?
Also getting errors when passing multiple prompts.