lucidrains / toolformer-pytorch

Implementation of Toolformer, Language Models That Can Use Tools, by MetaAI
MIT License
1.94k stars 124 forks source link

run example in readme get errors #6

Open hxf829 opened 1 year ago

hxf829 commented 1 year ago

run example in readme, get errors as follows, how to fix? thank you image

shiqichen17 commented 1 year ago

Same here.

lzp492456131 commented 1 year ago

AssertionError: Torch not compiled with CUDA enabled

DengXianqi commented 1 year ago

Hello, did you find out how to fix this error?

NNNNNaaaaaa commented 1 year ago

Same here

lianggaoquan commented 1 year ago

same error

wentinghome commented 1 year ago

same error

wentinghome commented 1 year ago

The following code does not augment original data with callable API starting '[' and ending with ']'. The error happens since after filtering, there is no callable API. data_with_api_calls = self.generate_data_with_api_calls(data)

A potential solution is to increase the size of PALM and load the pre-trained model checkpoint so that the prompted data contains API. I haven't noticed the checkpoint loading in the current repo. If anyone finds, please let me know.

yonatanbitton commented 1 year ago

I receive the same error. Was someone able to fix it? @lucidrains

line 882, in forward
    assert len(filtered_data_with_api_calls) > 0, 'your model failed to follow instructions and make API calls. please try a better model or do some better prompt engineering'
AssertionError: your model failed to follow instructions and make API calls. please try a better model or do some better prompt engineering
shiqichen17 commented 1 year ago

I've received your e-mail , and I'll reply soon:)Shiqi Chen

yixin-zhu commented 1 year ago

I met the same problem. Does anyone find out how to fix it? image

vectorx666 commented 1 year ago

Has anyone found a solution to this? Thanks

shiqichen17 commented 1 year ago

I've received your e-mail , and I'll reply soon:)Shiqi Chen

Kawai1Ace commented 1 year ago

does anyone solve this problem?

sungwo101 commented 7 months ago

does anyone found a solution for this???

shiqichen17 commented 7 months ago

I've received your e-mail , and I'll reply soon:)Shiqi Chen

Empth commented 6 months ago

I'm getting the same error, does anyone know of any solutions?

Alston16 commented 5 months ago

AssertionError: Torch not compiled with CUDA enabled

If you have a cuda compatible GPU, reinstall pytorch with the command from this link Or just remove .cuda() to have the model loaded in CPU

shiqichen17 commented 5 months ago

I've received your e-mail , and I'll reply soon:)Shiqi Chen

LiaoNiuyu commented 4 months ago

I've received your e-mail , and I'll reply soon:)Shiqi Chen

Have you solved the problem? thanks

shiqichen17 commented 4 months ago

I've received your e-mail , and I'll reply soon:)Shiqi Chen

ritalily commented 2 months ago

I've received your e-mail , and I'll reply soon:)Shiqi Chen

Hi Shiqi, I am facing the same error as this post when running with the example in readme. May you guide me on how to resolve this too? Much appreciated!

f840b93c01c46d6e64db9cd243a4b0c4
shiqichen17 commented 2 months ago

I've received your e-mail , and I'll reply soon:)Shiqi Chen

ritalily commented 2 months ago

I've received your e-mail , and I'll reply soon:)Shiqi Chen

Have you solved the problem? thanks

Hi LiaoNiuya, excuse me, just want to check my luck with you. Were you managed to resolve the error at the end?