Closed darenr closed 3 years ago
Do you have a GPU? We don't use a NN without the GPU due to low training speed on the CPU.
Your example works on kaggle notebook, but it has another python version. Link to kaggle kernel https://www.kaggle.com/simakov/notebookf7a1bdef0f
So if you have the GPU, the reason is python. We haven't tested LAMA with python 3.9+.
One more thing. If you want to specify bert model for the NN, you should do it directly:
nn_params={'opt_params': {'lr': 1e-5},
'max_length': 128, 'bs': 32, 'n_epochs': 1,
'lang': 'en',
'bert_model': 'distilbert-base-uncased'
},
@darenr Hi Daren, does the message above fix your issue?
yes, thank you
I might have coded this incorrectly (very likley) but I'm getting the error:
Reproducible test case:
nlp.zip
lightautoml==0.2.14 huggingface-hub==0.0.8 transformers==4.6.1 python=3.9.5