Finetuning H5-File
JanFreise opened this issue · 0 comments
JanFreise commented
I pretrained a new LM using the below stated tutorial.
https://keras.io/examples/nlp/pretraining_BERT/
This produces a TensorFlow checkpoint tf_model.h5 file for subsequent Finetuning.
Can i somehow finetune this with T-NER or is an pytorch_model.bin file obligatory?
The ERROR-Message suggest a way to do so - so i guess there can be done implementing a switch in T-NER:
YOURMODEL does not appear to have a file named pytorch_model.bin but there is a file for TensorFlow weights.
Use from_tf=True
to load this model from those weights.
Files available:
README.md
config.json
special_tokens_map.json
tf_model.h5
tokenizer.json
tokenizer_config.json
vocab.txt