Kamal Raj Kanakarajan comments

Results 57 comments of


                                            Kamal Raj Kanakarajan

Unused weights from saved model

yes

Suggestion on Readme.MD

@birdmw I will try adding that. Inference on the trained model examples.

save model issue in TF 2.0 (tf.saved_model.save)

@shawei3000 Post a simple code snippet , where i could reproduce the above error Try saving using https://www.tensorflow.org/api_docs/python/tf/keras/models/save_model

How to convert all cpp,header files to DLL file?

Ask in the PyTorch forum

is this using CoNll dataset for training NER tag ? dose the code support ontonotes as well?

https://github.com/kamalkraj/BERT-NER/issues/52 @ranjeetds has trained BERT ner using ontonotes 5.0 dataset.

Seems that F1 score of self trained model based on bert-base-uncased is unresonable

@yangheng95 use the branch `dev`

c++ inference

Hi @simonefrancia, I ported the model from Pytorch to libtorch.

c++ inference

@simonefrancia https://pytorch.org/tutorials/advanced/cpp_export.html

c++ inference

@simonefrancia tokenizer i re-wrote in C++.

c++ inference

@nggih I will try to add the model converter by this weekend