Hugging Face's logo

Spaces:
mr2along
/
vietnamese_hate_speech_detection
Sleeping

App Files Community
vietnamese_hate_speech_detection / knowledge_distillation.py

Commit History

Increase threshold
99575b1

jesse-tong commited on

Change eval
6e2887d

jesse-tong commited on

Edit evaluation to use weighted instead as there are imbalance labels
bad72d7

jesse-tong commited on

Fix an error
6bfe0f4

jesse-tong commited on

print to console
d26ac21

jesse-tong commited on

Fix a bug in evaluation in multi-class
80b78df

jesse-tong commited on

Forget num_classes
6cf4c1f

jesse-tong commited on

Change loss function for multi-category because CrossEntropyLoss only accept outputs of (batch, x) and labels with shape of (batch)
cb428cb

jesse-tong commited on

First commit
f292cd1

jesse-tong commited on

Found a critical bug in LSTM inference when distillation using BERT Tokenizer but inference uses custom LSTM tokenizer
77bc910

jesse-tong commited on

Update LSTM gradient clipping
a770449

jesse-tong commited on

Save vocab size with model_state_dict
ca56e1c

jesse-tong commited on

Edit some saving and loading details
626f169

jesse-tong commited on

Add LSTM fine tuning
ae47555

jesse-tong commited on