Kashgari is a production-level NLP Transfer learning framework built on top of tf.keras for text-labeling and text-classification, includes Word2Vec, BERT, and GPT2 Language Embedding.
model.pre_processor
to model.processor
predict_top_k_class
for classification model to get predict probabilities (#146)label2idx
, token2idx
properties to Embeddings and Modelstokenizer
property for BERT Embedding. (#136)predict_kwargs
for models predict()
function🎉🎉 tf.keras version 🎉🎉
tf.keras
. Discussion: #77
sequence_labeling_tokenize_add_bos_eos
consigBOS
and EOS
in sequence labeling result, fix #31AVCNNModel
, KMaxCNNModel
, RCNNModel
, AVRNNModel
, DropoutBGRUModel
, DropoutAVRNNModel
model to classification task.