search

sassoftware / python-dlpy

The SAS Deep Learning Python (DLPy) package provides the high-level Python APIs to deep learning methods in SAS Visual Data Mining and Machine Learning. It allows users to build deep learning models using friendly Keras-like APIs.
Apache License 2.0
224 stars 131 forks source link

TextClassification reset weights #245

Closed fede-citti closed 4 years ago

fede-citti commented 4 years ago

Is there a way to reset the weights in a TextClassification application?

I have the following situation: Define model1: model1 = TextClassification(s, neurons=10, n_blocks= 3, rnn_type='GRU') Train it: model1.fit(data=trainData, inputs='text', texts='text', target='target_new', text_parms=TextParms(init_input_embeddings='wiki_embeddings'), mini_batch_size=1, max_epochs=50, lr=0.1, log_level=2, valid_table=validData) Define a new model: model2 = TextClassification(s, neurons=10, n_blocks= 3, rnn_type='GRU') then model2 inherits the same weights as model1, that's not what I want.

sasxianhu commented 4 years ago

can you use another model_table name? The default is 'text_classifier'.

sasxianhu commented 4 years ago

can you use another model_table name? The default is 'text_classifier'.

fede-citti commented 4 years ago

Thanks, the issue is solved