How can I clear a model created with Keras and Tensorflow(as backend)?
Question:
I have a problem when training a neural net with Keras in Jupyter Notebook. I created a sequential model with several hidden layers. After training the model and saving the results, I want to delete this model and create a new model in the same session, as I have a for loop that checks the results for different parameters. But as I understand the errors I get, when changing the parameters, when I loop over, I am just adding layers to the model (even though I initialise it again with network = Sequential() inside the loop). So my question is, how can I completely clear the previous model or how can I initialise a completely new model in the same session?
Answers:
keras.backend.clear_session() should clear the previous model. From https://keras.io/backend/:
Destroys the current TF graph and creates a new one.
Useful to avoid clutter from old models / layers.
I know it is a bit old thread, but I was looking for something to clear the session. For TensorFlow 2.8 I think you need to use tf.keras.backend.clear_session()
I have a problem when training a neural net with Keras in Jupyter Notebook. I created a sequential model with several hidden layers. After training the model and saving the results, I want to delete this model and create a new model in the same session, as I have a for loop that checks the results for different parameters. But as I understand the errors I get, when changing the parameters, when I loop over, I am just adding layers to the model (even though I initialise it again with network = Sequential() inside the loop). So my question is, how can I completely clear the previous model or how can I initialise a completely new model in the same session?
keras.backend.clear_session() should clear the previous model. From https://keras.io/backend/:
Destroys the current TF graph and creates a new one.
Useful to avoid clutter from old models / layers.
I know it is a bit old thread, but I was looking for something to clear the session. For TensorFlow 2.8 I think you need to use tf.keras.backend.clear_session()