Module: tf_agents.networks.value_rnn_network
Stay organized with collections
Save and categorize content based on your preferences.
Sample Keras Value Network with LSTM cells .
Implements a network that will generate the following layers:
[optional]: preprocessing_layers # preprocessing_layers
[optional]: (Add | Concat(axis=-1) | ...) # preprocessing_combiner
[optional]: Conv2D # conv_layer_params
Flatten
[optional]: Dense # input_fc_layer_params
[optional]: LSTM # lstm_cell_params
[optional]: Dense # output_fc_layer_params
Dense -> 1 # Value output
Classes
class ValueRnnNetwork: Recurrent value network. Reduces to 1 value output per batch item.
Other Members |
|---|
|
absolute_import
|
Instance of __future__._Feature
|
|
division
|
Instance of __future__._Feature
|
|
print_function
|
Instance of __future__._Feature
|
Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.
Last updated 2024-04-26 UTC.
[null,null,["Last updated 2024-04-26 UTC."],[],[],null,["# Module: tf_agents.networks.value_rnn_network\n\n\u003cbr /\u003e\n\n|--------------------------------------------------------------------------------------------------------------------|\n| [View source on GitHub](https://github.com/tensorflow/agents/blob/v0.19.0/tf_agents/networks/value_rnn_network.py) |\n\nSample Keras Value Network with LSTM cells .\n\nImplements a network that will generate the following layers:\n\n\\[optional\\]: preprocessing_layers # preprocessing_layers\n\\[optional\\]: (Add \\| Concat(axis=-1) \\| ...) # preprocessing_combiner\n\\[optional\\]: Conv2D # conv_layer_params\nFlatten\n\\[optional\\]: Dense # input_fc_layer_params\n\\[optional\\]: LSTM # lstm_cell_params\n\\[optional\\]: Dense # output_fc_layer_params\nDense -\\\u003e 1 # Value output\n\nClasses\n-------\n\n[`class ValueRnnNetwork`](../../tf_agents/networks/value_rnn_network/ValueRnnNetwork): Recurrent value network. Reduces to 1 value output per batch item.\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n| Other Members ------------- ||\n|-----------------|-----------------------------------|\n| absolute_import | Instance of `__future__._Feature` |\n| division | Instance of `__future__._Feature` |\n| print_function | Instance of `__future__._Feature` |\n\n\u003cbr /\u003e"]]
View source on GitHub