Basic Networks¶
Basic Networks in RLzoo¶
Basic neural networks
-
rlzoo.common.basic_nets.CNN(input_shape, conv_kwargs=None)[source]¶ Multiple convolutional layers for approximation Default setting is equal to architecture used in DQN
Parameters: - input_shape – (tuple[int]) (H, W, C)
- conv_kwargs – (list[param]) list of conv parameters for tl.layers.Conv2d
- Return:
- input tensor, output tensor
-
rlzoo.common.basic_nets.CNNModel(input_shape, conv_kwargs=None)[source]¶ Multiple convolutional layers for approximation Default setting is equal to architecture used in DQN
Parameters: - input_shape – (tuple[int]) (H, W, C)
- conv_kwargs – (list[param]) list of conv parameters for tl.layers.Conv2d
- Return:
- tl.model.Model
-
rlzoo.common.basic_nets.MLP(input_dim, hidden_dim_list, w_init=<tensorflow.python.ops.init_ops_v2.Orthogonal object>, activation=<function relu>, *args, **kwargs)[source]¶ Multiple fully-connected layers for approximation
Parameters: - input_dim – (int) size of input tensor
- hidden_dim_list – (list[int]) a list of dimensions of hidden layers
- w_init – (callable) initialization method for weights
- activation – (callable) activation function of hidden layers
- Return:
- input tensor, output tensor
-
rlzoo.common.basic_nets.MLPModel(input_dim, hidden_dim_list, w_init=<tensorflow.python.ops.init_ops_v2.Orthogonal object>, activation=<function relu>, *args, **kwargs)[source]¶ Multiple fully-connected layers for approximation
Parameters: - input_dim – (int) size of input tensor
- hidden_dim_list – (list[int]) a list of dimensions of hidden layers
- w_init – (callable) initialization method for weights
- activation – (callable) activation function of hidden layers
- Return:
- input tensor, output tensor