Some of the problems

[Eroslon]

Hi, blogger, I am very glad to read your article, but I have some problems when running the code, I would like to ask you for help, the following picture is the problem I met, I look forward to your advice, thank you very much! ![Uploading image.png…]() ![Uploading image.png…]()

[rddy]

Can you please post the original text of the error message? The screenshot doesn't contain the entire backtrace, so it's hard to tell what the issue is.

[Eroslon]

Ok, thank you very much. The error it prompts is：

TypeError Traceback (most recent call last)

in 11 if tutor_name.startswith('RL'): 12 agent = build_tutor(n_items) ---> 13 R[k, i, :, j] = agent.train(rl_env, n_eps=n_eps) 14 else: 15 if 'Thresh' in tutor_name: in train(self, gym_env, n_eps) 9 policy = CategoricalGRUPolicy( 10 env_spec=env.spec, hidden_dim=32, ---> 11 state_include_action=False) 12 self.raw_policy = LoggedTRPO( 13 env=env, E:\Anaconda\lib\site-packages\rllab\policies\categorical_gru_policy.py in __init__(self, env_spec, hidden_dim, feature_network, state_include_action, hidden_nonlinearity) 72 hidden_nonlinearity=hidden_nonlinearity, 73 output_nonlinearity=TT.nnet.softmax, ---> 74 name="prob_network" 75 ) 76 E:\Anaconda\lib\site-packages\rllab\core\network.py in __init__(self, input_shape, output_dim, hidden_dim, hidden_nonlinearity, output_nonlinearity, name, input_var, input_layer) 226 nonlinearity=output_nonlinearity, 227 W=l_output_flat.W, --> 228 b=l_output_flat.b, 229 ) 230 E:\Anaconda\lib\site-packages\lasagne\layers\dense.py in __init__(self, incoming, num_units, W, b, nonlinearity, **kwargs) 62 b=init.Constant(0.), nonlinearity=nonlinearities.rectify, 63 **kwargs): ---> 64 super(DenseLayer, self).__init__(incoming, **kwargs) 65 self.nonlinearity = (nonlinearities.identity if nonlinearity is None 66 else nonlinearity) E:\Anaconda\lib\site-packages\lasagne\layers\base.py in __init__(self, incoming, name) 42 self.params = OrderedDict() 43 ---> 44 if any(d is not None and d <= 0 for d in self.input_shape): 45 raise ValueError(( 46 "Cannot create Layer with a non-positive input_shape "： E:\Anaconda\lib\site-packages\lasagne\layers\base.py in (.0) 42 self.params = OrderedDict() 43 ---> 44 if any(d is not None and d <= 0 for d in self.input_shape): 45 raise ValueError(( 46 "Cannot create Layer with a non-positive input_shape " TypeError: '<=' not supported between instances of 'tuple' and 'int' The error comes from lasagne\layers\base.py, the specific part of the code is as follows: from collections import OrderedDict from .. import utils __all__ = [ "Layer", "MergeLayer", ] # Layer base class class Layer(object): """ The :class:`Layer` class represents a single layer of a neural network. It should be subclassed when implementing new types of layers. Because each layer can keep track of the layer(s) feeding into it, a network's output :class:`Layer` instance can double as a handle to the full network. """ def __init__(self, incoming, name=None): """ Instantiates the layer. Parameters ---------- incoming : a :class:`Layer` instance or a tuple The layer feeding into this layer, or the expected input shape. name : a string or None An optional name to attach to this layer. """ if isinstance(incoming, tuple): self.input_shape = incoming self.input_layer = None else: self.input_shape = incoming.output_shape self.input_layer = incoming self.name = name self.params = OrderedDict() if any(d is not None and d <= 0 for d in self.input_shape): raise ValueError(( "Cannot create Layer with a non-positive input_shape " "dimension. input_shape=%r, self.name=%r") % ( self.input_shape, self.name))

[Eroslon]

I'm very sorry for disturbing you again because the previous problem has not been solved. Do you have any solution? Thank you very much!

[rddy]

This could be an rllab version issue. Try using rllab 0.1.0