tensorflow:Model 是用形状 (None, None, 6) 构造的，但它是在形状不兼容的输入上调用的

Question

The complete error is

WARNING:tensorflow:Model was constructed with shape (None, None, 6) for input KerasTensor(type_spec=TensorSpec(shape=(None, None, 6), dtype=tf.float32, name='dense_input'), name='dense_input', description="created by layer 'dense_input'"), but it was called on an input with incompatible shape (None, 6).
2021-02-10 17:56:08.288146: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11
2021-02-10 17:56:08.544550: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublasLt.so.11
WARNING:tensorflow:Model was constructed with shape (None, None, 6) for input KerasTensor(type_spec=TensorSpec(shape=(None, None, 6), dtype=tf.float32, name='dense_4_input'), name='dense_4_input', description="created by layer 'dense_4_input'"), but it was called on an input with incompatible shape (None, 6).
WARNING:tensorflow:Model was constructed with shape (None, None, 6) for input KerasTensor(type_spec=TensorSpec(shape=(None, None, 6), dtype=tf.float32, name='dense_input'), name='dense_input', description="created by layer 'dense_input'"), but it was called on an input with incompatible shape (20, 6).
WARNING:tensorflow:Model was constructed with shape (None, None, 6) for input KerasTensor(type_spec=TensorSpec(shape=(None, None, 6), dtype=tf.float32, name='dense_input'), name='dense_input', description="created by layer 'dense_input'"), but it was called on an input with incompatible shape (20, 6).

I am sure this is an input shape error, this is the relevant code:

def create_model(self):
    model = tf.keras.models.Sequential()  # a basic feed-forward model
    model.add(tf.keras.layers.Dense(6, activation=tf.nn.relu, input_shape=env.OBSERVATION_SPACE_VALUES))  # a simple fully-connected layer, 6 units, relu activation
    model.add(tf.keras.layers.Dense(128, activation=tf.nn.relu))  # a simple fully-connected layer, 128 units, relu activation
    model.add(tf.keras.layers.Dense(64, activation=tf.nn.relu))  # a simple fully-connected layer, 64 units, relu activation
    model.add(tf.keras.layers.Dense(27, activation=tf.nn.softmax))  # our output layer. 27 units for 27 actions 
    model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])


    return model

The env.OBSERVATION_SPACE_VALUES is (N0NE, 6) the input is basically a 1 dimensional array with 6 values

Answer 1

You should not specify the batch dimension in the input_shape argument of the Layer constructor.

env.OBSERVATION_SPACE_VALUES = (6,)

The documentation of the Input layer states ( emphasis is mine):

input_shape: Shape tuple (not including the batch axis) , or TensorShape instance (not including the batch axis) .