Init Signature at the end of output

12 views
Skip to first unread message

Vivek Katakam

unread,
Dec 21, 2019, 12:27:46 PM12/21/19
to spyder
Hi All,

I get getting the following message at the end of output after running my mnist dataset program in the Spyder IDE:


Init signature:
Dense(
    units,
    activation=None,
    use_bias=True,
    kernel_initializer='glorot_uniform',
    bias_initializer='zeros',
    kernel_regularizer=None,
    bias_regularizer=None,
    activity_regularizer=None,
    kernel_constraint=None,
    bias_constraint=None,
    **kwargs,
)
Docstring:    
Just your regular densely-connected NN layer.

`Dense` implements the operation:
`output = activation(dot(input, kernel) + bias)`
where `activation` is the element-wise activation function
passed as the `activation` argument, `kernel` is a weights matrix
created by the layer, and `bias` is a bias vector created by the layer
(only applicable if `use_bias` is `True`).

Note: if the input to the layer has a rank greater than 2, then
it is flattened prior to the initial dot product with `kernel`.

# Example

```python
    # as first layer in a sequential model:
    model = Sequential()
    model.add(Dense(32, input_shape=(16,)))
    # now the model will take as input arrays of shape (*, 16)
    # and output arrays of shape (*, 32)

    # after the first layer, you don't need to specify
    # the size of the input anymore:
    model.add(Dense(32))
```

# Arguments
    units: Positive integer, dimensionality of the output space.
    activation: Activation function to use
        (see [activations](../activations.md)).
        If you don't specify anything, no activation is applied
        (ie. "linear" activation: `a(x) = x`).
    use_bias: Boolean, whether the layer uses a bias vector.
    kernel_initializer: Initializer for the `kernel` weights matrix
        (see [initializers](../initializers.md)).
    bias_initializer: Initializer for the bias vector
        (see [initializers](../initializers.md)).
    kernel_regularizer: Regularizer function applied to
        the `kernel` weights matrix
        (see [regularizer](../regularizers.md)).
    bias_regularizer: Regularizer function applied to the bias vector
        (see [regularizer](../regularizers.md)).
    activity_regularizer: Regularizer function applied to
        the output of the layer (its "activation").
        (see [regularizer](../regularizers.md)).
    kernel_constraint: Constraint function applied to
        the `kernel` weights matrix
        (see [constraints](../constraints.md)).
    bias_constraint: Constraint function applied to the bias vector
        (see [constraints](../constraints.md)).

# Input shape
    nD tensor with shape: `(batch_size, ..., input_dim)`.
    The most common situation would be
    a 2D input with shape `(batch_size, input_dim)`.

# Output shape
    nD tensor with shape: `(batch_size, ..., units)`.
    For instance, for a 2D input with shape `(batch_size, input_dim)`,
    the output would have shape `(batch_size, units)`.
File:           c:\programdata\anaconda3\lib\site-packages\keras\layers\core.py
Type:           type
Subclasses:    

I did not choose any help topic. even though the above help message is comming at the end of output.
Is this an error in the code or something to do with settings.

Thanks and Regards,
Vivek

Bhanupratap Sahoo

unread,
Dec 21, 2019, 12:48:05 PM12/21/19
to spyd...@googlegroups.com
Sorry I cant understand.I'm a student and new to use spyder.
Can you tell me how can I update spyder 4 by Command Prompt or Anaconda command prompt?
Have a good day
Thank you

Bhanupratap


--
You received this message because you are subscribed to the Google Groups "spyder" group.
To unsubscribe from this group and stop receiving emails from it, send an email to spyderlib+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/spyderlib/8840f0f7-7145-44ea-8566-56b27645f2cf%40googlegroups.com.

Savin Beniwal

unread,
Dec 22, 2019, 8:28:48 PM12/22/19
to spyder
Hi B. Sahoo
To update spyder to 4.0 follow the below steps:

You can easily install if you use Anaconda by running:

conda update qt pyqt
conda install -c spyder-ide spyder=4.0.0rc1

Hope it helps you.
Best
------
To unsubscribe from this group and stop receiving emails from it, send an email to spyd...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages