stream ceed1a31db old argument targets->labels of tf.nn.softmax_cross_entropy_with_logits for tf 1.0		hace 8 años
..
data	41c52d60fe Spatial Transformer model	hace 9 años
README.md	d816971032 Use tf.softmax_cross_entropy_with_logits to calculate loss (#181)	hace 9 años
cluttered_mnist.py	ceed1a31db old argument targets->labels of tf.nn.softmax_cross_entropy_with_logits for tf 1.0	hace 8 años
example.py	31f1af580a Changed deprecated tf.initialize_all_variables() to tf.global_variables_initializer()	hace 8 años
spatial_transformer.py	052e5e8b6e Converted the models repo to TF 1.0 using the upgrade script	hace 8 años
tf_utils.py	41c52d60fe Spatial Transformer model	hace 9 años

Spatial Transformer Network

The Spatial Transformer Network [1] allows the spatial manipulation of data within the network.

API

A Spatial Transformer Network implemented in Tensorflow 0.7 and based on [2].

How to use

transformer(U, theta, out_size)

Parameters

U : float 
    The output of a convolutional net should have the
    shape [num_batch, height, width, num_channels]. 
theta: float   
    The output of the
    localisation network should be [num_batch, 6].
out_size: tuple of two ints
    The size of the output of the network

Notes

To initialize the network to the identity transform init theta to :

identity = np.array([[1., 0., 0.],
                    [0., 1., 0.]]) 
identity = identity.flatten()
theta = tf.Variable(initial_value=identity)

Experiments

We used cluttered MNIST. Left column are the input images, right are the attended parts of the image by an STN.

All experiments were run in Tensorflow 0.7.

References

[1] Jaderberg, Max, et al. "Spatial Transformer Networks." arXiv preprint arXiv:1506.02025 (2015)

[2] https://github.com/skaae/transformer_network/blob/master/transformerlayer.py

README.md