Forgive me if I got this wrong but it seems like the b_IJ are shared between all examples within a single batch (see reduce_sum and the shape).

I didn't see any mention of the batches in the paper, so I have assumed that there is a separate set of b_IJ weights for every batch. Why do you think that it's better to share those variables?

Edit: I've corrected the statement:

b_IJ are shared between all batches

to:

b_IJ are shared between all examples within a single batch

which is was I originally meant.

I think we should make a wechat group here for who interest this kind of subject. My wechat is bn31201 . Hope your adding, make some deep communicating.

Thanks for writing the code so shortly after the article was released. I'm trying to change the structure such that the capsule network can be trained for any image(x,y,z), but I am having trouble re-structuring the code. Can you help me identify which lines needs to be modified. I am guessing all lines with ... 28, 28, 1) -> ... 32, 32, 3) for CIFAR 10. But I am still not able to make it work.

Thank you again 👍

To the owner and all other visitors:

I do not mean to be offensive, but I decided to speak out my understanding of this routing algorithm as I have not seen any correct implementation so far yet.

The correct implementation of the routing algorithm should be treated something like the dynamic RNN in TensorFlow. In other words, if you implement it in a static way, and if you do 3 iterations, the two caps layers are actually 6 such layers. The primary layer performs line 4 and output to the digits layer, and then the digits layer performs line 5, 6, and 7 with b_ij updated, and then loop back to the primary layer again. This will need to use tf.while_loop if you use a dynamic way.

What confuses me or stops me from implementing myself is I am not sure how the weights and biases associated with the conv units are updated, as I assume other than the weights and biases associated with the capsules, each individual conv unit inside still carries its own parameters. Maybe I missed this by reading the paper.

Feel free to correct me if you believe I am wrong. Thanks.

Evaluating/Testing the trained model using python main.py --is_training=False gives the following error ValueError: Can't load save_path when it is None.

Hi!

I'm a student interested in Speech Synthesis with neural networks. I suppose this CapsNet might improve the quality of synthesized speech, so I try to apply this great program to the other program to generate artificial speech with neural network.

I would like to ask whether this CapsNet could replace other popular neural networks like CNN.

Thank you for answering.

the tf.contrib.layers.conv2d applies a relu activation,the PrimaryCap convolution does not included a relu activation before grouping neurons into capsules and then squashed, or did I miss something from the paper

https://github.com/naturomics/CapsNet-Tensorflow/blob/894c79cd8434d7de7784006c1646ff3107bb2a4f/capsLayer.py#L59

If squashing is done per capsule, then whey is the input dimensions to it 32, 1152, 8, 1, where 32 is the batch size? Shouldn't it be 32, 668, 32, 1?

https://github.com/naturomics/CapsNet-Tensorflow/blob/1e0668037447e89aca9173a688ce1965bf6a43c1/capsNet.py#L45

Why num_outputs is set in this while it will not be used?

Hi Huadong,

I've been running succesful tests of CapsNets with Pytorch and would like to compare notes with you. Maybe we can take our discussion offline? My email is: firstname.lastname[@]gmail.com

Let me know!

Tarry

https://github.com/naturomics/CapsNet-Tensorflow/blob/master/capsLayer.py#L151

            # then matmul in the last tow dim: [16, 1].T x [16, 1] => [1, 1], reduce mean in the
            # batch_size dim, resulting in [1, 1152, 10, 1, 1]
            v_J_tiled = tf.tile(v_J, [1, 1152, 1, 1, 1])
            u_produce_v = tf.matmul(u_hat, v_J_tiled, transpose_a=True)
            assert u_produce_v.get_shape() == [cfg.batch_size, 1152, 10, 1, 1]
            b_IJ += tf.reduce_sum(u_produce_v, axis=0, keep_dims=True)

Why would you need to average b across batch dimension? I don't see why would that be good, since that would make the model batch-size dependent. If there is any mention on this in the paper or other source, can you point out where and send a link, appreciated.

Is there a way to save this model in SavedModel or frozen_graph format? I am trying to do inference using Tensorrt, which only supports these two formats.

thanks your codes. I finished the normal version, but when I used the dist_version, it occurs a issue. The errors shows "AssertionError :assert not np.isnan(loss_value)" in distributed_train.py. in line 132. how's it going? can you help me ?

1. Y = valY[:num_val_batch * cfg.batch_size].reshape((-1, 1))

what is the use of this line in main.py? 2.Can we use the code for RGB image dataset? 3.What is the use of channel parameter in CapsNet function?