A bug to solve

[i3esn0w]

Traceback (most recent call last): File "clock_gated_rnn.py", line 63, in model.compile(loss='binary_crossentropy', optimizer='adam', class_mode="binary") File "/usr/local/lib/python2.7/dist-packages/keras/models.py", line 287, in compile self.y_train = self.get_output(train=True) File "/usr/local/lib/python2.7/dist-packages/keras/layers/containers.py", line 51, in get_output return self.layers[-1].get_output(train) File "/usr/local/lib/python2.7/dist-packages/keras/layers/core.py", line 223, in get_output X = self.get_input(train) File "/usr/local/lib/python2.7/dist-packages/keras/layers/core.py", line 31, in get_input return self.previous.get_output(train=train) File "/usr/local/lib/python2.7/dist-packages/keras/layers/core.py", line 341, in get_output X = self.get_input(train) File "/usr/local/lib/python2.7/dist-packages/keras/layers/core.py", line 31, in get_input return self.previous.get_output(train=train) File "build/bdist.linux-x86_64/egg/einstein/layers/recurrent.py", line 714, in get_output File "/usr/local/lib/python2.7/dist-packages/theano/scan_module/scan.py", line 745, in scan condition, outputs, updates = scan_utils.get_updates_and_outputs(fn(*args)) File "build/bdist.linux-x86_64/egg/einstein/layers/recurrent.py", line 694, in _step AttributeError: 'module' object has no attribute 'ifelse'

i use the code that you implement in example , but it can not run. And I check the code .there is no error , how i fix it ?

[gaoyuankidult]

Dear i3esn0w

Thanks for your input. However, this repo uses an older version of Keras and is outdated. Now it is only used for storing my previous experiment files.

I changed description to be

This repo depends on an older version of keras and is outdated. Now it only serves as a place to store my previous experiments. One may face a lot of errors trying to run these codes.

ClockworkSGU is a clockwork version of my previous model SGU. I don not know what would you like to do with it.

If you would like to port Clockwork to some other libraries, you can look at this class (link). I think anyone with experiences with older keras should know how to port it.

If you would like to check the code of SGU and DSGU, please have a look at this link. Actually both SGU and DSGU have some problems with multi-class classification (it only gives best class but does not provide a probabilistic distribution over all classes).

Please do not hesitate to tell me what would you like to do with the code, so I can help you with it.

[white54503]

I've been using DSGU w/ softmax l2_activation for multi-label classification in an RL-setting, with stellar results. GRUs comprise the hidden state, DSGU does the output. Network inputs range (-inf, inf). Do you see any reason this shouldn't be a viable approach? Thanks, and great work... I was sad to see the arxiv paper come down.

[gaoyuankidult]

Dear white54503

Did you get a good result ?

As mentioned in the paper, DSGU uses a sigmoid function as an activation function. Combining it with categorical cross entropy can make the classification right but output does not follow a probabilistic distribution. This is not useful in many cases (probably more experiments are needed if somebody want to continue investigation on multiplicative gate). That is why I withdrawn my paper.

Using DSGU + softmax may not be a good choice as DSGU+softmax did not work on my experiments (DSGU + sigmoid worked).

Actually, I suggest you look at this paper. By using batch normalization on LSTM, they managed to reach 99% on MNIST dataset (with softmax output layer), which is at least higher than my result. Keras recently has a pr about it (link). However, I never implemented batch normalization on my model.

[white54503]

Performance is quite good where other architectures have struggled mightily... I'm using DSGU in an asynchronous actor-critic setup similar to http://arxiv.org/pdf/1602.01783v1.pdf; loss is the negative of actor's advantage vs critic baseline. Classification by the max of the network's output yields control vectors; there is no probabilistic interpretation of the class labels. I'll test sigmoid vs. softmax and report results in a week or two.

[i3esn0w]

First of all, thanks for your reply.I use it to sentiment analyse,and I need tri-class classification.My old code are base keras 1.0 . So I will consider to port it to keras 1.0.

[gaoyuankidult]

@white54503 That is quite interesting. My original purpose of designing this network is for control problem as well, which does not require any probabilistic distribution. I am really interested in how did you use softmax+DSGU for your problem. I will read the paper and try to understand your system. Let us continue discussion latter. btw, maybe we can discuss here.

[gaoyuankidult]

@i3esn0w Which one would you like to port ? DSGU or ClockworkRNN ? If you would like to port DSGU, there is a ported version here. If you would like to port ClockworkRNN, you should look at this class.

如果你懂中文也可以中文说。

[i3esn0w]

好吧……我想试试ClockworkRNN

[gaoyuankidult]

@i3esn0w 我最近比较忙，可能没时间完成转成keras 1.0 的工作，不过我的step 函数 已经写recurrent.py里面了。

我建议你写一个支持Keras 1.0的ClockworkRNN函数，这样大家也会用的。

不过Clockwork RNN 非常慢，我不是很推荐用这个做sentiment analysis。