kendemu

Oh, I found this method in the test-clstm.py. SGD+momenum. ``` python clstm.sgd_update(net) ``` Is there adagrad, NAG, or faster solver? SGD is quite slow.

OK, I saw the [ocropus-lpred](https://github.com/tmbdev/ocropy/blob/master/ocropus-lpred#L47) code using clstm module add the following code. Hoping it's going to work. ``` python #setup codec charset_file = open("./lang/charset_jp.txt","r") self.charset = charset_file.read() charset_file.close() self.charset...

Thank you for the advice! I succeeded loading the neural network model and the inference! However, the Japanese model I found on the Internet seemed that the data is corrupt(it...

Hello, I'm doing a dynamic driver configuration from electron UI & IPC by using require. I think you can re-implement this code using Socket.io or MQTT. But probably you need...

I'm also stuck with the same problem. By reading the code, I found that caffe-tensorflow only implemented Softmax Layer for FCN. Use softmax or you have to define def map_sigmoid()...

@xuyinbo @chiggum In Network class from [kaffe/tensorflow/network.py](https://github.com/ethereon/caffe-tensorflow/blob/master/kaffe/tensorflow/network.py), implement method sigmoid. The transformer.py have an ability to do reflection of the Network class methods, so you don't have to modify transformer.py....