anastasios-stamoulis / deep-learning-with-csharp-and-cntk Goto Github PK

How would you implement the LSTM example with non binary data. I was wondering if you could provide a quick explanation or example?

Try to run this but found this little problem:

"Error with "Ch_06_Using_Word_Embeddings"

System.IO.DirectoryNotFoundException: 'Could not find a part of the path 'C:\Users\anastasios\Downloads\aclImdb\train\neg'.'"

Hi,
It's great work you made.
recently I'm studying your code, I wanna trying CRNN network using CNTK && C#,
finally I failed, I'm not clear how to feed CNN's output into LSTM.
would you like make one example about this ?
thanks!

I get this error
DllNotFoundException: Unable to load DLL 'Cntk.Core.CSBinding-2.3.1.dll': The specified module could not be found. (Exception from HRESULT: 0x8007007E)
Can you help?
Thanks,
Stefano
PS: Windows 10
Visual Studio Comunity 2017

Nice work. Please make a LSTM - GRU Example...
Μπράβο 👍

I'm very interested in sorting this out but it may be beyond either of us. When I try to implement your Neural Style Transfer exercise with my solution (which uses CNTK 2.6 ) I come up stuck against this error:

"Input('combinationImage', [206 x 216 x 3], [, #]) -> Output('conv5_2', [13 x 14 x 512], [, #]), Output('conv1_1', [206 x 216 x 64], [, #]), Output('conv2_1', [103 x 108 x 128], [, #]), Output('conv3_1', [52 x 54 x 256], [, #]), Output('conv4_1', [26 x 27 x 512], [, #]), Output('conv5_1', [13 x 14 x 512], [*, #])' with more than one output cannot be implicitly converted to a Variable"

Does this error come when you switch to version 2.6 (or even 2.5.1)?

Hi, I'm Trying to get prediction of the model I have. for example in ch2 first look at a neural network.
but there is no function I can find to do so using cntk.Can you help me with that?

Great work, really helpful!

I caught one error in the with Using Pre-trained _Convnet Models - "'Cannot open file 'D:\kaggle_cats_dogs\cat.582.jpg"

The same error is thrown with "...Convnets with Small Datasets"

Hi
Thank for this great job ! very interesting.
I'm wondering if in evaluation phase with evaluation_phase() , the tensor to use is more x_test than x_train ?

Best regards
Thierry