karpathy / neuraltalk

NeuralTalk is a Python+numpy project for learning Multimodal Recurrent Neural Networks that describe images with sentences.
5.4k stars 1.32k forks source link

predict_on_images.py: error: too few arguments #33

Open RanAR90 opened 8 years ago

RanAR90 commented 8 years ago

Hello.. Thanks for the code and the very helpful read me files.. I tried to call the predict_on_images.py on the examples folder you supported but got this error C:\neuraltalk-master>python predict_on_images.py usage: predict_on_images.py [-h] [-r ROOT_PATH] predict_on_images.py: error: too few arguments

I would appreciate any help ...

Regards

phiresky commented 8 years ago

Yf Am 28.10.2015 15:11 schrieb "RaniemAR" notifications@github.com:

When I am calling the code I am getting this error:

Outputs will be scaled by mean and standard deviation specified in NC file. Computing outputs for data fraction 1...FAILED: CUBLAS matrix multiplication failed

it happens after creating the neural network is done and this error message shows after :

Creating the neural network... done. Layers: (0) input size: 4096 lstm size: 512, bias: 1.0, weights: 9440768 softmax size: 8148, bias: 1.0, weights: 4179924 multiclass_classification [size: 8148] Total weights: 13620692

Any help would be appreciated .. Thank you very much..

— Reply to this email directly or view it on GitHub.