search

ZFTurbo / KAGGLE_DISTRACTED_DRIVER

Solutions
91 stars 53 forks source link

AssertionError: AbstractConv2d Theano optimization failed #2

Open Stacktohack opened 8 years ago

Stacktohack commented 8 years ago

This is the error encountered when running model.fit AssertionError: AbstractConv2d Theano optimization failed: there is no implementation available supporting the requested options. Did you exclude both "conv_dnn" and "conv_gemm" from the optimizer? If on GPU, is cuDNN available and does the GPU support it? If on CPU, do you have a BLAS library installed Theano can link against?

Can you please tell me what install am i missing?

ZFTurbo commented 8 years ago

You should check that your config of theano is correct: http://deeplearning.net/software/theano/library/config.html

Stacktohack commented 8 years ago

Thanks @ZFTurbo for the link. I just ran the code without border_mode='same', init='he_normal' . And this resolved the error. But How critical were these parameters? As i am getting a submitted test log loss scoreof about 1.7. (Validation log loss was 1.4)

guddulrk commented 7 years ago

Hi I am getting following error. Any help please

AssertionError: AbstractConv2d Theano optimization failed: there is no implementation available supporting the requested options. Did you exclude both "conv_dnn" and "conv_gemm" from the optimizer? If on GPU, is cuDNN available and does the GPU support it? If on CPU, do you have a BLAS library installed Theano can link against?

I am running python3.5 with CPU mode