Closed oorosh closed 4 years ago
I'm trying to run this solution
https://github.com/dotnet/machinelearning-samples/tree/master/samples/csharp/getting-started/DeepLearning_ImageClassification_Training
with GPU , but im getting this error.
"Can't copy 80080 bytes of a tensor into another with 40040 bytes buffer. [[{{node _arg_input_1/BottleneckInputPlaceholder_0_0}}]]"
in this line of code
ITransformer trainedModel = pipeline.Fit(trainDataView);
This are versions of nuget packages that I'm using
I have setup environment following this article
https://github.com/dotnet/machinelearning/blob/master/docs/api-reference/tensorflow-usage.md
I have CUDA 10.0 and CUDDN cudnn-10.0-windows10-x64-v7.6.5.32
GPU geforce GTX 950M with 2GB GDDR5
OS Windows 10 pro 64bit
framework core 3.1
I have fix my issue by updating to at this moment latest version of nuget packages
I'm trying to run this solution
https://github.com/dotnet/machinelearning-samples/tree/master/samples/csharp/getting-started/DeepLearning_ImageClassification_Training
with GPU , but im getting this error.
"Can't copy 80080 bytes of a tensor into another with 40040 bytes buffer. [[{{node _arg_input_1/BottleneckInputPlaceholder_0_0}}]]"
in this line of code
ITransformer trainedModel = pipeline.Fit(trainDataView);
This are versions of nuget packages that I'm using
I have setup environment following this article
https://github.com/dotnet/machinelearning/blob/master/docs/api-reference/tensorflow-usage.md
I have CUDA 10.0 and CUDDN cudnn-10.0-windows10-x64-v7.6.5.32
GPU geforce GTX 950M with 2GB GDDR5
OS Windows 10 pro 64bit
framework core 3.1