search

sunwj / CAR

Content adaptive resampler for image downscaling
GNU General Public License v3.0
472 stars 80 forks source link

Unable to run pre-trained model in Google Colab free GPU #6

Open AnanthaNatarajan opened 4 years ago

AnanthaNatarajan commented 4 years ago

I'm trying to make use of the trained model in Colab, but i get the following error during evaluation :

CUDA error at adaptive_gridsampler_kernel.cu:76 code=209(cudaErrorNoKernelImageForDevice) "cudaGetLastError()"

Colab Specs:

nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2019 NVIDIA Corporation

colab error

Built on Sun_Jul_28_19:07:16_PDT_2019 Cuda compilation tools, release 10.1, V10.1.243

physical_device_desc: "device: 0, name: Tesla K80, pci bus id: 0000:00:04.0, compute capability: 3.7"

Please let me know the exact specs required to run the trained model

davertor commented 3 years ago

I have the same issue. Impossible to run inference on colab for me.

Phylake1337 commented 3 years ago

Managed to infer the model on google colab, here's how did I do it..

after the compiling step, you gotta download python file, then you should copy the file into the project file /CAR, This file is a modified version of run.py.

make sure to make a folder that contains all the low-resolution images you need to upscale in png format.

then run this command..

python3 run_LR.py --scale 4 --img_dir path_to_images_folder --model_dir path_to_pretrained_models \
--output_dir path_to_output