Python version: 3.9.16 (main, Dec 7 2022, 01:11:51) [GCC 9.4.0] (64-bit runtime)
Python platform: Linux-5.10.147+-x86_64-with-glibc2.31
Is CUDA available: True
CUDA runtime version: 11.8.89
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration: GPU 0: Tesla T4
Nvidia driver version: 525.85.12
cuDNN version: Probably one of the following:
/usr/lib/x86_64-linux-gnu/libcudnn.so.8.7.0
/usr/lib/x86_64-linux-gnu/libcudnn_adv_infer.so.8.7.0
/usr/lib/x86_64-linux-gnu/libcudnn_adv_train.so.8.7.0
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_infer.so.8.7.0
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_train.so.8.7.0
/usr/lib/x86_64-linux-gnu/libcudnn_ops_infer.so.8.7.0
/usr/lib/x86_64-linux-gnu/libcudnn_ops_train.so.8.7.0
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True
Versions of relevant libraries:
[pip3] efficientnet-pytorch==0.7.1
[pip3] numpy==1.22.4
[pip3] pytorch-ranger==0.1.1
[pip3] segmentation-models-pytorch==0.3.2
[pip3] torch==1.13.1+cu116
[pip3] torch-optimizer==0.3.0
[pip3] torchaudio==0.13.1+cu116
[pip3] torchmetrics==0.11.3
[pip3] torchsummary==1.5.1
[pip3] torchtext==0.14.1
[pip3] torchvision==0.14.1+cu116
[conda] Could not collect
Composer information
Composer version: 0.13.2
Composer commit hash: None
Host processor model name: Intel(R) Xeon(R) CPU @ 2.20GHz
Host processor core count: 1
Number of nodes: 1
Accelerator model name: Tesla T4
Accelerators per node: 1
CUDA Device Count: 1
Go to 127.0.0.1:8888 and click on medical_image_segmentation.ipynb and Run all cells
Expected behavior
Composer should train a Unet model on the SIIM Pneumothorax dataset.
Additional context
It looks like the dataset has been removed from Kaggle Datasets. So in the notebook running the command !kaggle datasets download -d seesee/siim-train-test will result in 403 Error.
You can see this by going to the dataset location in your browser.
There is another Kaggle Dataset that can used from here . I created a branch on my fork with what I did to get it working (I updated the download location, some minor preprocessing steps and a few other things) . I can create a PR if the changes are okay.
Environment PyTorch information
PyTorch version: 1.13.1+cu116 Is debug build: False CUDA used to build PyTorch: 11.6 ROCM used to build PyTorch: N/A
OS: Ubuntu 20.04.5 LTS (x86_64) GCC version: (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0 Clang version: 10.0.0-4ubuntu1 CMake version: version 3.25.2 Libc version: glibc-2.31
Python version: 3.9.16 (main, Dec 7 2022, 01:11:51) [GCC 9.4.0] (64-bit runtime) Python platform: Linux-5.10.147+-x86_64-with-glibc2.31 Is CUDA available: True CUDA runtime version: 11.8.89 CUDA_MODULE_LOADING set to: LAZY GPU models and configuration: GPU 0: Tesla T4 Nvidia driver version: 525.85.12 cuDNN version: Probably one of the following: /usr/lib/x86_64-linux-gnu/libcudnn.so.8.7.0 /usr/lib/x86_64-linux-gnu/libcudnn_adv_infer.so.8.7.0 /usr/lib/x86_64-linux-gnu/libcudnn_adv_train.so.8.7.0 /usr/lib/x86_64-linux-gnu/libcudnn_cnn_infer.so.8.7.0 /usr/lib/x86_64-linux-gnu/libcudnn_cnn_train.so.8.7.0 /usr/lib/x86_64-linux-gnu/libcudnn_ops_infer.so.8.7.0 /usr/lib/x86_64-linux-gnu/libcudnn_ops_train.so.8.7.0 HIP runtime version: N/A MIOpen runtime version: N/A Is XNNPACK available: True
Versions of relevant libraries: [pip3] efficientnet-pytorch==0.7.1 [pip3] numpy==1.22.4 [pip3] pytorch-ranger==0.1.1 [pip3] segmentation-models-pytorch==0.3.2 [pip3] torch==1.13.1+cu116 [pip3] torch-optimizer==0.3.0 [pip3] torchaudio==0.13.1+cu116 [pip3] torchmetrics==0.11.3 [pip3] torchsummary==1.5.1 [pip3] torchtext==0.14.1 [pip3] torchvision==0.14.1+cu116 [conda] Could not collect
Composer information
Composer version: 0.13.2 Composer commit hash: None Host processor model name: Intel(R) Xeon(R) CPU @ 2.20GHz Host processor core count: 1 Number of nodes: 1 Accelerator model name: Tesla T4 Accelerators per node: 1 CUDA Device Count: 1
** To reproduce
Steps to reproduce the behavior:
In Colab
Open in Colab
Runtime -> Run all
On Linux machine
examples
directoryjupyter notebook
127.0.0.1:8888
and click onmedical_image_segmentation.ipynb
andRun all cells
Expected behavior
Composer should train a Unet model on the SIIM Pneumothorax dataset.
Additional context
It looks like the dataset has been removed from Kaggle Datasets. So in the notebook running the command
!kaggle datasets download -d seesee/siim-train-test
will result in403 Error
.You can see this by going to the dataset location in your browser.
There is another Kaggle Dataset that can used from here . I created a branch on my fork with what I did to get it working (I updated the download location, some minor preprocessing steps and a few other things) . I can create a PR if the changes are okay.