hmchuong / ml-bonesuppression Goto Github PK

Hi,
I know its not a issue, but can you share your model weights (already trained)?
I'm trying a new approach and I would like something to compare.
Thanks!

I found another one though from MIT.

I am running the command:
python3 test.py --model ./model/model.index --config ./config/train.cfg --input test.png --output output2.png

I am getting the following errors:

1. File "/home/iplab/BoneSuppression/lib/python3.8/site-packages/tensorflow/python/client/session.py", line 1451, in _call_tf_sessionrun
   return tf_session.TF_SessionRun_wrapper(self._session, options, feed_dict,
   tensorflow.python.framework.errors_impl.NotFoundError: Tensor name "Variable" not found in checkpoint files ./model/model.index
   [[{{node save/RestoreV2}}]]
2. Traceback (most recent call last):
   File "/home/iplab/BoneSuppression/lib/python3.8/site-packages/tensorflow/python/training/py_checkpoint_reader.py", line 69, in get_tensor
   return CheckpointReader.CheckpointReader_GetTensor(
   RuntimeError: _CHECKPOINTABLE_OBJECT_GRAPH not found in checkpoint file

Is there any solution to this issue?

Thanks & Regards,
Sourav Das

Can you share link of your paper,thank you

hello, i cloned your repo in my computer, i created virtual environment and are you sure that these packages are compatible with each other? i have hard time to install all these packages

Hello, can I use only normal X-ray images with a pre-trained model to obtain bone-suppressed images as the output?

Hi! I try to run train process, but it's too slow couse of CPU training.
I've found this line in reqiuirments: tensorflow-gpu==1.11.0.
And in readme:

I have trained on an instance with 1 NVIDIA GTX 1080Ti (11GB VRAM) and it takes approximately 14 hours.

Could you please share training code for GPU? Thanks in advance for your reply.

Hi,
Thank you for sharing this project.
I'm wondering if you know how bone suppression was achieved on the original training images? The Kaggle dataset does not mention how.