Abhilash Mahendrakar
Abhilash Mahendrakar
Was able to reproduce the issue with TF v2.2 and TF-nightly. Please find the gist of it [here](https://colab.research.google.com/gist/amahendrakar/26897d798c5f763191014d5a558564e3/40727.ipynb). Thanks!
Was able to reproduce the issue with TF v2.3. Please find the gist of it [here](https://colab.research.google.com/gist/amahendrakar/b5978e7b96b2e44a0b473bfd8309ecdb/44873.ipynb). Thanks!
@dd1923, On running the usage example given in the [MeanIoU documentation](https://www.tensorflow.org/api_docs/python/tf/keras/metrics/MeanIoU), the output I got was similar to the example. Please find the gist of it [here](https://colab.research.google.com/gist/amahendrakar/4ee1f3307e5ceb596b6c6628408ecd17/39173.ipynb). Could you please...
Was able to reproduce the issue with [TF v2.1](https://colab.research.google.com/gist/amahendrakar/d79131d083e48803fbe32b9ed3fd5c43/39173-2-1.ipynb), [TF v2.2.0-rc4](https://colab.research.google.com/gist/amahendrakar/d375b59b12d50c91dcd8d474b274fe32/39173.ipynb) and [TF-nightly](https://colab.research.google.com/gist/amahendrakar/8ab2ef9f8ff4d22de6e32aabbf29eced/39173-tf-nightly.ipynb). Please find the attached gist. Thanks!
Was able to reproduce the issue with TF 2.1 and TF nightly. Please find the Gist [here](https://colab.sandbox.google.com/gist/amahendrakar/098f60ba73339d773791c538b7fdea14/36911.ipynb). Thanks!
@Flamefire, TensorFlow 2.4 is tested and built against CUDA 11 and cuDNN 8. For more information, please take a look at the [tested build configurations](https://www.tensorflow.org/install/source#gpu). Could you please check if...
@Flamefire, In order to expedite the trouble-shooting process, could you please provide the exact sequence of commands / steps that you executed before running into the error. Also, TensorFlow 2.4...
Was able to reproduce the issue with TF v2.3 and TF v2.4. Please find the gist of it [here](https://colab.research.google.com/gist/amahendrakar/3c18dcfe8acc0183de0281eaf35abca9/45849.ipynb). Thanks!
@matthieucoquet, > It seems using nested name_scope can lead to memory leak. For example, in my setup, the memory grows indefinitely using this code. I did not observe any memory...
Was able to reproduce the issue with the latest [TF-nightly](https://colab.research.google.com/gist/amahendrakar/2e47c65e4cb30ab59a3fc83d3e288137/45322-tf-nightly.ipynb), i.e. v2.5.0-dev20201208. However, I did not observe much increase in the memory usage with [TF v2.2](https://colab.research.google.com/gist/amahendrakar/03b1a1862050fc7013ad3e217f828d3a/45322-2-2.ipynb) and [TF v2.3](https://colab.research.google.com/gist/amahendrakar/f5b75de4bef3971db0edd28effa642a7/45322-2-3.ipynb). Please...