DeepOBS

DeepOBS copied to clipboard

get_performance_dictionary doesn't provide the desired metric

1

The interface of the function is ``` def get_performance_dictionary( optimizer_path, mode="most", metric="valid_accuracies", conv_perf_file=None ): ``` But, despite providing e.g. "valid_accuracies", the function returns sometimes the "test_accuracies". The explanation can be...

andres-fr

Bumps [tensorflow](https://github.com/tensorflow/tensorflow) from 1.13.1 to 2.7.2. Release notes Sourced from tensorflow's releases. TensorFlow 2.7.2 Release 2.7.2 This releases introduces several vulnerability fixes: Fixes a code injection in saved_model_cli (CVE-2022-29216) Fixes...

dependabot[bot]

Feature Request: Add support for pytorch

8

(This is a test issue. As mentioned in the responses to #3 and #4 , there is a development branch that supports pytorch. find it here: https://github.com/abahde/DeepOBS) ## Expected behavior...

ludwigbald

Introduces a PyTorch version of the Tolstoi Char RNN (previously only in Tensorflow).

schaefertim

An attempt to upgrade deepOBS tensorflow to 2.0 - [x] run `tf_upgrade_v2` script on `deepobs` (cf. https://www.tensorflow.org/guide/upgrade), commit 5550bea - [x] try to fix errors in `reports.txt` with replacements from...

FelixBenning

- Uses `extras_require=` in the ` setup.py` to achieve the amount of custom installation variants currently achieved with many different `requirements.txt` e.g. ```txt pip install -e .[dev] ``` - Update...

FelixBenning

- Runners should also log the wall-clock time for logged values. We can extract the aforementioned time to achieve a target error in the post-processing step. - For completeness, and...

fsschneider

Problem: `cifar100_wrn404` forward pass is not working when adding regularization. This is caused by a missing return statement that creates the regularization groups. Demo: - Before: Forward pass without regularization...

f-dangel

Added DCGAN test problem class for PyTorch: - mnist_dcgan - fmnist_dcgan - afhq_dcgan - celeba_dcgan

Vanessa-Ts

The `quadratic_deep` problem for PyTorch has been updated and slightly changed. It is now re-written as a sequential "neural network". This allows compatibility for example with BackPACK. The TensorFlow version...

fsschneider