framework-reproducibility
framework-reproducibility copied to clipboard
NVIDIA
Providing reproducibility in deep learning frameworks
A list of all the torch operations and their respective determinism status would be very useful.
For example: In my project (https://github.com/edwardyehuang/CAR/blob/master/carnet.py), line 163 (instantiate SegManaged). If I create a wrapper class that inherited SegManaged and place it on line 163, e.g. ``` class gbb(SegManaged): pass...
## Problem Overview I train my model on the same dataset in two different setups: A) single-gpu, B) multi-gpu. The former leads to deterministic results, the latter leads to non-deterministic...
Model nondeterminism when using `K.layers.DepthwiseConv2D` / `tf.nn.depthwise_conv2d` (known issue)
Dear @duncanriach, I'm using TF2.3 on Ubuntu 16.04. To get deterministic results, I followed your instructions. Please check the attached code that is very simple MNIST example. After running the...
Dear @duncanriach, Thank you for your contributions, work and guidance towards making _tensorflow_ deterministic in the recent releases. Unfortunately, for popular keras NLP models (BERT) some problems seem to remain...
I am getting AttributeError for tf 1.9, saying that tensorflow doesn't have "version" as attribute. Can you help?
### Prerequisites I am using latest TensorFlow NGC container nvcr.io/nvidia/tensorflow:20.09-tf2-py3 ### Issue When running my Deep Learning model on 1 GPU (Tesla T4) within the NGC container 20.09, I don't...
Running TF 2.4.1 with seeds and envs set I'm getting different results each run for this guy: https://github.com/garder14/byol-tensorflow2 I currently suspect it's the gradient tape. Not sure how to handle...
A branch for dynamic patch of (sparse)_softmax_cross_entropy_with_logits. It is updated with the head of origin/master as of 02/17/2021 to make later merge as smooth as possible.
Model not deterministic/reproducible with all seeds set and os.environ['TF_DETERMINISTIC_OPS'] = '1'
in reference of this [kaggle notebook](https://www.kaggle.com/devang/transfer-learning-with-keras-and-efficientnets) i've written this [notebook](https://drive.google.com/file/d/1VyaUGXgaCDIbWbt0-F0iHTIdokxWesJH/view?usp=sharing) to try transfer learning through multiple datasets. through multiple soft runs on the notebook, I came to found out that...
