mauro-belgiovine
/ tf_env_variables.sh
Created
June 12, 2024 22:02
This bash script exports all the CUDA libraries in Ubuntu/Linux environment for correct visibility of GPU devices. This is especially useful when using Tensorflow (> 3.15) in a Conda environment, which is not a default library location.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # HOW TO RUN IN TERMINAL: | |
| # $ source tf_env_variables.sh | |
| NVIDIA_DIR=$(dirname $(dirname $(python -c "import nvidia.cudnn;print(nvidia.cudnn.__file__)"))) | |
| for dir in $NVIDIA_DIR/*; do | |
| if [ -d "$dir/lib" ]; then | |
| export LD_LIBRARY_PATH="$dir/lib:$LD_LIBRARY_PATH" | |
| fi | |
| done |
mauro-belgiovine
/ tf_cond_example_gradients.py
Last active
April 18, 2024 17:17
Example of tf.cond usage to allow gradient propagation through conditional branches
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| import tensorflow as tf | |
| # X, Y and Z are all watched variables, although they will produce a gradient (i.e. not None) only if the branch executed has uses that variable | |
| with tf.GradientTape() as tape: | |
| x = tf.Variable(5.0, dtype=tf.float32, name='X') | |
| y = tf.Variable(6.0, dtype=tf.float32, name='Y') | |
| z = tf.Variable(8.0, dtype=tf.float32, name='Z') | |
| cond = tf.cond(pred = x < y, | |
| true_fn = lambda: tf.add(x, z), |