From 65aca0abe26e39a8e8b671fbf2afabf5af71b96f Mon Sep 17 00:00:00 2001 From: Wirawan Purwanto Date: Wed, 20 Jul 2022 23:11:35 -0400 Subject: [PATCH] * Initial notes on TensorFlow/GPU troubleshooting. --- tensorflow/20220601.GPU-troubleshooting.md | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) create mode 100644 tensorflow/20220601.GPU-troubleshooting.md diff --git a/tensorflow/20220601.GPU-troubleshooting.md b/tensorflow/20220601.GPU-troubleshooting.md new file mode 100644 index 0000000..66c43eb --- /dev/null +++ b/tensorflow/20220601.GPU-troubleshooting.md @@ -0,0 +1,20 @@ +Troubleshooting TensorFlow / KERAS GPU Process +============================================== + + +Common issues: + +* Model too big + (print the KERAS model---is the number of parameters doable with the GPU?) + +* Batch size too large + +* Memory not released from previous processing + + (use `with tf.Session() as sess:` code block) + + + + + +