Training on a GPU

[vvkv]

I am attempting to accelerate my training process by training this network on an NVIDIA titan x.

To do this I go into my conda environment and do the following:

conda create -n tensorflowGPU python=3.5
source activate tensorflowGPU
conda install pandas matplotlib jupyter notebook scipy scikit-learn
pip install tensorflow-gpu

I then follow this up with the training code as follows:

python retrain.py
--bottleneck_dir=bottlenecks
--how_many_training_steps=500
--model_dir=inception
--summaries_dir=training_summaries/basic
--output_graph=retrained_graph.pb
--output_labels=retrained_labels.txt
--image_dir=flower_photos

The expectation here is for the training speed to accelerate, however this does not seem to be happening. Is there an explicit declaration/change that needs to be made on the retrain.py script to run it with a GPU?

Thank you

[martinblampied]

Ever get an answer to this?