Overview / Usage

This optimized architecture speeds up the training process by at least 2x; it is also observed that convergence is achieved in fewer epochs than with Cycle-GAN. Also, by using optimization techniques specific to Intel AI DevCloud, up to 18x speed-up can be achieved.

Methodology / Approach

Intel AI DevCloud works especially well for testing research ideas. This is because it can independently perform computations on multiple nodes of the cluster. Thus, several ideas can be tested simultaneously without waiting for others to complete execution.

I have created an architecture which is modular in nature and each module can be trained independently on separate compute node from Intel AI DevCloud and communicate with other modules using shared storage. Utilization of each compute node effectively contributes for speeding up the training process.

Technologies Used

Intel AI DevCloud, Intel Optimized Tensorflow Framework, Intel® Xeon® Gold 6128 processors

Other links

• Improving Cycle-GAN using Intel® AI DevCloud
• An Optimized Architecture for Unpaired Image-to-Image Translation