We’re excited to share that weight pruning is now part of the TensorFlow Model Optimization Toolkit! Learn how sparsity can dramatically reduce model sizes with negligible accuracy loss. Read here → goo.gle/2vYWBrO

@TensorFlow Wooooo faster inference time

@TensorFlow @fchollet Excellent ✨🙂

@TensorFlow Is this project related to the work that was done to reduce the model size for google assistant (which has lead to the local one device deployment) ?

@TensorFlow It looks like GNMT EN-DE improves by .19 with 80% forced sparsity... did any other models in your experiments show improvement? Are we seeing in improvement because forced sparsity help to combat over fitting and allows the model to generalize better in some scenarios?

@TensorFlow Nice.

@TensorFlow woohooo nice.... gonna implement soon 😀😀