Hello trend,
This is Satya Mallick from LearnOpenCV.com.
In today's post, we will take a deep dive into Model Optimization using the Tensorflow Model Optimization toolkit. The TensorFlow Model Optimization Toolkit is a suite of tools for optimizing ML models for deployment and execution. The model optimization techniques that we will discuss in this post are:
- Pruning: We remove unnecessary connections in the network and thus reduce the size of the model.
- Weight Clustering: It is a technique designed to decrease model storage requirements using clustering and data compression algorithms. The weights are saved using lesser number of bits.
- Quantization Aware Training (QAT): Quantization results in loss of information. So, Quantization Aware Training tries to minimize the quantization loss by adding it to the loss function and minimize it during training. Thus, achieving a more robust quantized model.
Without further ado, let's get started.
| TensorFlow Model Optimization Toolkit |
You can check out the code by clicking on the button below, and star us on GitHub to say thanks!
| Download Code (GitHub) |
The TFLite and Model Optimization Series
This is the complete list of posts in the TFLite and model optimization Series:
- TensorFlow Lite: Model Optimization for On-Device Machine Learning
- TensorFlow Lite Model Maker: Create Models for On-Device Machine Learning
- TensorFlow Model Optimization Toolkit: Deeper Dive into Model Optimization
Please let us know if you want us to write more articles on TFLite and model optimization by replying to this email.
Cheers!
Satya
Courses / YouTube / Facebook / LinkedIn / Twitter / Instagram
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.