Rashmi Nagpal - Build your Machine Learning Model on Edge with React Native

React

Build machine learning models on edge devices with React Native, optimizing for limited memory and computing resources, and enabling faster response times and reduced latency.

Key takeaways

Machine learning (ML) on edge devices is the trend, and React Native is one way to achieve this.
Edge devices have limited memory and computing resources, which need to be optimized.
Bulky ML models won’t work on edge devices; they need to be fine-tuned.
Transfer learning, pruning, and quantization are techniques to optimize ML models.
Edge computing enables faster response times and reduced latency.
Machine learning frameworks like TensorFlow.js, OpenAI’s Chart GPT, and Core ML are available for edge device development.
Generative modeling and deep learning are subsets of machine learning.
React Native provides tools for building mobile applications with ML capabilities.
The use cases for ML on edge devices include image classification, object detection, and prediction.
Machine learning models need to be fine-tuned for edge devices, with parameters such as epochs, batch size, and model size.
Quantization is significant in reducing the precision of numbers in ML models.
Edge computing can lead to reduced cloud costs and improved data security.
React Native allows developers to reuse code and leverage existing ML frameworks.
Machine learning models on edge devices can be used for personalized recommendations and predictions.
Edge devices have limitations, such as memory constraint, latency, and computation resources, which need to be addressed.
The future of machine learning is in edge devices, and React Native is one way to achieve this.
React Native provides a framework for building machine learning models on edge devices.

Rashmi Nagpal - Build your Machine Learning Model on Edge with React Native

More talks