Exploring Tensor Flow on
summary
Builds a spam classifier from a 5,000‑message dataset using Python and TensorFlow and layers created with TFlearn framework, about 30 lines of code. Training completes in roughly 12 seconds on the demo and the model predicts a sample spam at ~36% spam score versus 0.9% for ham. Exports a frozen .pb protobuf (checkpoint/index/meta files → frozen graph), embeds it in Android assets, runs CPU inference with the TensorFlow Android API, and the app effectively filters spam messages.
AI-written summary · about
chapters (7)
- 0:00Introduction to TensorFlow
- 3:19Machine learning basics
- 8:53TensorFlow concepts and graphs
- 13:07Building the spam classifier
- 18:21Training the neural network
- 20:26Evaluating and deploying model
- 23:18Summary and Q&A
related talks
46mTensorFlow for Mobile Poets
31mSupercharging Android Apps with Tensorflow
31mTensorflow for Janitors
47mDeep Learning: a practical introduction from an Android dev perspective
30mTensorFlow for Mobile Heroes
26mMachine Learning with TensorFlow
52mPractical Uses of Scala Tensorflow for Image Analysis
37mSimplifying Training Deep & Serving Learning Models...using Tensorflow