Tensorflow: Learning Machines
summary
Covers hands-on TensorFlow usage with live demos: linear regression via y=mx+b and gradient descent, a browser neural-net playground classifying spirals, and retrained Inception on 300 images (100 per class: ramen, carbonara, pancit) using bottlenecks and 500 training steps. Shows TensorBoard embeddings and loss/accuracy plots, exports a served model, and reports ~97–99% train/test accuracy on held-out images. Notes CPU vs GPU/TPU, recommends running training on Google Cloud, and mentions AutoML for automating model/step selection.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 0:43What is Tensorflow
- 3:20Why Tensorflow
- 7:47Can I use Tensorflow
- 9:19How does it work
- 11:10Coding
- 12:44Learning
- 14:21Descent Gradient
- 16:25Code
- 19:23Google Team
- 19:39Classification
- 22:30Optimization
- 34:12Estimations
- 36:56Embedding
- 38:27Network
- 39:30Versioning
- 41:25Who uses it
related talks
51mMachine Learning Exposed Introduction to Machine Learning
44mTensorflow, Tensorflow, What is it that you see?
21mA Journey into the World of Machine Learning with TensorFlow.js
45mThe most soft and fluffy intro into Machine Learning and Deep Neural Networks
54mAdventures in Javascript Machine Learning
13mTaking Machine Learning to the Next Level
49mMachine Learning Exposed: Introduction to Machine Learning
01h 06mDeep Learning with TensorFlow