Deep Learning Like a Viking: Building Convolutional Neural Networks with Keras
summary
Build a convolutional neural network in Python using Keras to recognize hand-drawn runes from the Younger Futhark, train the model on rune images, and integrate it into a web app with Flask and client-side JavaScript so users can draw runes and get predictions.
AI-written summary · about
chapters (20)
- 0:00Start
- 4:35The Younger Futhark
- 8:05Harald Bluetooth
- 9:02Younger Futhark vs. Latin & Greek
- 9:40Common Ancestors
- 11:21Logic Gates with Actual Numbers
- 11:48Neurons with Actual Numbers
- 12:08Inside a Neuron
- 13:50Even Mathier
- 15:30Step Function
- 16:17Sigmoid Function
- 16:56tanh Function
- 17:23Linear Function
- 18:11Rectified Linear Units (ReLU)
- 20:22Bad at Recognizing Runes
- 23:47Convolutions with Fancy Animation
- 24:40Convolutions with Numbers
- 25:10Convolutions with Familiar Looking Numbers
- 28:14Convolutional Neural Networks
- 42:56Image Credits
related talks
44mDeep Learning for Developers
21mRecognizing Human features using Deep Networks
37mCode First, Math Later: Learning Neural Nets Through Implementation and Examples
49mExploring Neural Networks for Recognition of Hand-drawn Shapes
39mRun Deep Learning models in the browser with JavaScript and ConvNetJS
24mConvolutional Neural Networks in Action
27mDemystifying Deep Neural Networks
51mA Practical Introduction to Convnets for Image Classification