Sentiment Classification & How To "Frame Problems" for a Neural Network
- neural networks, forward and back-propagation
- stochastic gradient descent
- mean squared error
- and train/test splits
- Leverage the recommended Course Reading Material - Grokking Deep Learning
-
Intro: The Importance of "Framing a Problem" (this lesson)
[Curate a Dataset](#lesson_1) [Developing a "Predictive Theory"](#lesson_2)
-
PROJECT 1: Quick Theory Validation
[Transforming Text to Numbers](#lesson_3)
-
PROJECT 2: Creating the Input/Output Data
Putting it all together in a Neural Network
-
PROJECT 3: Building our Neural Network
[Understanding Neural Noise](#lesson_4)
-
PROJECT 4: Making Learning Faster by Reducing Noise
[Analyzing Inefficiencies in our Network](#lesson_5)
-
PROJECT 5: Making our Network Train and Run Faster
[Further Noise Reduction](#lesson_6)
-
PROJECT 6: Reducing Noise by Strategically Reducing the Vocabulary
[Analysis: What's going on in the weights?](#lesson_7)
NOTE: unzip file reviews.rar into reviews.txt before run code