Grokking Machine Learning, video edition

Table of contents

1. Chapter 1. What is machine learning? It is common sense, except done by a computer
2. Chapter 1. What is machine learning?
3. Chapter 1. Some examples of models that humans use
4. Chapter 1. Example 4: More?
5. Chapter 2. Types of machine learning
6. Chapter 2. Supervised learning: The branch of machine learning that works with labeled data
7. Chapter 2. Unsupervised learning: The branch of machine learning that works with unlabeled data
8. Chapter 2. Dimensionality reduction simplifies data without losing too much information
9. Chapter 2. What is reinforcement learning?
10. Chapter 3. Drawing a line close to our points: Linear regression
11. Chapter 3. The remember step: Looking at the prices of existing houses
12. Chapter 3. Some questions that arise and some quick answers
13. Chapter 3. Crash course on slope and y-intercept
14. Chapter 3. Simple trick
15. Chapter 3. The linear regression algorithm: Repeating the absolute or square trick many times to move the line closer to the points
16. Chapter 3. How do we measure our results? The error function
17. Chapter 3. Gradient descent: How to decrease an error function by slowly descending from a mountain
18. Chapter 3. Real-life application: Using Turi Create to predict housing prices in India
19. Chapter 3. Parameters and hyperparameters
20. Chapter 4. Optimizing the training process: Underfitting, overfitting, testing, and regularization
21. Chapter 4. How do we get the computer to pick the right model? By testing
22. Chapter 4. A numerical way to decide how complex our model should be: The model complexity graph
23. Chapter 4. Another example of overfitting: Movie recommendations
24. Chapter 4. Modifying the error function to solve our problem: Lasso regression and ridge regression
25. Chapter 4. An intuitive way to see regularization
26. Chapter 4. Polynomial regression, testing, and regularization with Turi Create
27. Chapter 4. Polynomial regression, testing, and regularization with Turi Create :The testing RMSE for the models follow:
28. Chapter 5. Using lines to split our points: The perceptron algorithm
29. Chapter 5. The problem: We are on an alien planet, and we don’t know their language!
30. Chapter 5. Sentiment analysis classifier
31. Chapter 5. The step function and activation functions: A condensed way to get predictions
32. Chapter 5. The bias, the y-intercept, and the inherent mood of a quiet alien
33. Chapter 5. Error function 3: Score
34. Chapter 5. Pseudocode for the perceptron trick (geometric)
35. Chapter 5. Bad classifier
36. Chapter 5. Pseudocode for the perceptron algorithm
37. Chapter 5. Coding the perceptron algorithm using Turi Create
38. Chapter 6. A continuous approach to splitting points: Logistic classifiers
39. Chapter 6. The dataset and the predictions
40. Chapter 6. Error function 3: log loss
41. Chapter 6. Formula for the log loss
42. Chapter 6. Pseudocode for the logistic trick
43. Chapter 6. Coding the logistic regression algorithm
44. Chapter 6. Classifying into multiple classes: The softmax function
45. Chapter 7. How do you measure classification models? Accuracy and its friends
46. Chapter 7. False positives and false negatives: Which one is worse?
47. Chapter 7. Recall: Among the positive examples, how many did we correctly classify?
48. Chapter 7. Combining recall and precision as a way to optimize both: The F-score
49. Chapter 7. A useful tool to evaluate our model: The receiver operating characteristic (ROC) curve
50. Chapter 7. The receiver operating characteristic (ROC) curve: A way to optimize sensitivity and specificity in a model
51. Chapter 7. A metric that tells us how good our model is: The AUC (area under the curve)
52. Chapter 7. Recall is sensitivity, but precision and specificity are different
53. Chapter 7. Summary
54. Chapter 8. Using probability to its maximum: The naive Bayes model
55. Chapter 8. Sick or healthy? A story with Bayes’ theorem as the hero: Let’s calculate this probability.
56. Chapter 8. Prelude to Bayes’ theorem: The prior, the event, and the posterior
57. Chapter 8. What the math just happened? Turning ratios into probabilities
58. Chapter 8. What the math just happened? Turning ratios into probabilities:Product rule of probabilities
59. Chapter 8. What about two words? The naive Bayes algorithm
60. Chapter 8. What about more than two words?
61. Chapter 8. Implementing the naive Bayes algorithm
62. Chapter 9. Splitting data by asking questions: Decision trees
63. Chapter 9. Picking a good first question
64. Chapter 9. The solution: Building an app-recommendation system
65. Chapter 9. Gini impurity index: How diverse is my dataset?
66. Chapter 9. Entropy: Another measure of diversity with strong applications in information theory
67. Chapter 9. Classes of different sizes? No problem: We can take weighted averages
68. Chapter 9. Beyond questions like yes/no
69. Chapter 9. The graphical boundary of decision trees
70. Chapter 9. Setting hyperparameters in Scikit-Learn
71. Chapter 9. Applications
72. Chapter 10. Combining building blocks to gain more power: Neural networks
73. Chapter 10. Why two lines? Is happiness not linear?
74. Chapter 10. The boundary of a neural network
75. Chapter 10. Potential problems: From overfitting to vanishing gradients
76. Chapter 10. Neural networks with more than one output: The softmax function
77. Chapter 10. Training the model
78. Chapter 10. Other architectures for more complex datasets
79. Chapter 10. How neural networks paint paintings: Generative adversarial networks (GAN)
80. Chapter 11. Finding boundaries with style: Support vector machines and the kernel method
81. Chapter 11. Distance error function: Trying to separate our two lines as far apart as possible
82. Chapter 11. Training SVMs with nonlinear boundaries: The kernel method
83. Chapter 11. Going beyond quadratic equations: The polynomial kernel
84. Chapter 11. A measure of how close points are: Similarity
85. Chapter 11. Overfitting and underfitting with the RBF kernel: The gamma parameter
86. Chapter 12. Combining models to maximize results: Ensemble learning
87. Chapter 12. Fitting a random forest manually
88. Chapter 12. Combining the weak learners into a strong learner
89. Chapter 12. Gradient boosting: Using decision trees to build strong learners
90. Chapter 12. XGBoost similarity score: A new and effective way to measure similarity in a set
91. Chapter 12. Building the weak learners: Split at 25
92. Chapter 12. Tree pruning: A way to reduce overfitting by simplifying the weak learners
93. Chapter 13. Putting it all in practice: A real-life example of data engineering and machine learning
94. Chapter 13. Using Pandas to study our dataset
95. Chapter 13. Turning categorical data into numerical data: One-hot encoding
96. Chapter 13. Feature selection: Getting rid of unnecessary features
97. Chapter 13. Testing each model’s accuracy
98. Chapter 13. Tuning the hyperparameters to find the best model: Grid search