See notebooks in the nb dir. Some extras:
- Intro
- A J.P. Morgan report on ML in banking incl. some experiments of their own
- Iris dataset exploration
- Credit fraud detection
- Text classification, Reuters21578 news dataset
- Pandas and Numpy Intro / Cheatsheets
6, 7. Logistic regression, decision trees, boosting, hyperparameter tuning -- discussed offline
8a. SVM -- a brief comparison of kernels
8b. Topic modeling applied to get extra features
9a. Some experiments with word2vec incl. a word analogies demo and text classification
9b. URLs to some curious resources about neural networks
- Vorontsov / Yandex / MIPT
- intro course, short simple videos (but with homework takes 7w)
- specialization based on the previous course
- the YDS course both above are actually based on
- its wiki page with slides and notes
- Natural language processing overview course
- Andrew Ng
- ODS community course
- classifiers: logreg, MLP, kNN, SVM, decision trees, RF, gradient boosting, naive Bayes
- dimensionality reduction, visualization: t-SNE (vis. only), PCA, UMAP
- text pre-processing: stemming / lemmatization, bag of words approach, TF-IDF
- topic modeling (SVD, LDA)
- word vector representations (word2vec, GloVe, fastText)
- sentence / document embeddings (SIF, doc2vec, StarSpace)
- advanced models: LSTM, GRU, CNN, ELMo, ULMFiT, Transformer
- python, numpy, scipy, pandas, matplotlib, seaborn, jupyter -- all the basics
- scikit-learn -- classical algorithms
- tensorflow -- neural networks and more
- gensim -- topic modeling
- nltk -- text processing
- spaCy -- advanced natural language processing