This repository was updated continously for a period of 31 days. With the aim to learn about machine learning. This is created during Coronavirus Quaratine time. Target is to do learn something daily and commit it in this repository.
Day1: Setting up Github and Python libraries.
Day2: Stochastic gradient descent implementation
Day3: Normalization the dataset.(Normalization before or after the test_train split). Normalization(Min-Max Scaler) and Standardization
Day4: Ensemble Learning, xgboost
Day5: Hackathon1 xgboost model
Day6: LGBM, Hackathon1 lgbm model
Day7: Hyper-parameter optimization.
Day8: Catboost model
Day9: Classification model
Day10: Model Validation Techniques
Day11: L1 and L2 regilarization techniques- Used to deal with over-fitting by introducing penalty. L1-lasso, L2-ridge
Day12: Handling missing values
Day13: Macro, Weighted and micro F1 score
Day14: Structure of uber, netflix
Day15: Web scrapping.
Day16: Pre-processing-Removing special char, Expanding contractions
Day17: Stemming and Lemmatization
Day18: Stop words
Day19: Part of Speech (POS)
Day20: PCA- Dimentionality Reduction
Day21: Gini impurity
Day22: Name Entity Recognition (NER)
Day23: Word Embedding: Representing word in vector format. Two type frequency and prediction based. Frequency type includes Count vector, TF-IDF, etc. Prediction based include Continous bag of words (CBOW), skip-gram.
Day24: TF-IDF
Day25: Softmax
Day26: Chatbot Day1
Day27: Sentiment analysis using TextBlob
Day28: Voice Enabling
Day29: Cosine similarity
Day30: Spell checker
Day31: Chatbot day2 (https://github.com/manzoormahmood/Tourist-Guide-Chatbot)