Practical Machine Learning with Python: A Problem-Solver's Guide to Building Real-World Intelligent Systems (Paperback)

PART I - Understanding Machine Learning Chapter 1: Machine Learning Basics Chapter Goal: This chapter familiarizes and acquaints readers with the basics of machine learning, industry standard workflows followed for machine learning processes and expands on the different types of machine learning and deep learning algorithms No of pages: 50-60 Sub -Topics 1. Brief on machine learning, definitions and concepts 2. Industry standard for data mining processes - CRISP - DM and adoption in ML 3. Brief on data processing, visualization, feature extraction\engineering concepts 4. Types of learning algorithms - supervised, unsupervised, reinforcement learning 5. Advanced models - time series, deep learning 6. Model building and validation concepts 7. Applications of machine learning Chapter 2: The Python Machine Learning Ecosystem Chapter Go al: This chapter introduces readers to the python language and the entire ecosystem built around machine learning with python tools, frameworks and libraries. Overview and code samples are given for each tool to depict its usage and effectiveness No of pages: 50 - 60 Sub - Topics 1. Brief on Python 2. Why is Python effective for machine learning and data science 3. Brief overview on the python ecosystem followed by data scientists (includes anaconda distribution) 4. Reproducible research with ipython 5. Data processing and computing with pandas, numpy, scipy 6. Statistical learning with statsmodels 7. ML frameworks - scikit-learn, pyml etc 8. NLP frameworks - nltk, pattern, spacy 9. DL frameworks - theano, tensorflow, keras PART II - The Machine Learning Pipeline Chapter 3: Processing, wrangling and visualizing data& amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;Sub - Topics: 1. Data Retrieval mechanisms (crawling, databases, APIs etc) 2. Data processing (handling various forms of data - SQL, JSON, XML, Images) 3. Data attributes and features (numeric, categorical etc) 4. Data Wrangling (cleaning, handling missing values, normalizing data) 5. Data Summarization 6. Data Visualization (bar, histogram, boxplot, line, scatter etc) Chapter 4: Feature Engineering and Selection Chapter Goal: T his chapter focuses on the next stage in the ML pipeline, feature extraction, engineering and selection. Readers will learn about both basic and advanced feature engineering methods for different data formats including numeric, text and images. We will also focus on methods for effective feature selection No of pages: 50 - 60 Sub - Topics: 1. Features - understanding your v2. Basic Feature engineering 3. Extracting features from numeric, categorical variables 4. Extracting features from date imestamp variables 5. Extracting Basic features from textual data (bag of words) 6. Advanced Feature engineering 7. Extracting complex features from textual data (word vectorization, tfidf, topic models) 8. Extracting features from images (pixels, edge detection, shapes) 9. Time series features 10. Feature scaling and standardization 11 Feature se lection techniques 12 Using forwardackward selection techniques 13 Using machine learning models like random forests 14 Other methods Chapter 5: Building, tuning and deploying models Chapter Goal: This chapter focuses on the final stage in the ML pipeline where readers will learn how to fit and build models on data features, how to optimize and tun e models and f learn ways of deploying models to use them in real-world scenarios for predictions\insights No of pages : 50-60 Sub - Topics: 1. Fitting and building models 2. Model evaluation techniques 3. Model optimization methods like gradient descent 4. Model tuning methodologies like cross validation, grid search 5. How to save and load models 6. Deploying models in action PART III - Real-world case studies in applied machine learning Chapt er 6: Analyzing bike sharing trends Chapter Goal: This chapter will focus on a real-world case study of analyzing and predicting bike sharing trends with a focus on regression models No