Generative AI Masters 2026 - From Python to Gen AI

Explore map, reduce, and filter in Python, using lambda functions to write concise, loop-free code and see practical examples like circle area and Celsius-to-Fahrenheit conversions.

Learn Python file handling by opening, reading, writing, and closing files, using test.txt examples, and exploring append and read line; briefly preview pandas read_csv for data science workflows.

Explore Python control structures with binary and relational operators and boolean logic. Apply if-else decision making to tasks like even/odd checks, age filters, and range and list comprehensions.

Master pandas in Python to read CSV data, explore data frames with head, tail, info, and describe, and perform sorting, selection, column creation, filtering, null handling, and joins.

Master numpy to perform fast, array-based numerical computation with 1d, 2d, and nd arrays, explore dtypes, shapes, indexing, and operations like zeros, ones, and random.

Explore data visualization in Python to turn numbers into visual insights, using matplotlib and seaborn to identify patterns, correlations, and trends with bar, line, area, and pie charts.

Explore matplotlib for Python data visualization, learning to plot histograms, bar charts, area charts, and pies using NumPy and pandas alongside seaborn in future lessons.

Discover seaborn, a data visualization library that makes hard plotting easier than matplotlib, with iris data, installation steps, and kernel density plots, distribution plots, pair plots, and heat maps.

Explore natural language processing foundations, from linguistic concepts to text preprocessing and tokenization, and see how NLP powers generative AI, LLMs, chatbots, and translation.

Trace the history and scope of natural language processing within AI, and highlight applications like language translation, sentiment analysis, and chatbots powered by large language models.

Explore NLP key challenges, including ambiguity types — pragmatic, lexical, syntactic, anaphoric — along with lack of standardization, ethical considerations, context understanding, and data sparsity.

Explore the four core topics of linguistics: phonetics, morphology, syntax, and semantics, and how they underpin NLP applications like speech recognition and machine translation.

Explore case folding as an NLP normalization that converts text to lowercase to improve generalization and reduce vocabulary size, with Python examples using lower and casefold, noting proper nouns.

Learn how special character removal, a crucial NLP pre-processing step, reduces noise and improves tokenization using re, spaCy, and NLTK for text used in generative AI.

Explore handling contractions in NLP preprocessing by expanding contractions with a Python library and regex, and understand the role of stopword and special character removal before model training.

Explore tokenization as a fundamental NLP step that breaks text into sentences and words, using sentence tokenizer, word tokenizer, and other tokenizers such as subword and character tokenizers.

Learn how stop words—common non-informative words—are identified and removed in NLP to improve signal-to-noise ratio, using Python with NLTK stopwords and word tokenization (and similar tools such as Spacy).

Explore n-grams, including unigrams, bigrams, and trigrams, and learn how language models predict the next word for tasks like text classification, spelling correction, and text summarization.

Convert textual data into numerical vectors to enable machine learning algorithms to process language, using bag of words, tf-idf, and embeddings like word2vec, glove, and fasttext.

Learn bag of words as a foundational NLP vectorization method, converting text to word-frequency features while noting its limitations and its role before more advanced techniques.

Apply bag of words practicals to build a spam detector from a ham vs. spam dataset, using count vectorizer and Multinomial Naive Bayes, with train/test split and evaluation.

Explore tf-idf, a term weighting method that uses term frequency and inverse document frequency to highlight significant words while downplaying common terms across documents.

Explore part of speech tagging as a pre-processing NLP step, apply it to named entity recognition, Q&A, and chatbots, and practice using spaCy while studying hidden Markov and Viterbi.

Practice named entity recognition with spaCy by identifying entities and labels in sentences, and customizing entities using doc ents with examples like Tesla as organization and money as value.

Explore word embeddings and the word2vec model, compare bag-of-words with one-hot encoding, and learn about cbow and skip-gram approaches and neural networks for word representations.

Explain word embeddings in two main types—count-based and prediction-based—covering bag-of-words, TF-IDF, glove, and word2vec with continuous bag of words and skip-gram, plus pre-trained models like Google's and ChatGPT.

Explore pretrained word2vec vectors from Google News, using Gensim to examine 300-dimensional word representations, compute similarities, and perform analogies like king minus man plus woman.

Discover the intuition behind word2vec by building a simple feature-based vector model, illustrating king minus man plus woman yielding queen, and comparing handcrafted vs automated neural features.

Explore the continuous bag of words model in word two vec, showing how context words predict a target word via a neural network with window size and backpropagation.

Explore the word2vec skip-gram model, predicting context from a target. Compare it to cbow and learn the neural network setup with input, hidden, and output layers, softmax and backpropagation.

Glove derives word embeddings from a word-to-word co-occurrence matrix, using matrix factorization to produce word vectors. Apply pre-processing, build the co-occurrence matrix, and explore word similarities like cosine similarity.

Explore fasttext as an upgraded word embeddings method and compare it with word2vec, highlighting character and gram-level units and the out of vocabulary problem.

Explore cosine similarity as the angle between non-zero vectors, using dot products and magnitudes to measure sentence or word similarity in NLP, with bag-of-words examples and practical use cases.