Masterclass in ChatGPT, Advanced Analytics & Python Projects

Explore basic prompts for large language models like ChatGPT and Bard, covering summarization, information extraction, question answering, text classification, conversation, code generation, and reasoning to build a practical baseline.

Frame a conversation with an AI research assistant by defining context and a technical tone. Then compare model responses to prompts on topics like black holes.

Explore how large language models exhibit critical reasoning, common errors, and prompt strategies like step-by-step solving to improve accuracy in reasoning tasks.

Break big projects into smaller tasks and prompts, starting simple and iterating to refine context, audience, and output formats, using versioned prompts, then develop chapters and subtopics step by step.

Master precision in prompts by being specific and direct, treating the model like a colleague; define length (two to three sentences) and audience to guide prompt engineering.

Explore zero-shot and few-shot prompting, with single-shot, five-shot, or ten-shot examples, and compare these techniques to fine tuning; learn how demonstrations guide model reasoning and reveal limits.

Explore editing with ChatGPT using the Python agent to crop and resize, rotate 180 degrees, adjust brightness and color to black and white, and save in various formats.

Edit images with ChatGPT code interpreter by cropping and resizing to 180 by 920, rotating 180 degrees, adjusting brightness and colors to black and white, and converting png to jpeg.

Learn to upload documents to ChatGPT, generate holistic summaries and insights, create visualizations, extract outliers, and craft SEO-friendly blog posts and analyst reports from text data.

Plot NVIDIA stock historical data using adjusted close and volume on dual axes, then show a $1 investment growth to illustrate 60-70% returns.

Explore the relationship between trading volume and price using correlation analysis, revealing a weak negative correlation and the value of lag-based tests with Nvidia data.

Explore correlation analysis across multiple stocks by calculating daily returns from historical prices and visualizing a correlation heat map to assess diversification and inter-stock linkage from a Kaggle stock dataset.

Upload a large zip dataset, inspect its folder structure (data, etfs, stocks), and prepare to extract data by correcting file formats from csv to txt for tickers like aapl.us.txt.

Explore stock price data through an end-to-end EDA workflow: extract from txt files, retain date and close, plot time series. Compute daily returns and visualize the correlation.

Explore how to compute stock return correlations, build a correlation heat map, assess portfolio diversification, and compare equal-weighted portfolio returns for 2017 across tech and non-tech stocks.

Learn to create a portfolio analysis report with an equally weighted stock portfolio, using diversification, daily returns, and a correlation matrix, then present visuals to boost investors' credibility online.

Explore how 50-day and 90-day simple moving averages smooth price data to reveal trends and drive a moving average crossover trading strategy.

Perform sentiment analysis on a McDonald's reviews dataset to identify the highest rated US stores by city and state, with visualizations and a Python-based word cloud.

Combine all reviews into a single text, lowercase and remove punctuation and stop words, then generate a word map and word cloud to visualize the most frequent terms.