sandrabotica Goto Github PK

belly-button-challenge

Module 14 Challenge involves building a dashboard to explore an external site with a microbe dataset. Horizontal bar chart, dropdown menu, bubble chart and metadata table of sample data. App deployed to a static hosting page.

credit-risk-classification

Module 20 Challenge uses Supervised Machine Learning to predict loan risk of borrowers. Train-test-split, Logistic Regression Model, Confusion Matrix and Classification Report analysis.

crowdfunding_etl

Module 13 Challenge is a ETL pipeline using Python dictionary methods and regular expressions to extract and transform data. CSVs used to create an ERD table schema and uploaded to a Postgres database.

cryptoclustering

Module 19 Challenge uses unsupervised machine Learning to predict if Cryptopcurrencies are affected by 24hour or 7day price changes. Normalise data, K-Means Clustering, elbow method and PCA analysis.

deep-learning-challenge

Module 21 Challenge uses a Neural Network Model as a binary classifier to predict if a funding organisation will be successful in their application for funds based on features in a dataset. Compile, Train, Evaluate, Optimise the model.

home_sales

Module 22 Challenge, Big Data, use of SparkSQL to determine key metrics about home sales data. Temporary views, partitioned data, cached and uncached a temporary table, and verifying table as uncached.

leaflet-challenge

Module 15 Challenge uses Leaflet to visualise an earthquake dataset from USGS. TileLayer loads and connects to a geojson API using D3 and tooltip for each data point, scaled with magnitude and colour based on depth.

ml_project_4

Supervised ML model implementation and optimisation for heart disease prediction. Logistic Regression, Random Forest and a Neural Network. Tableau visualisation of data.

module-18-challenge

Module 18 Challenge involves aggregating data from an external site, the CitiBikeNY History log to present 2 unexpected phenomena in a Tableau Dashboard. Metrics included, analytic rigor, readability and visual appeal.

module11-challenge

Module 11 Challenge involves web-scraping and data analysis by identifying HTML elements (id/classes attributes) on a page. Extract info with Splinter and HTML parsing using Beautiful Soup.

module5-matplotlib

Module 5 Challenge using Matplotlib, reporting clinical study results. Prepare data, generate summary statistics, create bar/pie/line/scatter/box charts/plots, calculate quartiles, correlation, regression and identify outliers.

nosql-challenge

Module 12 Challenge uses MongoDB to connect and update database collections on restaurant ratings. Using a Jupyter notebook to perform queries to answer questions about restaurants.

ny-city-bikes

nytimes-scraping

pandas-challenge

Module 4 Challenge using Pandas and Jupyter Notebook to aggregate trends in school performance to create a report on observable trends based on the data.

project-3

A data story on global coffee trade. Dashboard of user-driven visualisations using Plotly. Python Flask API created which includes HTML/CSS, JavaScript and a database using ERD and PostgresSQL.

python-api-challenge

Module 6 Challenge using Python requests, APIs (OpenWeatherMap & Geoapify) and JSON traversals. Scatter plots and linear regression showcase weather variables and latitude relationships. Mapping hotels within a certain radius of ideal weather conditions.

python-challenge

Module 3 Challenge using Python scripts to analyse financial records and vote-counts from polling data. Import/read/write csv files, tuples, lists, dictionaries, iterating through data and debugging.

sql-challenge

Module 9 Challenge involves data modelling, engineering and analysis. Use of QuickDBD to sketch an ERD and table schemas, data type, primary and foreign key relationships. SQL queries for data analysis.

sqlalchemy-challenge

Module 10 Challenge is a climate analysis of a vacation destination. SQLAlchemy ORM queries, Pandas and Matplotlib used for climate data exploration. Precipitation and Stations data converted to json and linked to Flask API.

vba-challenge

Module 2 Challenge using VBA scripting to analyse generated stock market data. Looping through data and reading/storing specific values, creating columns, conditional formatting and calculations.