Brandon Martinez's Projects
100 data puzzles for pandas, ranging from short and simple to super tricky (60% complete)
These are daily challenges from HackerRank using a number of object-oriented programming languages. I used Python for the challenge.
This repo contains exercises for advanced topics like Cross Validation
Analysis of community college data to identify if 95% of students have a formal academic plan in place. I am to identify any areas that may be the cause of being below the goal.
This repo contains exercises for Anomaly Detection.
Practice with various ML algorithms
Config files for my GitHub profile.
This repo contains exercises for Machine Learning Classification Methods
This repo contains exercises for Machine Learning Clustering Methods.
For this project I will be working with the zillow dataset. Using the 2017 properties and predictions data for single unit / single family homes. This project is meant to incorporate clustering methodologies.
I wanted to determine whether I could develop a profile of customers who sleep separately from their significant other. The business case was to decide which types of customers mattress companies could market two or more beds to. Looking at survey responses like age group and census region, several factors seemed to have relevance to my initial goal. Unfortunately, I found that the current features did not have predictive power. The next steps would be to update the survey to mandatory responses to reduce null values, gather more quantitative data, and classify the open ended responses.
I was tasked with analyzing the data from Codeup's curriculum logs from 2018 to November 2020. I looked into which lessons were most and least accessed based on cohort and program. For Data Science, the fundamentals, regression, and classification modules were the most popular. For Web Development, it was the JavaScript, Java, and HTML lessons. I investigated user behavior to inspect for any suspicious activity finding accounts of cross access among programs, user ID's associated with multiple cohorts, and users with up to 29 IP addresses. My findings can be detailed on my GitHub page.
I am developing an algorithm to classify jobs in the data field as: Data Scientist, Data Engineer, Data Analyst, or Machine Learning Engineer. I want to create a list of common words and phrases from each job posting so that applicants like myself can utilize them in creating resumes and cover letters that stand out to potential employers!
Created common scripts/functions to use when iterating through each section of the data science pipeline.
This repo contains exercises for MySQL practice.
This package is for reading and writing operations to a database easily using a configuration file and user credentials.
This repo contains exercises for extra practice in topics like python, SQL, and statistics.
You're at work, minding your business. Suddenly, a mad scientist busts through the door and starts shouting various phrases! He hands you a USB thumb drive and expects the "next two weeks" of missing data. What could that mean? The data is from an unknown individual's FitBit device. You know very little details but you're expected to decipher that information. See more inside!
Experimental fitness app that will generate a full body workout routine with 8 exercises.
This repo contains exercises for Flask.
Practice submitting to GitHub from my local computer
These are practice problems from HackerRank involving different components of Python.
These are practice problems from HackerRank involving different components of writing SQL queries.
Automatic digit image recognition practice work. Deep Learning techniques makes it possible for object recognition in image data. This practice problem is meant to help me kick start in deep learning.
Repository cloned from https://github.com/MicrosoftDocs/ml-basics to complete the exercises from Microsoft's Ignite Azure Data Science Challenge
This repo contains exercises for Natural Language Processing
This project entailed an analysis of the readme text in various GitHub repositories. A partner and I, had the objective to build a classification model to predict the primary programming language a repo was using based on the content within the readme. We identified distinct and common words among repositories that were written in Python or Javascript. We produced a Logistic Regression model using TF-IDF that predicted with 90% accuracy on over 300 unseen repositories.
100 numpy exercises (with solutions)