120-data-science-interview-questions

Answers to 120 commonly asked data science interview questions.

adversarial_models

Training several deep learning models on MNIST handwritten dataset using keras. These models were tested on perturned images with different perturbations and then trained using fsgm and pgd images to improve robustness of the models.

allanah1

applied-ml

📚 Papers & tech blogs by companies sharing their work on data science & machine learning in production.

francis

fuzy_logic

This project involved being able to determine the speed and steering angle of a robot based on rules set by the programmer. These rules were dependent on 2 precedents - distance to an object and angle between the robot and the object.

hello-world

A Hello-world program

image_generation

Variational auto encoder and General Adversarial network to generate images

imdb_movies_nlp

Multiple deep learning models were trained on IMDB movie reviews using NLP and pretrained glove word embeddings and used to classify these reviews

model_for_predicting_borrower_credibility

A project involving predicting whether or not a borrower would pay back using LendingClub dataset obtained from kaggle.com

multilabel-tree-methods

Python implementation of HOMER, Parabel and CraftML algorithms for XML. Also includes EDA for XML datasets.

rnn_stock_data

Comparing LSTM, GRU, Bidirectional LSTM and Bidirectional GRU on some stock data

scipy

SciPy library main repository

short_term_load_forecasting

Short term Load Forecasting

simulated_annealing

Achieve state of the art results on traveling salesman problem using simulated annealing with a different type of stopping conditions

treeclassifiers

Decision Trees, Random Forests, Bagging from scratch using only numpy