Saksham Midha's repositories
deep-learning-v2-pytorch
Projects and exercises for the latest Deep Learning ND program https://www.udacity.com/course/deep-learning-nanodegree--nd101
myappsample
Sample app for tutorial
capsule_net_pytorch
Readable implementation of a Capsule Network as described in "Dynamic Routing Between Capsules" [Hinton et. al.]
Coursera_Capstone
This repository is a demonstration of how to create a Github repository and properly set it up.
datasharing
The Leek group guide to data sharing
phishing-URL-detection
Phishing website detection system provides strong security mechanism to detect and prevent phishing domains from reaching user. This project presents a simple and portable approach to detect spoofed webpages and solve security vulnerabilities using Machine Learning. It can be easily operated by anyone since all the major tasks are happening in the backend. The user is required to provide URL as input to the GUI and click on submit button. The output is shown as “YES” for phishing URL and “NO” for not phished URL. PYTHON DEPENDENCIES: • NumPy, Pandas, Scikit-learn: For Data cleaning, Data analysis and Data modelling. • Pickle: For exporting the model to local machine • Tkinter, Pyqt, QtDesigner: For building up the Graphical User Interface (GUI) of the software. To avoid the pain of installing independent packages and libraries of python, install Anaconda from www.anaconda.com. It is a Python data science platform which has all the ML libraries, Data analysis libraries, Jupyter Notebooks, Spyder etc. built in it which makes it easy to use and efficient. Steps to be followed for running the code of the software: • Install anaconda in the system. • gui.py : It contains the code for the GUI and is linked to other modules of the software. • Feature_extractor.py: It contains the code of Data analysis and data modelling. • Rf_model.py: It contains the trained machine learning model. • Only gui.py is to be run to execute the whole software.
twitter_corpus
The twitter sentiment corpus created by Sanders Analytics, it consists of 5513 hand-classified tweets(however, 400 tweets missing due to the scripts created by the company). Each tweet was classified with respect to one of four different topics. And a twitter account password hash file is included as well.