Lending Club Loans: Brief Introduction LendingClub is a US peer-to-peer lending company, headquartered in San Francisco, California.[3] It was the first peer-to-peer lender to register its offerings as securities with the Securities and Exchange Commission (SEC), and to offer loan trading on a secondary market. LendingClub is the world's largest peer-to-peer lending platform. Objective Building a model that predicts whether the borrower can payback the loan or not, so in the future we can assess the customer and whether or not he's likely to payback his loan. Main Strategy Our main objective is not lending a person that is not going to payback his loan which would be a Type 1 error, Therefore we must depend on the recall score of the loans not payed category by doing methods that might reduce our accuracy but ultimately increasing our recall. Step 1: Exploratory Data Analysis and Feature Engineering Getting a general idea of datatypes and null values for each column Using Seaborn to visualize the data by plotting charts We see that the labels are imbalanced with a ratio of 4:1 (fully_paid, charged_off) Imbalance between Fully Paid and being Charged Off can negatively affect our model that tries to have a high recall score for the Charged Off label, a poor recall score could be achieved when overfitting to the Fully Paid portion We downsample the fully paid portion to the size of the charged off portion. Removing outliers that may result in misleading interpretations. Handling nan values and transforming strings to numeric data-types Dropping columns if they won't be used in our model or be used in feature engineering. Extracting the zip-code from the address column. Encoding columns and getting dummy variables. Filling na values with the mean for values that don't have a high correlation with the loan status. Using Random Forest Regressor to predict the missing values in the mort_acc column as it is highly correlated to the loan_status. Step 2: Building the model Splitting the data into train and test data 80, 20 split Taking the test data and upsampling the fully paid portion to get a realistic summary of the metrics Using a sequential model for our ANN model Building the model to have 4 layers and an activation of rectified linear unit(except the last layer which is sigmoid), a Dropout of 0.2 and building it for binary classification using the Adam optimizer. Saving the model and checking the losses for the model. Checking the predictions. Conclusion: We get a well rounded classification report, and we get a recall score of 0.81 and an accuracy score of 0.80, we can further tune our model and get better recall score for charging off for example but that may affect our overall accuracy and that depends on how we want our model to perform.
Lending-Club-Loans - Using TensorFlow to create an ANN model to predict whether people would charge off or pay back their loans.
Overview
Decode the Ontario proof of vaccination QR code
Decode the contents of the Ontario Proof of Vaccination (the "Smart Health Card QR Code") Output This is from my QR code, hopefully fully redacted alt
Freqtrade 3commas wrapper for python
Freqtrade 3commas wrapper The aim of this project is to provide an easy way to integrate freqtrade with 3commas. The main reason someone would want to
This Is Advanced Version Of Old Radio Player, An Telegram Bot to Play Radio/Music in Channel or Group Voice Chats.
Telegram Radio Player V2 An Telegram Bot to Play Radio/Music in Channel or Group Voice Chats. This is also the source code of the bot which is being u
A python script to send sms anonymously with SMS Gateway API. Works on command line terminal.
incognito-sms-sender A python script to send sms anonymously with SMS Gateway API. Works on command line terminal. Download and run script Go to API S
This project checks the weather in the next 12 hours and sends an SMS to your phone number if it's going to rain to remind you to take your umbrella.
RainAlert-Request-Twilio This project checks the weather in the next 12 hours and sends an SMS to your phone number if it's going to rain to remind yo
An Simple Advance Auto Filter Bot Complete Rewritten Version Of Adv-Filter-Bot
Adv Auto Filter Bot V2 This Is Just An Simple Advance Auto Filter Bot Complete Rewritten Version Of Adv-Filter-Bot.. Just Sent Any Text As Query It Wi
A python based all-in-one tool for Google Drive
gdrive-tools A python based all-in-one tool for Google Drive Uses For Gdrive-Tools ✓ generate SA ✓ Add the SA and Add them to TD automatically ✓ Gener
Discord Token Generator - Python (Generates Tokens and Joins your Server Automatically) hCaptcha Bypass **FREE**
Best Discord Token Generator {hCaptcha bypass FREE Unlimited Memberboost} Install few requirements & run main.py it will redirect you to the Download
Policy and data administration, distribution, and real-time updates on top of Open Policy Agent
⚡ OPAL ⚡ Open Policy Administration Layer OPAL is an administration layer for Open Policy Agent (OPA), detecting changes to both policy and policy dat
Telegram bot that search for the classrooms status of the chosen day and then return all the free classrooms using your preferred time slot
Aule Libere Polimi Since the PoliMi site no longer allows people to search for free classrooms this bot was necessary! It simply search for the classr
An open source development framework to help you build data workflows and modern data architecture on AWS.
AWS DataOps Development Kit (DDK) The AWS DataOps Development Kit is an open source development framework for customers that build data workflows and
PyFIR - Python implementations of Finite Impulse Response (FIR) filters
pyFIR Python implementations of Finite Impulse Response (FIR) filters. The algorithms are mainly (but not strictly) the ones described in WEFERS, Fran
Bulk NFT uploader to OpenSea!
Bulk NFT Uploader Description Simple easy peasy python script which logins to opensea account using metamask and bulk uploads NFT to your default coll
Small cloudfoundry client implemented in python
Cloudfoundry python client The cf-python-client repo contains a Python client library for Cloud Foundry. Installing Supported versions warning: Starti
Zen-Userbot - Userbot gabut With Python
Zen-Userbot Disclaimer ⚠️ PERINGATAN UNTUK ANDA ⚠️ ️ Zen-Userbot
Disqus API bindings for Python
disqus-python Let's start with installing the API: pip install disqus-python Use the API by instantiating it, and then calling the method through dott
A Telegram Bot for adding Footer caption beside main caption of Telegram Channel Messages.
Footer-Bot A Telegram Bot for adding Footer caption beside main caption of Telegram Channel Messages. Best for Telegram Movie Channels. Made by @AbirH
Invites link generator for telegram(made for channel referral links)
InviteLinkGen Invites link generator for telegram(for channel referral links) made for @HelakuruEsana channel Spotify Giveaway
Want to get your driver's license? Can't get a appointment because of COVID? Well I got a solution for you.
NJDMV-appoitment-alert Want to get your driver's license? Can't get a appointment because of COVID? Well I got a solution for you. We'll get you one i
This is a discord bot, which tells you food recipes.
Discord Chef Bot You have a friend, familiy or other group / channel where the topic is the food? You cannot really decide what's for Saturday lunch?