ner-aim/Credit-Card-Approval-Prediction ner-aim/Credit-Card-Approval-Prediction

  • Stars
    star
    1
  • Language
    Jupyter Notebook
  • Created almost 3 years ago
  • Updated almost 3 years ago
Twitter logo Share LinkedIn logo Share Facebook logo Share
ner-aim/Credit-Card-Approval-Prediction
github

ner-aim

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

More Repositories

1

TwinMemories-Prototype

C#
3
star
2

ner-aim.github.io

Visit my portfolio: https://ner-aim.github.io
CSS
2
star
3

HR-Analytics-Dashboard

2
star
4

Facial-Expression-Detection

Jupyter Notebook
2
star
5

A-Tile-Game

A platformer game.
C#
2
star
6

Super-Sparty-Bros

This game is the part of a course and a clone to Super Mario Bros.
C#
2
star
7

Reinforcement-Learning-using-Unity

Reinforcement learning with Unity
ASP.NET
2
star
8

Capital-One-Cards-Fraud-Detection-MLOps

Jupyter Notebook
2
star
9

Course-Projects

C#
1
star
10

academic-kickstart

Shell
1
star
11

Retreving-Financial-Estimates-using-Estimize

This project revolves around getting the financial estimates for a few companies of the financial year 2019
Jupyter Notebook
1
star
12

Capital-One-Credit-Card-Fraud-Predicition

Jupyter Notebook
1
star
13

Color-Game

A basic color game where you have to guess the color of the given RGB composition.
JavaScript
1
star
14

TicTacToe

Just a simple TicTacToe game made using python tkinter
Python
1
star
15

US-Prevailing-Wages

Operated with a team of 5 and preprocessed data using Pandas by clearing problematic rows and brought consistency and leveraged 10 common features Generated visualizations using Plotly for gathering hidden insights of wage growth, geographic, NAICS, job experience
Jupyter Notebook
1
star
16

Predicting-seasurface-temperatures

Comparing all 3 models, using the Residual Mean Squared Error as our metric, we see that the BSTS model performs the best with the lowest RMSE score. We suspect that the Bayesian approach worked so well because it can infer causal impacts within time series. In this case, whether 1 or more layers greatly influenced the temperatures of the layers beneath it. In our coefficient analysis for a model size of 1 we confirmed this was the case. We faced a few issues in this project. For the SARIMA models, we started out with a trial and error method to determine the order of the seasonal model required to fit the data. This method proved difficult since there are numerous combinations of numbers that could have made the model. Reading the ACF and PACF model and using that as a guide resulted in producing the best model we have provided above. Another issue we ran into was trying to configure the predict method to work with our data. The predict method produced predictions for the entire dataset instead of just future data. Using the one step ahead prediction and the forecast function was able to rectify this error.
Jupyter Notebook
1
star