Skip to content

Latest commit

 

History

History
71 lines (46 loc) · 3.32 KB

README.md

File metadata and controls

71 lines (46 loc) · 3.32 KB

Dogeship

A project to predict the price of Doge cryptocurrency of any date based on previous data.

A Machine Learning model with an architecture including Linear regression and the model was trained on it.

The project website can be found here. Doge price gets produced when the user inputs the parameters and clicks the Predict button. The website was built using HTML, CSS and JS along with an API request and response which deployed in heroku. On pressing the Predict button, an API GET request gets called which returns the price. The source code regarding the api and website can be found in the api and frontend folders respectively.

Team members

  1. Haleel sada
  2. Nanda Kishor M Pai

Team Id

ML / 2

Link to product walkthrough

Product Walkthrough

How it Works ?

  1. Go to website and give date and other parameters to predict the Doge price of next day
  2. On pressing the Predict button, an API GET request gets called which returns the price
  3. the api uses Linear regression algorithm on previous data to predict the price of Dogecoin
  4. The website was built using HTML, CSS and JS along with an API request and response which deployed in heroku

Libraries used

pandas - 1.3.5

numpy - 1.22.3

joblib - 1.1.0

matplotlib - 3.2.2

scikit-learn - 1.0.2

flask - 1.1.4

How to configure

Inorder to train the model, load the python jupyter notebook found here in a Google Colab and make a copy of it for use. To predict the price use this website.

The source code regarding the api and website can be found in the api and frontend directories respectively.

API has been built on this Regression Model. URL could be found here here

User has to send a POST request to the given api with a dictionary of Input Features.

import requests

columns = ['Open', 'High', 'Low', 'Close', 'Volume', '7day_open', '7day_close',
               '7day_high', '7day_low', '40day_open', '40day_close', '40day_high', '40day_low']
# Sample Inputs
arr = [0.000293, 0.000299, 0.00026, 0.000268, 1463600, 0.000287714285714,
           0.000290571428571, 0.000325, 0.00026, 0.000300025, 0.000298775, 0.000467, 0.000223]
dic = {columns[i]: arr[i] for i in range(len(arr))}

url="https://dogecoin-prediction-bot.herokuapp.com/"
r = requests.post(url,json = dic)
print(r.text)

Output

'{"close":0.0003021457613296437}\n'

How to Run

The project website could be run from here