ML Part of Product Based Bangkit Capstone Project Team C22-PS316
Name | Bangkit ID | Contacts |
---|---|---|
Muhammad Syah Zichrullah Habibie | M2322F2811 | Github |
Naufaldi Hafidhigbal | M2224W2072 | Github |
We are making a crop recommendation system for recommending the top 5 crops to be grown based on the users city and environment.
Click here ( Deployed using Google Cloud Run )
Endpoint | Method | Return |
---|---|---|
/predict | GET | JSON |
Parameter | Expected input | Explanation |
---|---|---|
city | str(city name) | - |
soil | int(1, 2 or 3) | 1: Pasir, 2: Lempung, 3: Liat |
light | int(1 or 2) | 1: Full sun, 2: Semi shade |
In our machine learning model we use a deep neural network with 4 layers, 1 layer for input with input_dim is 5 which will enter 5 data, namely temperature, humidity, rainfall, soil, and light, then 2 layers for the hidden layer, and the last 1 layer for the output, in the output layer we use softmax activation so that later we can take a sequence of 5 predictions with the highest accuracy. Here we can see after we train the model we get 33% loss and 89% accuracy, from this training result as you can see highest accuracy prediction and actual data are the same and then it will be followed up to second-highest accuracy to the fifth-highest accuracy
- Clone this repo
- Open terminal and go to this project's root directory
- Create your own OpenWeatherMap API_KEY (*Notes)
- Create a config.py file and type
api_key="your_api_key"
then save - Create and activate a python virtual environment if you want
- Type
pip install -r requirements.txt
- Serve the Flask app by typing
flask run
- It will run on
http://127.0.0.1:5000