Predictive Agriculture [SVE 03]

• 

  Histogram representation crop yielding countries

• 

  Representation of crop yielding items (without duplicate items)

• 

  Behaviour of Features

• 

  KDE plot of yield

• 

  Representation of crop yielding items (with duplicate items)

• 

  Box plot representation of crop yield range

• 

  Representation of crop yielding areas

• 

  Representation of crop yielding areas

Inspiration

As farmers of today facing a lots of trouble due to climatic factor and harvesting wrong type of crop, they yield very less amount of crops and result in loss of money as well as profit. To overcome this problem, we have created a model, which predicts the crop yield production by analyzing the data such as climatic factors, rainfall and temperature.

What it does

Crop yield prediction model uses machine learning to predict the future agricultural outcomes such as crop yield, live stock production and pest and disease outbreaks .These models can be used to help farmers, researches and agribusiness make better decision. It will also give the mean square error, mean absolute error and r2 scores and data visualization which helps the farmers to predict which crop and which area will give high production rate at what time. It can also be helpful for the researches to develop new agricultural technologies.

How we built it

We have used Python as a programming language and created and deployed our model in Linux One System. We have trained and deployed our model to achieve our desired goal. we Picked our desired dataset from Kaggle repository.

Challenges we ran into

Data Availability and quality. Model selection and model deployment. Model Complexity and model interpretability.

Accomplishments that we're proud of

Crop yield prediction model helps to develop a more sustainable agricultural system. Our ultimate accomplishment is to help the farmers to increase their profitability and to decrease their risk.

What we learned

The Importance of agriculture , data, model deployment.

What's next for Predictive Agriculture

More sophisticated and accurate models More widely deployed models New applications for predictive agriculture

Built With

• linuxone
• matplotlib
• numpy
• pandas
• python
• seaborn
• shap
• sklearn.compose
• sklearn.ensemble
• sklearn.linear-model
• sklearn.metrics
• sklearn.model-selection
• sklearn.neighbours
• sklearn.preprocessing
• sklearn.tree