Welcome to PYCS!

)(

(Predict Your CropS, pronounced like pisces)

Crop Yield Prediction

And What-If Tool

PYCS is a What-If tool. Submit your data in .csv format, and PYCS will give you its best prediction of crop yields (alfalfa) given your data. PYCS uses machine learning models to forecast crop yields, so this tool can help you plan for shortages or surpluses based on factors like rain, temperature, solar radiation, and more. What if reality ends up closely resembling the details you provide? What if historical trends continue into the future? These are the questions PYCS intends to answer! (Please allow 1 week for results.)

Annualized Mode: Off

File Format: .csv, minimum 2 columns - (1) date_of_cut (mm/dd/yyyy) and (2) yield (tons/acre).

File Format: .csv, minimum 2 columns - (1) year_of_cut (yyyy) and (2) yield (tons/acre).

Your email address:

URL to your .csv dataset: (on GitHub, Google Drive, etc.)

Your location: (City, State)

Name this dataset: (ex.: what_if_low_rain, actual_radiation, etc.)

Features You Are Providing:

Yield

Date of cut

Year

Rain

Minimum Average Temperature

Maximum Average Temperature

Accummulated solar radiation

Date sown

Variety

# Previous Cuts

# Previous Years

Prediction Accuracy:

3 classes (hi, md, lo)

Nearest ton

Nearest tenth ton

Nearest hundredth ton

Train/Test:

Split submitted dataset into train & test

Test on submitted dataset, train on (select):

Train on submitted dataset, test on dataset in other submission

Test on submitted dataset, train on dataset in other submission

Increase training dataset size with data synthesis

# synth samples (<= 500K)

Notes: