Wazihub Soil Moisture Prediction Challenge
$8,000 USD
Predict soil humidity using sensor data from low-cost DIY Internet of Things in Senegal
29 July–20 October 2019 23:59
709 data scientists enrolled, 96 on the leaderboard
Application of dataset
published 5 Sep 2019, 12:22

my questions are a bit much

1. so now to my own understanding, for this particular Train,csv data set you gave us, we are to creat 4new datasets from it for each humiidity and irrigation processes?

2. if we are to split the Training set into training and test set, whats gonna be our ratio?

3. must our submission data be of 5783 rows as in the sample submission?

you might not need to answer the the whole quesstion, just answer those you can in a way we can all understand.



You need to make one model that can be applied to each field individually. You will need to train the model on field 1 and implement on field 1, you will then need to use the same model but train it on field 2 and implement on field 2, and so forth for fields 3 and 4. In other words, the model type and hyperparameters remain the same, but it can be re-trained on a field-by-field basis.

You can choose your own train and test size. The csv submission we want from you is the 5783 row.

Hello, can we use the values given by the irrigation column for the tests data ?