This repository contains data to validate the next milk predictions from the Veerkracht VLAIO project. Please request write access to the repository via [email protected] If extra information is needed, this can be requested.
In the './Data/' folder a dataset of individual cow milk recording events of 100 farms has been collected.
The dataset is partitioned into 10 partitioned csvs (UTF-8 charset) each containing milk recordings
- Timestamp format is 'yyyy-MM-dd'T'HH:mm:ss[.SSS][XXX]'
- Line seperator is '\n'
- Seperator is ','
Eah row of the dataset contains:
- HerdIdentifier - Unique identification number of the farm - string
- AnimalIdentifier - Unique identification number of the animal - string
- BirthDate - Birth date of the animal - timestamp
- CalvingDate - Calving date of the lactation - timestamp
- LactationNumber - Lactation number of the lactation - integer
- TestDate - Date of milk recording - timestamp
- MilkYieldKg - Milk yield of the milk recording - double
- DaysInMilk - Days in milk at the milk recording - integer
- FatPercentage - Fat content (%) of the milk recording - double
- ProteinPercentage - Protein content (%) of the milk recording - double
Please create a folder in the './Predictions/' with your own milk yield predictions using the same unique identifiers
The predictions for the first milk yield prediction should have the following format
- HerdIdentifier - Unique identification number of the farm - string
- AnimalIdentifier - Unique identification number of the animal - string
- CalvingDate - Calving date of the lactation for which the first yield is pedicted - timestamp
- LactationNumber - Lactation number of the lactation - integer
- TestDate - Date of milk recording - timestamp
- MilkYieldKg - Milk yield (kg) of the milk recording - double
- DaysInMilk - Days in milk at the milk recording - integer
- PredictedMilkYieldKg - Predicted milk yield (kg) - double