... | @@ -4,7 +4,7 @@ In this example we want to predict TD Principal of the [Apache Kafka](https://gi |
... | @@ -4,7 +4,7 @@ In this example we want to predict TD Principal of the [Apache Kafka](https://gi |
|
|
|
|
|
Once the server is running, open your web browser and navigate to the following URL:
|
|
Once the server is running, open your web browser and navigate to the following URL:
|
|
|
|
|
|
http://127.0.0.1:5000/ForecasterToolbox/TDForecasting?horizon=10&project=apache_kafka_measures®ressor=auto&ground_truth=yes&test=no
|
|
http://127.0.0.1:5000/ForecasterToolbox/TDForecasting?horizon=10&project=apache_kafka_measures®ressor=auto&ground_truth=no&test=no
|
|
|
|
|
|
The response of the submitted command is a JSON object containing the forecasted TD values of the Apache Kafka test project for versions 151 to 160 (10 steps) ahead, using the Ridge regressor, which the back-end selected as the most appropriate model for this occasion. This JSON object is illustrated below:
|
|
The response of the submitted command is a JSON object containing the forecasted TD values of the Apache Kafka test project for versions 151 to 160 (10 steps) ahead, using the Ridge regressor, which the back-end selected as the most appropriate model for this occasion. This JSON object is illustrated below:
|
|
|
|
|
... | @@ -64,3 +64,50 @@ The response of the submitted command is a JSON object containing the forecasted |
... | @@ -64,3 +64,50 @@ The response of the submitted command is a JSON object containing the forecasted |
|
"status": 200
|
|
"status": 200
|
|
}
|
|
}
|
|
```
|
|
```
|
|
|
|
|
|
|
|
## Energy Forecasting Example
|
|
|
|
|
|
|
|
In this example we want to predict Energy Consumption of the [Sbeamer Gapbs](https://github.com/sbeamer/gapbs) test project for 5 steps ahead. Sbeamer Gapbs dataset contains 60 versions, so we expect the response to contain predicted TD values for versions 61 to 65. Again, we set the *horizon* parameter to *5* and the *regressor* parameter to *auto*, in order to allow the Energy Forecaster to choose the best model based on training error minimization.
|
|
|
|
|
|
|
|
Once the server is running, open your web browser and navigate to the following URL:
|
|
|
|
|
|
|
|
http://127.0.0.1:5000/ForecasterToolbox/EnergyForecasting?horizon=5&project=sbeamer_gapbs_energy_measures®ressor=auto&ground_truth=no&test=no
|
|
|
|
|
|
|
|
The response of the submitted command is a JSON object containing the forecasted Energy values of the Sbeamer Gapbs test project for versions 61 to 65 (5 steps) ahead, using the Lasso regressor, which the back-end selected as the most appropriate model for this occasion. This JSON object is illustrated below:
|
|
|
|
|
|
|
|
```json
|
|
|
|
{
|
|
|
|
"message": "The request was fulfilled.",
|
|
|
|
"results": {
|
|
|
|
"forecasts": [{
|
|
|
|
"value": 51.089354864448154,
|
|
|
|
"version": 61
|
|
|
|
},
|
|
|
|
{
|
|
|
|
"value": 50.32746962241154,
|
|
|
|
"version": 62
|
|
|
|
},
|
|
|
|
{
|
|
|
|
"value": 51.481911189167526,
|
|
|
|
"version": 63
|
|
|
|
},
|
|
|
|
{
|
|
|
|
"value": 51.3345537282448,
|
|
|
|
"version": 64
|
|
|
|
},
|
|
|
|
{
|
|
|
|
"value": 51.490599251851116,
|
|
|
|
"version": 65
|
|
|
|
}
|
|
|
|
],
|
|
|
|
"parameters": {
|
|
|
|
"ground_truth": "no",
|
|
|
|
"horizon": 5,
|
|
|
|
"project": "sbeamer_gapbs_energy_measures",
|
|
|
|
"regressor": "lasso",
|
|
|
|
"test": "no"
|
|
|
|
}
|
|
|
|
},
|
|
|
|
"status": 200
|
|
|
|
}
|
|
|
|
``` |
|
|
|
\ No newline at end of file |