Run Statsforecast with MLFlow.
unique_id | ds | y | static_0 | |
---|---|---|---|---|
0 | 0 | 2000-01-01 | 12.073897 | 43 |
1 | 0 | 2000-01-02 | 59.734166 | 43 |
2 | 0 | 2000-01-03 | 101.260794 | 43 |
3 | 0 | 2000-01-04 | 143.987430 | 43 |
4 | 0 | 2000-01-05 | 185.320406 | 43 |
mlflow
and mlflavors
are needed. Install them
with:
statsforecast
model can be loaded from the MLFlow registry using
the mlflow.statsforecast.load_model
function and used to generate
predictions.
ds | AutoARIMA | AutoARIMA-lo-90 | AutoARIMA-hi-90 | |
---|---|---|---|---|
unique_id | ||||
0 | 2000-02-13 | 55.894432 | 44.343880 | 67.444984 |
0 | 2000-02-14 | 97.818054 | 86.267502 | 109.368607 |
0 | 2000-02-15 | 146.745422 | 135.194870 | 158.295975 |
0 | 2000-02-16 | 188.888336 | 177.337784 | 200.438904 |
0 | 2000-02-17 | 231.493637 | 219.943085 | 243.044189 |
ds | AutoARIMA | AutoARIMA-lo-90 | AutoARIMA-hi-90 | |
---|---|---|---|---|
unique_id | ||||
0 | 2000-02-13 | 55.894432 | 44.343880 | 67.444984 |
0 | 2000-02-14 | 97.818054 | 86.267502 | 109.368607 |
0 | 2000-02-15 | 146.745422 | 135.194870 | 158.295975 |
0 | 2000-02-16 | 188.888336 | 177.337784 | 200.438904 |
0 | 2000-02-17 | 231.493637 | 219.943085 | 243.044189 |
pyfunc
flavor to a
local REST API endpoint and subsequently requesting a prediction from
the served model. To serve the model run the command below where you
substitute the run id printed during execution training code.
ds | AutoARIMA | AutoARIMA-lo-95 | AutoARIMA-lo-90 | AutoARIMA-hi-90 | AutoARIMA-hi-95 | |
---|---|---|---|---|---|---|
0 | 2000-02-13T00:00:00 | 55.894432 | 42.131100 | 44.343880 | 67.444984 | 69.657768 |
1 | 2000-02-14T00:00:00 | 97.818054 | 84.054718 | 86.267502 | 109.368607 | 111.581390 |
2 | 2000-02-15T00:00:00 | 146.745422 | 132.982086 | 135.194870 | 158.295975 | 160.508759 |
3 | 2000-02-16T00:00:00 | 188.888336 | 175.125015 | 177.337784 | 200.438904 | 202.651672 |
4 | 2000-02-17T00:00:00 | 231.493637 | 217.730301 | 219.943085 | 243.044189 | 245.256973 |