Request Prediction Intervals
This page describes how to request prediction intervals when scoring.
Example Usage for Data Frames
import h2o_mlops_scoring_client
import pandas
DATA_FRAME = pandas.read_csv("/Users/jgranados/datasets/BNPParibas.csv")
ID_COLUMN = "ID"
MLOPS_ENDPOINT_URL = "https://model.internal.dedicated.h2o.ai/aa7a5860-cfb5-4c5a-948c-85a81a061c0d/model/score"
h2o_mlops_scoring_client.score_data_frame(
mlops_endpoint_url=MLOPS_ENDPOINT_URL,
data_frame=DATA_FRAME,
id_column=ID_COLUMN,
request_prediction_intervals=True
)
24/02/27 16:45:14 INFO h2o_mlops_scoring_client: Connecting to H2O.ai MLOps scorer at 'https://model.internal.dedicated.h2o.ai/aa7a5860-cfb5-4c5a-948c-85a81a061c0d/model/score'
24/02/27 16:45:15 INFO h2o_mlops_scoring_client: Starting scoring data frame
24/02/27 16:45:47 INFO h2o_mlops_scoring_client: Scoring complete
24/02/27 16:45:47 INFO h2o_mlops_scoring_client: Total run time: 0:00:33
24/02/27 16:45:47 INFO h2o_mlops_scoring_client: Scoring run time: 0:00:32
| ID | target | target.lower | target.upper | |
|---|---|---|---|---|
| 0 | 3 | 0.861840 | 0.068284 | 1.293299 |
| 1 | 4 | 0.701454 | -0.092101 | 1.132914 |
| 2 | 5 | 0.794855 | 0.001299 | 1.226315 |
| 3 | 6 | 0.950734 | 0.157178 | 1.382193 |
| 4 | 8 | 0.956176 | 0.162620 | 1.387635 |
| ... | ... | ... | ... | ... |
| 114316 | 228708 | 0.913621 | 0.120065 | 1.345081 |
| 114317 | 228710 | 0.935338 | 0.141782 | 1.366797 |
| 114318 | 228711 | 0.969223 | 0.175667 | 1.400683 |
| 114319 | 228712 | 0.767552 | -0.026004 | 1.199012 |
| 114320 | 228713 | 0.887224 | 0.093668 | 1.318683 |
Example Usage for Source/Sink
import h2o_mlops_scoring_client
ID_COLUMN = "ID"
MLOPS_ENDPOINT_URL = "https://model.internal.dedicated.h2o.ai/aa7a5860-cfb5-4c5a-948c-85a81a061c0d/model/score"
SOURCE_DATA = "file:///Users/jgranados/datasets/BNPParibas.csv"
SINK_LOCATION = "file:///Users/jgranados/datasets/output/"
SOURCE_FORMAT = h2o_mlops_scoring_client.Format.CSV
SINK_FORMAT = h2o_mlops_scoring_client.Format.CSV
SINK_WRITE_MODE = h2o_mlops_scoring_client.WriteMode.OVERWRITE
def preprocess(spark_df):
return spark_df.repartition(30)
h2o_mlops_scoring_client.score_source_sink(
mlops_endpoint_url=MLOPS_ENDPOINT_URL,
id_column=ID_COLUMN,
source_data=SOURCE_DATA,
source_format=SOURCE_FORMAT,
sink_location=SINK_LOCATION,
sink_format=SINK_FORMAT,
sink_write_mode=SINK_WRITE_MODE,
preprocess_method=preprocess,
request_prediction_intervals=True
)
24/02/27 16:45:49 INFO h2o_mlops_scoring_client: Connecting to H2O.ai MLOps scorer at 'https://model.internal.dedicated.h2o.ai/aa7a5860-cfb5-4c5a-948c-85a81a061c0d/model/score'
24/02/27 16:45:52 INFO h2o_mlops_scoring_client: Applying preprocess method
24/02/27 16:45:52 INFO h2o_mlops_scoring_client: Starting scoring from 'file:///Users/jgranados/datasets/BNPParibas.csv' to 'file:///Users/jgranados/datasets/output/'
24/02/27 16:46:16 INFO h2o_mlops_scoring_client: Scoring complete
24/02/27 16:46:16 INFO h2o_mlops_scoring_client: Total run time: 0:00:29
24/02/27 16:46:16 INFO h2o_mlops_scoring_client: Scoring run time: 0:00:24
Feedback
- Submit and view feedback for this page
- Send feedback about H2O MLOps to cloud-feedback@h2o.ai