DataRobot API resources > API reference documentation > Training predictions

Training predictions¶

class datarobot.models.training_predictions.TrainingPredictionsIterator¶

Lazily fetches training predictions from DataRobot API in chunks of specified size and then iterates rows from responses as named tuples. Each row represents a training prediction computed for a dataset’s row. Each named tuple has the following structure:

Variables:
- row_id (int) – id of the record in original dataset for which training prediction is calculated
- partition_id (str or float) – The ID of the data partition that the row belongs to. “0.0” corresponds to the validation partition or backtest 1.
- prediction (float or str or list of str) – The model’s prediction for this data row.
- prediction_values (list of dict) – An array of dictionaries with a schema described as PredictionValue.
- timestamp (str or None) – (New in version v2.11) an ISO string representing the time of the prediction in time series project; may be None for non-time series projects
- forecast_point (str or None) – (New in version v2.11) an ISO string representing the point in time used as a basis to generate the predictions in time series project; may be None for non-time series projects
- forecast_distance (str or None) – (New in version v2.11) how many time steps are between the forecast point and the timestamp in time series project; None for non-time series projects
- series_id (str or None) – (New in version v2.11) the id of the series in a multiseries project; may be NaN for single series projects; None for non-time series projects
- prediction_explanations (list of dict or None) – (New in version v2.21) The prediction explanations for each feature. The total elements in the array are bounded by max_explanations and feature count. Only present if prediction explanations were requested. Schema described as PredictionExplanations.
- shap_metadata (dict or None) – (New in version v2.21) The additional information necessary to understand SHAP based prediction explanations. Only present if explanation_algorithm equals datarobot.enums.EXPLANATIONS_ALGORITHM.SHAP was added in compute request. Schema described as ShapMetadata.

Notes

Each PredictionValue dict contains these keys:

label : describes what this model output corresponds to. For regression projects, it is the name of the target feature. For classification and multiclass projects, it is a label from the target feature.

value : the output of the prediction. For regression projects, it is the predicted value of the target. For classification and multiclass projects, it is the predicted probability that the row belongs to the class identified by the label.

Each PredictionExplanations dictionary contains these keys:

label (str) : describes what output was driven by this prediction explanation. For regression projects, it is the name of the target feature. For classification projects, it is the class whose probability increasing would correspond to a positive strength of this prediction explanation.

feature (str) : the name of the feature contributing to the prediction

feature_value (object) : the value the feature took on for this row. The type corresponds to the feature (boolean, integer, number, string)

strength (float) : algorithm-specific explanation value attributed to feature in this row

ShapMetadata dictionary contains these keys:

shap_remaining_total (float) : The total of SHAP values for features beyond the max_explanations. This can be identically 0 in all rows, if max_explanations is greater than the number of features and thus all features are returned.

shap_base_value (float) : the model’s average prediction over the training data. SHAP values are deviations from the base value.

warnings (dict or None) : SHAP values calculation warnings (e.g. additivity check failures in XGBoost models). Schema described as ShapWarnings.

ShapWarnings dictionary contains these keys:

mismatch_row_count (int) : the count of rows for which additivity check failed

max_normalized_mismatch (float) : the maximal relative normalized mismatch value

Examples

import datarobot as dr

# Fetch existing training predictions by their id
training_predictions = dr.TrainingPredictions.get(project_id, prediction_id)

# Iterate over predictions
for row in training_predictions.iterate_rows()
    print(row.row_id, row.prediction)

class datarobot.models.training_predictions.TrainingPredictions¶

Represents training predictions metadata and provides access to prediction results.

Variables:
- project_id (str) – id of the project the model belongs to
- model_id (str) – id of the model
- prediction_id (str) – id of generated predictions
- data_subset (datarobot.enums.DATA_SUBSET) –
  
  data set definition used to build predictions. Choices are: - datarobot.enums.DATA_SUBSET.ALL : for all data available. Not valid for models in datetime partitioned projects. - datarobot.enums.DATA_SUBSET.VALIDATION_AND_HOLDOUT : for all data except training set. Not valid for models in datetime partitioned projects. - datarobot.enums.DATA_SUBSET.HOLDOUT : for holdout data set only. - datarobot.enums.DATA_SUBSET.ALL_BACKTESTS : for downloading the predictions for all backtest validation folds. Requires the model to have successfully scored all backtests. Datetime partitioned projects only. * explanation_algorithm (datarobot.enums.EXPLANATIONS_ALGORITHM) – (New in version v2.21) Optional. If set to shap, the response will include prediction explanations based on the SHAP explainer (SHapley Additive exPlanations). Defaults to null (no prediction explanations). * max_explanations (int) – (New in version v2.21) The number of top contributors that are included in prediction explanations. Max 100. Defaults to null for datasets narrower than 100 columns, defaults to 100 for datasets wider than 100 columns. * shap_warnings (list) – (New in version v2.21) Will be present if explanation_algorithm was set to datarobot.enums.EXPLANATIONS_ALGORITHM.SHAP and there were additivity failures during SHAP values calculation.

Notes

Each element in shap_warnings has the following schema:

partition_name (str) : the partition used for the prediction record.

value (object) : the warnings related to this partition.

The objects in value are:

mismatch_row_count (int) : the count of rows for which additivity check failed.

max_normalized_mismatch (float) : the maximal relative normalized mismatch value.

Examples

Compute training predictions for a model on the whole dataset

import datarobot as dr

# Request calculation of training predictions
training_predictions_job = model.request_training_predictions(dr.enums.DATA_SUBSET.ALL)
training_predictions = training_predictions_job.get_result_when_complete()
print('Training predictions {} are ready'.format(training_predictions.prediction_id))

# Iterate over actual predictions
for row in training_predictions.iterate_rows():
    print(row.row_id, row.partition_id, row.prediction)

List all training predictions for a project

import datarobot as dr

# Fetch all training predictions for a project
all_training_predictions = dr.TrainingPredictions.list(project_id)

# Inspect all calculated training predictions
for training_predictions in all_training_predictions:
    print(
        'Prediction {} is made for data subset "{}"'.format(
            training_predictions.prediction_id,
            training_predictions.data_subset,
        )
    )

Retrieve training predictions by id

import datarobot as dr

# Getting training predictions by id
training_predictions = dr.TrainingPredictions.get(project_id, prediction_id)

# Iterate over actual predictions
for row in training_predictions.iterate_rows():
    print(row.row_id, row.partition_id, row.prediction)

classmethod list(project_id)¶

Fetch all the computed training predictions for a project.

Parameters: project_id (str) – id of the project
Return type: A list of TrainingPredictions objects

classmethod get(project_id, prediction_id)¶

Retrieve training predictions on a specified data set.

Parameters:
- project_id (str) – id of the project the model belongs to
- prediction_id (str) – id of the prediction set
Returns: object which is ready to operate with specified predictions
Return type: TrainingPredictions

iterate_rows(batch_size=None)¶

Retrieve training prediction rows as an iterator.

Parameters: batch_size (Optional[int]) – maximum number of training prediction rows to fetch per request
Returns: iterator – an iterator which yields named tuples representing training prediction rows
Return type: TrainingPredictionsIterator

get_all_as_dataframe(class_prefix='class_', serializer='json')¶

Retrieve all training prediction rows and return them as a pandas.DataFrame.

Returned dataframe has the following structure: : - row_id : row id from the original dataset - prediction : the model’s prediction for this row - class_ : the probability that the target is this class (only appears for classification and multiclass projects) - timestamp : the time of the prediction (only appears for out of time validation or time series projects) - forecast_point : the point in time used as a basis to generate the predictions (only appears for time series projects) - forecast_distance : how many time steps are between timestamp and forecast_point (only appears for time series projects) - series_id : he id of the series in a multiseries project or None for a single series project (only appears for time series projects)

Parameters:
- class_prefix (Optional[str]) – The prefix to append to labels in the final dataframe. Default is class_ (e.g., apple -> class_apple)
- serializer (Optional[str]) – Serializer to use for the download. Options: json (default) or csv.
Returns: dataframe
Return type: pandas.DataFrame

download_to_csv(filename, encoding='utf-8', serializer='json')¶

Save training prediction rows into CSV file.

Parameters:
- filename (str or file object) – path or file object to save training prediction rows
- encoding (Optional[str]) – A string representing the encoding to use in the output file, defaults to ‘utf-8’
- serializer (Optional[str]) – Serializer to use for the download. Options: json (default) or csv.