two version of R2R are here HEAD master

1 files changed, 123 insertions, 0 deletions

diff --git a/.venv/lib/python3.12/site-packages/azure/ai/ml/constants/_monitoring.py b/.venv/lib/python3.12/site-packages/azure/ai/ml/constants/_monitoring.py
new file mode 100644
index 00000000..224102ca
--- /dev/null
+++ b/.venv/lib/python3.12/site-packages/azure/ai/ml/constants/_monitoring.py
@@ -0,0 +1,123 @@
+# ---------------------------------------------------------
+# Copyright (c) Microsoft Corporation. All rights reserved.
+# ---------------------------------------------------------
+
+from enum import Enum
+
+from azure.core import CaseInsensitiveEnumMeta
+
+from azure.ai.ml._utils._experimental import experimental
+
+
+ALL_FEATURES = "all_features"
+
+
+AZMONITORING = "azmonitoring"
+
+DEPLOYMENT_MODEL_INPUTS_NAME_KEY = "data_collector.collections.model_inputs.data.name"
+DEPLOYMENT_MODEL_INPUTS_VERSION_KEY = "data_collector.collections.model_inputs.data.version"
+DEPLOYMENT_MODEL_OUTPUTS_NAME_KEY = "data_collector.collections.model_outputs.data.name"
+DEPLOYMENT_MODEL_OUTPUTS_VERSION_KEY = "data_collector.collections.model_outputs.data.version"
+DEPLOYMENT_MODEL_INPUTS_COLLECTION_KEY = "data_collector.collections.model_inputs.enabled"
+DEPLOYMENT_MODEL_OUTPUTS_COLLECTION_KEY = "data_collector.collections.model_outputs.enabled"
+
+
+SPARK_INSTANCE_TYPE_KEY = "compute.spark.resources.instance_type"
+SPARK_RUNTIME_VERSION = "compute.spark.resources.runtime_version"
+
+COMPUTE_AML_TYPE = "AmlToken"
+COMPUTE_MANAGED_IDENTITY_TYPE = "ManagedIdentity"
+
+DEFAULT_DATA_DRIFT_SIGNAL_NAME = "data-drift-signal"
+DEFAULT_PREDICTION_DRIFT_SIGNAL_NAME = "prediction-drift-signal"
+DEFAULT_DATA_QUALITY_SIGNAL_NAME = "data-quality-signal"
+DEFAULT_TOKEN_USAGE_SIGNAL_NAME = "token-usage-signal"
+
+
+@experimental
+class MonitorSignalType(str, Enum, metaclass=CaseInsensitiveEnumMeta):
+    DATA_DRIFT = "data_drift"
+    DATA_QUALITY = "data_quality"
+    PREDICTION_DRIFT = "prediction_drift"
+    MODEL_PERFORMANCE = "model_performance"
+    FEATURE_ATTRIBUTION_DRIFT = "feature_attribution_drift"
+    CUSTOM = "custom"
+    GENERATION_SAFETY_QUALITY = "generation_safety_quality"
+    GENERATION_TOKEN_STATISTICS = "generation_token_statistics"
+
+
+@experimental
+class MonitorMetricName(str, Enum, metaclass=CaseInsensitiveEnumMeta):
+    JENSEN_SHANNON_DISTANCE = "jensen_shannon_distance"
+    NORMALIZED_WASSERSTEIN_DISTANCE = "normalized_wasserstein_distance"
+    POPULATION_STABILITY_INDEX = "population_stability_index"
+    TWO_SAMPLE_KOLMOGOROV_SMIRNOV_TEST = "two_sample_kolmogorov_smirnov_test"
+    PEARSONS_CHI_SQUARED_TEST = "pearsons_chi_squared_test"
+    NULL_VALUE_RATE = "null_value_rate"
+    DATA_TYPE_ERROR_RATE = "data_type_error_rate"
+    OUT_OF_BOUND_RATE = "out_of_bounds_rate"
+    NORMALIZED_DISCOUNTED_CUMULATIVE_GAIN = "normalized_discounted_cumulative_gain"
+    ACCURACY = "accuracy"
+    PRECISION = "precision"
+    RECALL = "recall"
+    F1_SCORE = "f1_score"
+    MAE = "MAE"
+    MSE = "MSE"
+    RMSE = "RMSE"
+
+
+@experimental
+class MonitorModelType(str, Enum, metaclass=CaseInsensitiveEnumMeta):
+    CLASSIFICATION = "classification"
+    REGRESSION = "regression"
+
+
+@experimental
+class MonitorFeatureType(str, Enum, metaclass=CaseInsensitiveEnumMeta):
+    NUMERICAL = "numerical"
+    CATEGORICAL = "categorical"
+    NOT_APPLICABLE = "not_applicable"
+    ALL_FEATURE_TYPES = "all_feature_types"
+
+
+@experimental
+class MonitorDatasetContext(str, Enum, metaclass=CaseInsensitiveEnumMeta):
+    MODEL_INPUTS = "model_inputs"
+    MODEL_OUTPUTS = "model_outputs"
+    TRAINING = "training"
+    TEST = "test"
+    VALIDATION = "validation"
+    GROUND_TRUTH_DATA = "ground_truth"
+
+
+class MonitorTargetTasks(str, Enum, metaclass=CaseInsensitiveEnumMeta):
+    CLASSIFICATION = "Classification"
+    REGRESSION = "Regression"
+    QUESTION_ANSWERING = "QuestionAnswering"
+
+
+class MonitorInputDataType(str, Enum, metaclass=CaseInsensitiveEnumMeta):
+    #: An input data with a fixed window size.
+    STATIC = "Static"
+    #: An input data which trailing relatively to the monitor's current run.
+    TRAILING = "Trailing"
+    #: An input data with tabular format which doesn't require preprocessing.
+    FIXED = "Fixed"
+
+
+class FADColumnNames(str, Enum, metaclass=CaseInsensitiveEnumMeta):
+    PREDICTION = "prediction"
+    PREDICTION_PROBABILITY = "prediction_probability"
+    CORRELATION_ID = "correlation_id"
+
+
+class MonitorFeatureDataType(str, Enum, metaclass=CaseInsensitiveEnumMeta):
+    NUMERICAL = "numerical"
+    CATEGORICAL = "categorical"
+
+
+class NumericalMetricThresholds(str, Enum, metaclass=CaseInsensitiveEnumMeta):
+    JENSEN_SHANNON_DISTANCE = "jensen_shannon_distance"
+    NORMALIZED_WASSERSTEIN_DISTANCE = "normalized_wasserstein_distance"
+    POPULATION_STABILITY_INDEX = "population_stability_index"
+    TWO_SAMPLE_KOLMOGOROV_SMIRNOV_TEST = "two_sample_kolmogorov_smirnov_test"