create genai MCD preprocessor component (#2294)

Co-authored-by: Richard Li <[email protected]>

assets/model_monitoring/components/model_monitor/genai_mdc_preprocessor/asset.yaml

@@ -0,0 +1,3 @@
+type: component
+spec: spec.yaml
+categories: ["Model Monitoring"]

assets/model_monitoring/components/model_monitor/genai_mdc_preprocessor/spec.yaml

@@ -0,0 +1,50 @@
+$schema: http://azureml/sdk-2-0/SparkComponent.json
+type: spark
+
+name: genai_mdc_preprocessor
+display_name: GenAI MDC - Preprocessor
+description: Filters the raw span log based on the window provided, and aggregates it to trace level.
+version: 0.0.1
+is_deterministic: true
+
+code: ../../src
+entry:
+  file: ./model_data_collector_preprocessor/genai_run.py
+
+inputs:
+  data_window_start:
+    type: string
+  data_window_end:
+    type: string
+  input_data:
+    type: uri_folder
+    mode: direct
+outputs:
+  preprocessed_span_data:
+    type: mltable
+    mode: direct
+  aggregated_trace_data:
+    type: mltable
+    mode: direct
+conf:
+  spark.driver.cores: 4
+  spark.driver.memory: 28g
+  spark.executor.cores: 4
+  spark.executor.memory: 28g
+  spark.dynamicAllocation.enabled: True
+  spark.dynamicAllocation.minExecutors: 1
+  spark.dynamicAllocation.maxExecutors: 4
+  spark.synapse.library.python.env: |
+    channels:
+      - conda-forge
+    dependencies:
+      - python=3.8
+      - pip:
+        - mltable~=1.3.0
+    name: momo-base-spark
+args: >-
+  --data_window_end ${{inputs.data_window_end}}
+  --data_window_start ${{inputs.data_window_start}}
+  --input_data ${{inputs.input_data}}
+  --preprocessed_span_data ${{outputs.preprocessed_span_data}}
+  --aggregated_trace_data ${{outputs.aggregated_trace_data}}

assets/model_monitoring/components/src/model_data_collector_preprocessor/genai_run.py

@@ -0,0 +1,6 @@
+# Copyright (c) Microsoft Corporation.
+# Licensed under the MIT License.
+
+"""Entry script for GenAI MDC Preprocessor."""
+
+# place holder file, will add content later