Add new environment aoai-evaluation (#3531)

* add new environment aoai-evaluation

* remove azureml-metrics from dependencies

* pin packages

* reorder logic for batch score result

* fix code style issue

assets/aml-benchmark/environments/evaluation/asset.yaml

@@ -0,0 +1,6 @@
+name: evaluation
+version: auto
+type: environment
+spec: spec.yaml
+extra_config: environment.yaml
+categories: ["Models"]

assets/aml-benchmark/environments/evaluation/context/Dockerfile

@@ -0,0 +1,23 @@
+FROM mcr.microsoft.com/azureml/openmpi4.1.0-ubuntu22.04:{{latest-image-tag}}
+
+RUN apt-get update && apt-get -y upgrade
+
+WORKDIR /
+ENV CONDA_PREFIX=/azureml-envs/aoai-evaluation
+ENV CONDA_DEFAULT_ENV=$CONDA_PREFIX
+
+COPY conda.yaml .
+# Create conda environment
+RUN conda env create -p $CONDA_PREFIX -f conda.yaml -q && \
+    rm conda.yaml && \
+    conda run -p $CONDA_PREFIX pip cache purge && \
+    conda clean -a -y
+
+# clean conda and pip caches
+RUN rm -rf ~/.cache/pip
+
+# Prepend path to AzureML conda environment
+ENV PATH=$CONDA_PREFIX/bin:$PATH
+
+# This is needed for mpi to locate libpython
+ENV LD_LIBRARY_PATH $CONDA_PREFIX/lib:$LD_LIBRARY_PATH

assets/aml-benchmark/environments/evaluation/context/conda.yaml

@@ -0,0 +1,19 @@
+name: evaluation
+channels:
+- conda-forge
+- anaconda
+dependencies:
+- python=3.10
+- pip=23.1
+- pip:
+  - azureml-mlflow=={{latest-pypi-version}}
+  - azure-ai-ml=={{latest-pypi-version}}
+  - azureml-core=={{latest-pypi-version}}
+  - azureml-telemetry=={{latest-pypi-version}}
+  - mltable=={{latest-pypi-version}}
+  - azureml-automl-core=={{latest-pypi-version}}
+  - rouge-score>=0.1.2
+  - nltk>=3.9.1
+  - pandas>=2.2.3
+  - tiktoken>=0.8.0
+  - openai>=1.52.2

assets/aml-benchmark/environments/evaluation/environment.yaml

@@ -0,0 +1,12 @@
+image:
+  name: azureml/curated/evaluation
+  os: linux
+  context:
+    dir: context
+    dockerfile: Dockerfile
+    template_files:
+    - Dockerfile
+    - conda.yaml
+  publish:
+    location: mcr
+    visibility: public

assets/aml-benchmark/environments/evaluation/spec.yaml

@@ -0,0 +1,15 @@
+$schema: https://azuremlschemas.azureedge.net/latest/environment.schema.json
+description: >-
+  Environment for evaluating models.
+
+name: "{{asset.name}}"
+version: "{{asset.version}}"
+
+build:
+  path: context/
+
+os_type: linux
+
+tags:
+  OS: Ubuntu22.04
+  Evaluation: ""