Deprecate GPU-related files, improve README, resolve CVEs (#6986)

Author: vtaskow

components/alibi-explain-server/Dockerfile

@@ -1,6 +1,6 @@
 ARG VERSION
 ARG BASE_IMAGE
-FROM ${BASE_IMAGE}:${VERSION} as base
+FROM ${BASE_IMAGE}:${VERSION} AS base
 
 ARG VERSION
 LABEL name="Seldon Alibi Wrapper" \
@@ -10,7 +10,7 @@ LABEL name="Seldon Alibi Wrapper" \
       summary="Alibi Explainer Wrapper for Seldon Core" \
       description="Allows Seldon Core inference models to run with a black box model explanation model from the Alibi:Explain project"
 
-FROM base as builder
+FROM base AS builder
 
 USER root
 
@@ -24,25 +24,21 @@ RUN mkdir microservice
 WORKDIR /microservice
 
 # Install Poetry
-ENV POETRY_HOME /microservice/.poetry
+ENV POETRY_HOME=/microservice/.poetry
 RUN /opt/conda/bin/pip install --no-cache-dir "poetry==2.1.2"
 
 # Replace vulnerable pip wheel embedded inside virtualenv(CVE-2025-8869)
-RUN find /opt/conda/lib/python3.12/site-packages/virtualenv/seed/wheels/embed/ -name "pip-*.whl" -delete && \
-    /opt/conda/bin/pip wheel pip==25.3.0 --wheel-dir /opt/conda/lib/python3.12/site-packages/virtualenv/seed/wheels/embed/
+RUN PIP_WHEEL_DIR=/opt/conda/lib/python3.12/ensurepip/_bundled && \
+    /opt/conda/bin/pip wheel "pip==25.3.0" "setuptools>=75.0.0,<76.0.0" "wheel>=0.44.0,<0.45.0" --wheel-dir $PIP_WHEEL_DIR && \
+    rm -f $PIP_WHEEL_DIR/pip-25.0.1-py3-none-any.whl
 
-ENV PATH "$POETRY_HOME/bin:$PATH"
-ENV POETRY_VIRTUALENVS_CREATE false
+ENV PATH="$POETRY_HOME/bin:$PATH"
+ENV POETRY_VIRTUALENVS_CREATE=false
 
 # Install the dependencies only
 COPY poetry.lock pyproject.toml ./
-## Disable Poetry's new installer to avoid JSONDecodeError
-## https://github.com/python-poetry/poetry/issues/4210
-## NOTE: Removing explicitly requirements.txt file from subdeps test
-## dependencies causing false positives in Snyk.
 RUN poetry install --no-root
 
-
 # Add licences
 RUN mkdir /licenses
 RUN mkdir ./licenses && pip-licenses --from=mixed --format=csv --output-file=./licenses/license_info.csv && \
@@ -62,7 +58,7 @@ COPY README.md README.md
 # Install the project code
 RUN poetry install
 
-FROM base as final
+FROM base AS final
 WORKDIR /microservice
 
 RUN microdnf -y update

components/alibi-explain-server/Makefile

@@ -18,7 +18,6 @@ get_apis:
 	$(MAKE) -C ${SELDON_CORE_DIR}/proto/tensorflow clean
 
 build_apis: get_apis
-	pip install -r requirements-apis.txt
 	cd alibiexplainer && python \
 		-m grpc.tools.protoc \
 		-I./ \
@@ -49,15 +48,9 @@ fmt:
 docker-build:
 	docker build --platform=linux/amd64 --file=Dockerfile --build-arg BASE_IMAGE=${BASE_IMAGE} --build-arg VERSION=${VERSION} -t ${IMAGE}:${VERSION} .
 
-docker-build-gpu:
-	docker build --file=Dockerfile.gpu -t ${IMAGE}-gpu:${VERSION} .
-
 docker-push:
 	docker push ${IMAGE}:${VERSION}
 
-docker-push-gpu:
-	docker push ${IMAGE}-gpu:${VERSION}
-
 kind_load: docker-build
 	kind load docker-image ${IMAGE}:${VERSION} --name ${KIND_NAME}
 
@@ -78,7 +71,6 @@ clean:
 #
 # Test Tabular Explanations
 #
-
 test_models/sklearn/iris:
 	mkdir -p test_models/sklearn/iris
 	gsutil cp -r gs://seldon-models/v1.11.0-dev/sklearn/iris test_models/sklearn
@@ -88,7 +80,7 @@ test_models/explainers/anchor_tabular:
 	python tests/make_test_models.py --model anchor_tabular --model_dir test_models/explainers/anchor_tabular
 
 anchor_tabular_model: test_models/sklearn/iris
-	docker run -it --rm --name "sklearnserver"  -p 9000:9000 -v ${PWD}/test_models:/models -e PREDICTIVE_UNIT_PARAMETERS='[{"type":"STRING","name":"model_uri","value":"/models/sklearn/iris"}]' ${REPO}/sklearnserver:${VERSION}
+	docker run -it --rm --name "sklearnserver"  -p 9000:9000 -v ${PWD}/test_models:/models -e PREDICTIVE_UNIT_PARAMETERS='[{"type":"STRING","name":"model_uri","value":"/models/sklearn/iris"}]' ${DOCKER_REGISTRY}/sklearnserver:${VERSION}
 
 anchor_tabular_predict:
 	curl -d '{"data": {"ndarray":[[5.964, 4.006, 2.081, 1.031]]}}'    -X POST http://localhost:9000/api/v1.0/predictions    -H "Content-Type: application/json"
@@ -102,19 +94,15 @@ anchor_tabular_docker: test_models/explainers/anchor_tabular
 anchor_tabular_explain:
 	curl -d '{"data": {"ndarray":[[5.964, 4.006, 2.081, 1.031]]}}'    -X POST http://localhost:8080/api/v1.0/explain    -H "Content-Type: application/json"
 
-
 #
 # Test Text Explanations
 #
-
-
 test_models/sklearn/moviesentiment:
 	mkdir -p test_models/sklearn
 	gsutil cp -r gs://seldon-models/sklearn/moviesentiment_sklearn_0.24.2 test_models/sklearn
 
 anchor_text_model: test_models/sklearn/moviesentiment
-	docker run -it --rm --name "sklearnserver"  -p 9000:9000 -v ${PWD}/test_models:/models -e PREDICTIVE_UNIT_PARAMETERS='[{"type":"STRING","name":"model_uri","value":"/models/sklearn/moviesentiment_sklearn_0.24.2"}]' ${REPO}/sklearnserver:${VERSION}
-
+	docker run -it --rm --name "sklearnserver"  -p 9000:9000 -v ${PWD}/test_models:/models -e PREDICTIVE_UNIT_PARAMETERS='[{"type":"STRING","name":"model_uri","value":"/models/sklearn/moviesentiment_sklearn_0.24.2"}]' ${DOCKER_REGISTRY}/sklearnserver:${VERSION}
 
 anchor_text_predict:
 	curl -d '{"data": {"ndarray":["a visually exquisite but narratively opaque and emotionally vapid experience of style and mystification"]}}'    -X POST http://localhost:9000/api/v1.0/predictions    -H "Content-Type: application/json"
@@ -128,11 +116,9 @@ anchor_text_docker:
 anchor_text_explain:
 	curl -d '{"data": {"ndarray":["a visually exquisite but narratively opaque and emotionally vapid experience of style and mystification"]}}'    -X POST http://localhost:8080/api/v1.0/explain    -H "Content-Type: application/json"
 
-
 #
 # Test Image Explanation
 #
-
 test_models/tfserving/cifar10/resnet32:
 	mkdir -p test_models/tfserving/cifar10
 	gsutil cp -r gs://seldon-models/tfserving/cifar10/resnet32 test_models/tfserving/cifar10
@@ -144,11 +130,9 @@ test_models/explainers/anchor_image:
 anchor_images_model: test_models/tfserving/cifar10/resnet32
 	docker run --name tfserver -it --rm  -p 8501:8501 -p 8500:8500 -v "${PWD}/test_models/tfserving/cifar10:/models" -e MODEL_NAME=resnet32 tensorflow/serving
 
-
 anchor_images_predict:
 	curl -d @./tests/data/input.json  -X POST http://localhost:8501/v1/models/resnet32:predict    -H "Content-Type: application/json"
 
-
 anchor_images: test_models/explainers/anchor_image
 	python -m alibiexplainer --model_name resnet32 --protocol tensorflow.http --storage_uri "${PWD}/test_models/explainers/anchor_image" --predictor_host localhost:8501 AnchorImages
 
@@ -158,12 +142,9 @@ anchor_images_docker: test_models/explainers/anchor_image
 anchor_images_explain:
 	curl -d @./tests/data/input.json  -X POST http://localhost:8080/v1/models/resnet32:explain    -H "Content-Type: application/json"
 
-
 #
 # Test Kernel Shap Explanation
 #
-
-
 test_models/sklearn/wine/model-py36-0.23.2:
 	mkdir -p test_models/sklearn/wine
 	gsutil cp -r gs://seldon-models/sklearn/wine/model-py36-0.23.2 test_models/sklearn/wine
@@ -173,24 +154,20 @@ test_models/explainers/kernel_shap:
 	python tests/make_test_models.py --model kernel_shap --model_dir test_models/explainers/kernel_shap
 
 kernel_shap_model: test_models/sklearn/wine/model-py36-0.23.2
-	docker run -it --rm --name "sklearnserver"  -p 9000:9000 -v ${PWD}/test_models:/models -e PREDICTIVE_UNIT_PARAMETERS='[{"type":"STRING","name":"model_uri","value":"/models/sklearn/wine/model-py36-0.23.2"},{"type":"STRING","name":"method","value":"decision_function"}]' ${REPO}/sklearnserver:${VERSION}
+	docker run -it --rm --name "sklearnserver"  -p 9000:9000 -v ${PWD}/test_models:/models -e PREDICTIVE_UNIT_PARAMETERS='[{"type":"STRING","name":"model_uri","value":"/models/sklearn/wine/model-py36-0.23.2"},{"type":"STRING","name":"method","value":"decision_function"}]' ${DOCKER_REGISTRY}/sklearnserver:${VERSION}
 
 kernel_shap_predict:
 	curl -d '{"data": {"ndarray":[[-0.24226334,  0.26757916,  0.42085937,  0.7127641 ,  0.84067236, -1.27747161, -0.60582812, -0.9706341 , -0.5873972 ,  2.42611713, -2.06608025, -1.55017035, -0.86659858]]}}'    -X POST http://localhost:9000/api/v1.0/predictions    -H "Content-Type: application/json"
 
-
 kernel_shap: test_models/explainers/kernel_shap
 	python -m alibiexplainer --model_name wine --protocol seldon.http --storage_uri "${PWD}/test_models/explainers/kernel_shap" --predictor_host localhost:9000 KernelShap
 
-
 kernel_shap_docker: test_models/explainers/kernel_shap
 	docker run -it --rm  --name "explainer" --network=host -p 8080:8080 -v ${PWD}/test_models:/models ${IMAGE}:${VERSION} --model_name wine --protocol seldon.http --storage_uri /models/explainers/kernel_shap --predictor_host localhost:9000 KernelShap
 
-
 kernel_shap_explain:
 	curl -d '{"data": {"ndarray":[[-0.24226334,  0.26757916,  0.42085937,  0.7127641 ,  0.84067236, -1.27747161, -0.60582812, -0.9706341 , -0.5873972 ,  2.42611713, -2.06608025, -1.55017035, -0.86659858]]}}'    -X POST http://localhost:8080/api/v1.0/explain    -H "Content-Type: application/json"
 
-
 #
 # Test Integrated Gradients
 #
@@ -201,15 +178,12 @@ test_models/keras/imdb:
 integrated_gradients: test_models/keras/imdb
 	python -m alibiexplainer --model_name imdb --protocol seldon.http --storage_uri ${PWD}/test_models/keras/imdb  IntegratedGradients IntegratedGradients --layer 1
 
-
 integrated_gradients_docker: test_models/keras/imdb
 	docker run -it --rm  --name "explainer" --network=host -p 8080:8080 -v ${PWD}/test_models:/models ${IMAGE}:${VERSION} --model_name adult --protocol seldon.http --storage_uri /models/keras/imdb IntegratedGradients --layer 1
 
 integrated_gradients_explain:
 	curl -d '{"data": {"ndarray":[[   0,    0,    0,    0,    0,    0,    0,    0,    0,    0,    0, 0,    0,    0,    0,    0,    0,    0,    0,    0,    0,    0, 0,    0,    0,    0,    0,    0,    0,    0,    0,    0,    1, 591,  202,   14,   31,    6,  717,   10,   10,    2,    2,    5, 4,  360,    7,    4,  177, 5760,  394,  354,    4,  123,    9, 1035, 1035, 1035,   10,   10,   13,   92,  124,   89,  488, 7944, 100,   28, 1668,   14,   31,   23,   27, 7479,   29,  220,  468, 8,  124,   14,  286,  170,    8,  157,   46,    5,   27,  239, 16,  179,    2,   38,   32,   25, 7944,  451,  202,   14,    6, 717]]}}'    -X POST http://localhost:8080/api/v1.0/explain    -H "Content-Type: application/json"
 
-
-
 #
 # Test Tree Shap
 # This is an odd case where we do not need a separate model (yet)
@@ -240,50 +214,16 @@ test_models/explainers/ale:
 	python tests/make_test_models.py --model ale --model_dir test_models/explainers/ale
 
 ale_model: test_models/sklearn/iris-0.23.2/lr_model
-	docker run -it --rm --name "sklearnserver"  -p 9000:9000 -v ${PWD}/test_models:/models -e PREDICTIVE_UNIT_PARAMETERS='[{"type":"STRING","name":"model_uri","value":"/models/sklearn/iris-0.23.2/lr_model"},{"type":"STRING","name":"method","value":"decision_function"}]' ${REPO}/sklearnserver:${VERSION}
+	docker run -it --rm --name "sklearnserver"  -p 9000:9000 -v ${PWD}/test_models:/models -e PREDICTIVE_UNIT_PARAMETERS='[{"type":"STRING","name":"model_uri","value":"/models/sklearn/iris-0.23.2/lr_model"},{"type":"STRING","name":"method","value":"decision_function"}]' ${DOCKER_REGISTRY}/sklearnserver:${VERSION}
 
 ale_predict:
 	curl -d '{"data": {"ndarray":[[6.1, 2.8, 4.7, 1.2]]}}'    -X POST http://localhost:9000/api/v1.0/predictions    -H "Content-Type: application/json"
 
-
 ale: test_models/explainers/ale
 	python -m alibiexplainer --model_name iris --protocol seldon.http --storage_uri "${PWD}/test_models/explainers/ale" --predictor_host localhost:9000 ALE
 
-
 ale_docker: test_models/explainers/ale
 	docker run -it --rm  --name "explainer" --network=host -p 8080:8080 -v ${PWD}/test_models:/models ${IMAGE}:${VERSION} --model_name iris --protocol seldon.http --storage_uri /models/explainers/ale --predictor_host localhost:9000 ALE
 
-
 ale_explain:
 	curl -d '{"data": {"ndarray":[[6.1, 2.8, 4.7, 1.2]]}}'    -X POST http://localhost:8080/api/v1.0/explain    -H "Content-Type: application/json"
-
-
-
-#
-# Test Triton Cifar10
-#
-# TODO: move the below in relevant place
-# as we are moving to mlserver alibi runtime for V2 protocol
-
-#test_models/triton/cifar10/tf_cifar10:
-#	mkdir -p test_models/triton/tf_cifar10
-#	gsutil cp -r gs://seldon-models/triton/tf_cifar10 test_models/triton
-#
-#
-#anchor_images_triton_model: test_models/triton/cifar10/tf_cifar10
-#	docker run --rm --shm-size=1g --ulimit memlock=-1 --ulimit stack=67108864 -p9000:9000 -p8001:8001 -p8002:8002 -p5001:5001 -v ${PWD}/test_models/triton/tf_cifar10:/models nvcr.io/nvidia/tritonserver:21.08-py3 /opt/tritonserver/bin/tritonserver --model-repository=/models --http-port=9000 --grpc-port=5001
-#
-#anchor_images_triton_predict:
-#	curl -H "Content-Type: application/json" http://0.0.0.0:9000/v2/models/cifar10/infer -d '@tests/data/truck-v2.json'
-#
-#
-#anchor_images_triton: test_models/explainers/anchor_image
-#	python -m alibiexplainer --model_name cifar10 --protocol kfserving.http --storage_uri ${PWD}/explainers/anchor_image --predictor_host localhost:9000 AnchorImages
-#
-#
-#anchor_images_triton_docker: test_models/explainers/anchor_image
-#	docker run -it --rm --name "explainer" --network=host -p 8080:8080 -v ${PWD}/test_models:/models ${IMAGE}:${VERSION} --model_name cifar10 --protocol kfserving.http --storage_uri /models/explainers/anchor_image --predictor_host localhost:9000 AnchorImages
-#
-#
-#anchor_images_triton_explain:
-#	curl -d @tests/data/truck-v2.json  -X POST http://localhost:8080/v2/models/cifar10/explain    -H "Content-Type: application/json"