Added demo for MXNet. And fix accuracy-checker build in docker-file

demo/accuracy_checker_configs/MXNet.xml

@@ -0,0 +1,17 @@
+<?xml version="1.0" encoding="utf-8"?>
+<Tests>
+    <Test>
+        <Model>
+            <Task>classification</Task>
+            <Name>samplenet-0000</Name>
+            <Precision>FP32</Precision>
+            <SourceFramework>MXNet</SourceFramework>
+            <Directory>/media/models</Directory>
+        </Model>
+        <Parameters>
+            <InferenceFramework>MXNet</InferenceFramework>
+            <Device>CPU</Device>
+            <Config>{CONFIG_PATH}</Config>
+        </Parameters>
+    </Test>
+</Tests>

demo/accuracy_checker_configs/MXNet.yml

@@ -0,0 +1,33 @@
+models:
+  - name: SampLeNet_example
+    launchers:
+      - framework: mxnet
+        device: CPU
+        model:   samplenet-0000.params
+        adapter: classification
+        inputs:
+        - name: 'data'
+          type: INPUT
+          shape: 3, 32, 32
+
+    datasets:
+      - name: sample_dataset
+        data_source: sample_dataset/test
+        annotation_conversion:
+          converter: cifar
+          data_batch_file: cifar-10-batches-py/test_batch
+          convert_images: True
+          converted_images_dir: sample_dataset/test
+          num_classes: 10
+
+        preprocessing:
+          - type: resize
+            size: 32
+          - type: bgr_to_rgb
+          - type: normalization
+            mean: (125.307, 122.961, 113.8575)
+            std: (51.5865, 50.847, 51.255)
+
+        metrics:
+          - type: accuracy
+            top_k: 1

demo/benchmark_configs/MXNet.xml

@@ -0,0 +1,35 @@
+<?xml version="1.0" encoding="utf-8"?>
+<Tests>
+    <Test>
+        <Model>
+            <Task>classification</Task>
+            <Name>SampleNet</Name>
+            <Precision>FP32</Precision>
+            <SourceFramework>MXNet</SourceFramework>
+            <ModelPath>/media/models/samplenet-symbol.json</ModelPath>
+            <WeightsPath>/media/models/samplenet-0000.params</WeightsPath>
+        </Model>
+        <Dataset>
+            <Name>ImageNET</Name>
+            <Path>/tmp/{DLI_DATASET_REPO_NAME}/Datasets/ImageNET/</Path>
+        </Dataset>
+        <FrameworkIndependent>
+            <InferenceFramework>MXNet</InferenceFramework>
+            <BatchSize>1</BatchSize>
+            <Device>CPU</Device>
+            <IterationCount>10</IterationCount>
+            <TestTimeLimit>0</TestTimeLimit>
+        </FrameworkIndependent>
+        <FrameworkDependent>
+            <Mode>Sync</Mode>
+            <InputName>data</InputName>
+            <InputShape>1 3 32 32</InputShape>
+            <Hybridize>True</Hybridize>
+            <Normalize>True</Normalize>
+            <Mean></Mean>
+            <Std></Std>
+            <ChannelSwap></ChannelSwap>
+            <Quantization></Quantization>
+        </FrameworkDependent>
+    </Test>
+</Tests>

demo/linux.sh

@@ -1,6 +1,6 @@
 #!/bin/bash
 
-supported_frameworks="OpenVINO_DLDT ONNXRuntime"
+supported_frameworks="OpenVINO_DLDT MXNet ONNXRuntime"
 
 usage() {
     echo "Usage: $0 [-l LOGIN] [-p PASSWORD] [-f FRAMEWORK] [-d GIT_LINK_TO_DATASET]"

docker/MXNet/Dockerfile

@@ -17,5 +17,6 @@ RUN pip uninstall -y numpy && pip install numpy==1.23.1
 # Download Accuracy Checker
 WORKDIR /tmp/
 WORKDIR /tmp/open_model_zoo/tools/accuracy_checker
+RUN python3 setup.py install_core
 RUN /bin/bash -c 'accuracy_check -c /tmp/dl-benchmark/docker/MXNet/config.yml -m data/test_models -s sample'
 WORKDIR /tmp/