AC: MXNet isolated import (#1599)

Author: eaidova

tools/accuracy_checker/accuracy_checker/launcher/mxnet_launcher.py

@@ -1,12 +1,9 @@
 """
 Copyright (c) 2018-2020 Intel Corporation
-
 Licensed under the Apache License, Version 2.0 (the "License");
 you may not use this file except in compliance with the License.
 You may obtain a copy of the License at
-
       http://www.apache.org/licenses/LICENSE-2.0
-
 Unless required by applicable law or agreed to in writing, software
 distributed under the License is distributed on an "AS IS" BASIS,
 WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
@@ -19,7 +16,6 @@
 from pathlib import Path
 
 import numpy as np
-import mxnet
 
 from .launcher import Launcher, LauncherConfigValidator, ListInputsField
 from ..config import PathField, StringField, NumberField, ConfigError
@@ -58,6 +54,13 @@ def parameters(cls):
         return parameters
 
     def __init__(self, config_entry: dict, *args, **kwargs):
+        try:
+            import mxnet # pylint: disable=C0415
+            self.mxnet = mxnet
+        except ImportError as import_error:
+            raise ValueError(
+                "MXNet isn't installed. Please, install it before using. \n{}".format(import_error.msg)
+            )
         super().__init__(config_entry, *args, **kwargs)
         self._delayed_model_loading = kwargs.get('delayed_model_loading', False)
 
@@ -78,9 +81,9 @@ def __init__(self, config_entry: dict, *args, **kwargs):
                 identifier = match.group('identifier')
                 if identifier is None:
                     identifier = 0
-                device_context = mxnet.gpu(int(identifier))
+                device_context = self.mxnet.gpu(int(identifier))
             else:
-                device_context = mxnet.cpu()
+                device_context = self.mxnet.cpu()
 
             # Get batch from config or 1
             self._batch = self.config.get('batch', 1)
@@ -109,7 +112,7 @@ def batch(self):
 
     def fit_to_input(self, data, input_layer, layout, precision):
         data = np.transpose(data, layout)
-        return mxnet.nd.array(data.astype(precision) if precision else data)
+        return self.mxnet.nd.array(data.astype(precision) if precision else data)
 
     @property
     def inputs(self):
@@ -125,9 +128,9 @@ def predict(self, inputs, metadata=None, **kwargs):
         """
         results = []
         for infer_input in inputs:
-            data_iter = mxnet.io.NDArrayIter(
+            data_iter = self.mxnet.io.NDArrayIter(
                 data=infer_input, label=None, batch_size=self.batch)
-            data_batch = mxnet.io.DataBatch(data=data_iter.data_list)
+            data_batch = self.mxnet.io.DataBatch(data=data_iter.data_list)
 
             # Infer
             self.module.forward(data_batch)

tools/accuracy_checker/tests/test_mxnet_launcher.py

@@ -15,7 +15,7 @@
 """
 
 import pytest
-pytest.importorskip('accuracy_checker.launcher.mxnet_launcher')
+pytest.importorskip('mxnet')
 import cv2
 import numpy as np