def setUpClass(self,
input_image_shape=(3, 224, 224),
model="resnet-50",
feature="awesome_image"):
"""
The setup class method for the basic test case with all default values.
"""
self.feature = feature
self.label = None
self.input_image_shape = input_image_shape
self.pre_trained_model = model
# Create the model
self.def_opts = {
"model": "resnet-50",
"verbose": True,
}
# Get deep features if needed
if self.feature.endswith("WithDeepFeature"):
data[self.feature] = get_deep_features(
data["awesome_image"],
self.feature.split('_WithDeepFeature')[0])
# Model
self.model = tc.image_similarity.create(data,
feature=self.feature,
label=None,
model=self.pre_trained_model)
self.nn_model = self.model.feature_extractor
self.lm_model = self.model.similarity_model
self.opts = self.def_opts.copy()
# Answers
self.get_ans = {
"similarity_model":
lambda x: type(x) == tc.nearest_neighbors.NearestNeighborsModel,
"feature":
lambda x: x == self.feature,
"training_time":
lambda x: x > 0,
"input_image_shape":
lambda x: x == self.input_image_shape,
"label":
lambda x: x == self.label,
"feature_extractor":
lambda x: callable(x.extract_features),
"num_features":
lambda x: x == self.lm_model.num_features,
"num_examples":
lambda x: x == self.lm_model.num_examples,
"model":
lambda x: (x == self.pre_trained_model or
(self.pre_trained_model == "VisionFeaturePrint_Screen"
and x == "VisionFeaturePrint_Scene")),
}
self.fields_ans = self.get_ans.keys()
def test_select_correct_feature_column_to_train(self):
# sending both, the correct extracted features colum and image column
if self.feature == "awesome_image":
test_data = data.select_columns([self.feature, self.target])
deep_features_col_name = self.pre_trained_model + "_WithDeepFeature"
test_data[deep_features_col_name] = get_deep_features(
data["awesome_image"], self.pre_trained_model)
test_model = tc.image_classifier.create(
test_data, target=self.target, model=self.pre_trained_model)
self.assertTrue(test_model.feature == deep_features_col_name)
def setUpClass(
self,
model="resnet-50",
feature="awesome_image",
input_image_shape=(3, 224, 224),
tol=0.02,
num_examples=100,
label_type=int,
):
self.feature = feature
self.target = "awesome_label"
self.input_image_shape = input_image_shape
self.pre_trained_model = model
self.tolerance = tol
# Get deep features if needed
if self.feature.endswith("WithDeepFeature"):
data[self.feature] = get_deep_features(
data["awesome_image"],
self.feature.split('_WithDeepFeature')[0])
self.model = tc.image_classifier.create(data,
target=self.target,
feature=self.feature,
model=self.pre_trained_model,
seed=42)
self.nn_model = self.model.feature_extractor
self.lm_model = self.model.classifier
self.max_iterations = 10
self.get_ans = {
"classifier":
lambda x: type(x) == tc.logistic_classifier.LogisticClassifier,
"feature":
lambda x: x == self.feature,
"classes":
lambda x: x == self.lm_model.classes,
"training_time":
lambda x: x > 0,
"input_image_shape":
lambda x: x == self.input_image_shape,
"target":
lambda x: x == self.target,
"feature_extractor":
lambda x: callable(x.extract_features),
"training_loss":
lambda x: x > 0,
"max_iterations":
lambda x: x == self.max_iterations,
"num_features":
lambda x: x == self.lm_model.num_features,
"num_examples":
lambda x: x == self.lm_model.num_examples,
"model":
lambda x: (x == self.pre_trained_model or
(self.pre_trained_model == "VisionFeaturePrint_Screen"
and x == "VisionFeaturePrint_Scene")),
"num_classes":
lambda x: x == self.lm_model.num_classes,
}
self.fields_ans = self.get_ans.keys()