def test_task():
dataset, _, test_dataset = ImageDataset.from_folders('https://autogluon.s3.amazonaws.com/datasets/shopee-iet.zip')
model_list = ImagePredictor.list_models()
classifier = ImagePredictor()
classifier.fit(dataset, hyperparameters={'epochs': 1, 'early_stop_patience': 3}, hyperparameter_tune_kwargs={'num_trials': 2})
assert classifier.fit_summary()['valid_acc'] > 0.1, 'valid_acc is abnormal'
test_result = classifier.predict(test_dataset)
single_test = classifier.predict(test_dataset.iloc[0]['image'])
single_proba = classifier.predict_proba(test_dataset.iloc[0]['image'])
classifier.save('classifier.ag')
classifier2 = ImagePredictor.load('classifier.ag')
fit_summary = classifier2.fit_summary()
test_acc = classifier2.evaluate(test_dataset)
# raw dataframe
df_test_dataset = pd.DataFrame(test_dataset)
test_acc = classifier2.evaluate(df_test_dataset)
assert test_acc[-1] > 0.2, f'{test_acc} too bad'
test_proba = classifier2.predict_proba(test_dataset)
test_feature = classifier2.predict_feature(test_dataset)
single_test2 = classifier2.predict(test_dataset.iloc[0]['image'])
assert isinstance(single_test2, pd.Series)
assert single_test2.equals(single_test)
# to numpy
test_proba_numpy = classifier2.predict_proba(test_dataset, as_pandas=False)
assert np.array_equal(test_proba.to_numpy(), test_proba_numpy)
test_feature_numpy = classifier2.predict_feature(test_dataset, as_pandas=False)
single_test2_numpy = classifier2.predict(test_dataset.iloc[0]['image'], as_pandas=False)
assert np.array_equal(single_test2.to_numpy(), single_test2_numpy)
def test_task_label_remap():
train_dataset, _, test_dataset = ImageDataset.from_folders(
'https://autogluon.s3.amazonaws.com/datasets/shopee-iet.zip')
label_remap = {0: 'd', 1: 'c', 2: 'b', 3: 'a'}
train_dataset = train_dataset.replace({"label": label_remap})
test_dataset = test_dataset.replace({"label": label_remap})
# rename label column
train_dataset = train_dataset.rename(columns={'label': 'my_label'})
predictor = ImagePredictor(label='my_label')
dataset_copy = copy.deepcopy(train_dataset)
predictor.fit(train_dataset, hyperparameters={'epochs': 2})
# assert input dataset not altered
assert dataset_copy.equals(train_dataset)
pred = predictor.predict(test_dataset)
pred_proba = predictor.predict_proba(test_dataset)
label_remap_inverse = {
col_name: i
for i, col_name in enumerate(list(pred_proba.columns))
}
from autogluon.core.metrics import accuracy, log_loss
score_accuracy = accuracy(y_true=test_dataset['label'], y_pred=pred)
score_log_loss = log_loss(
y_true=test_dataset['label'].replace(label_remap_inverse),
y_pred=pred_proba.to_numpy())
assert score_accuracy > 0.2 # relax
def test_invalid_image_dataset():
invalid_test = download(
'https://autogluon.s3-us-west-2.amazonaws.com/miscs/test_autogluon_invalid_dataset.zip'
)
invalid_test = unzip(invalid_test)
df = ImageDataset.from_csv(os.path.join(invalid_test, 'train.csv'),
root=os.path.join(invalid_test, 'train_images'))
predictor = ImagePredictor(label="labels")
predictor.fit(df, df.copy(), time_limit=60)
def test_image_predictor(fit_helper):
from autogluon.vision import ImageDataset
train_data, _, test_data = ImageDataset.from_folders('https://autogluon.s3.amazonaws.com/datasets/shopee-iet.zip')
feature_metadata = FeatureMetadata.from_df(train_data).add_special_types({'image': ['image_path']})
predictor = TabularPredictor(label='label').fit(
train_data=train_data,
hyperparameters={'AG_IMAGE_NN': {'epochs': 2, 'model': 'resnet18_v1b'}},
feature_metadata=feature_metadata
)
leaderboard = predictor.leaderboard(test_data)
assert len(leaderboard) > 0
def test_image_predictor_presets():
train_dataset, _, test_dataset = ImageDataset.from_folders(
'https://autogluon.s3.amazonaws.com/datasets/shopee-iet.zip')
for preset in [
'medium_quality_faster_train',
]:
predictor = ImagePredictor()
predictor.fit(train_dataset,
tuning_data=test_dataset,
presets=[preset],
time_limit=300,
hyperparameters={'epochs': 1})