# Copyright (c) Microsoft Corporation. All rights reserved.
# Licensed under the MIT License.

import os
import numpy as np
from utils_cv.classification.widget import AnnotationWidget, ResultsWidget


def test_annotation_widget(tiny_ic_data_path, tmp):
    ANNO_PATH = os.path.join(tmp, "cvbp_ic_annotation.txt")

    w_anno_ui = AnnotationWidget(
        labels=["can", "carton", "milk_bottle", "water_bottle"],
        im_dir=os.path.join(tiny_ic_data_path, "can"),
        anno_path=ANNO_PATH,
        im_filenames=None,  # Set to None to annotate all images in IM_DIR
    )
    w_anno_ui.update_ui()


def test_results_widget(model_pred_scores):
    learn, pred_scores = model_pred_scores

    w_results = ResultsWidget(
        dataset=learn.data.valid_ds,
        y_score=pred_scores,
        y_label=[
            learn.data.classes[x] for x in np.argmax(pred_scores, axis=1)
        ],
    )
    w_results.update()