Remove unused variables and clean functions in preprocessing subpackage

Author: ruancomelli

boiling_learning/preprocessing/visualize.py

@@ -489,7 +489,6 @@ def _make_ds(params: Parameters) -> tf.data.Dataset:
         params[['creator', 'value', 'image_dataset']] = img_ds
 
         ds = _make_ds(params)
-        ds_train, ds_val, ds_test = ds
         for col, (split_name, ds_split) in enumerate(
             zip(('train', 'val', 'test'), ds)
         ):
@@ -501,7 +500,7 @@ def _make_ds(params: Parameters) -> tf.data.Dataset:
                 # wspace=0.1,
                 # hspace=0.1
             )
-            for sample, (img, data) in enumerate(
+            for sample, (img, _) in enumerate(
                 ds_split.take(n_samples).as_numpy_iterator()
             ):
                 ax = plt.Subplot(fig, inner[sample])
@@ -538,56 +537,3 @@ def _make_ds(params: Parameters) -> tf.data.Dataset:
 
                 fig.add_subplot(ax)
                 fig.show()
-
-
-# def interact_frames(
-#         image_datasets: Sequence[ImageDataset]
-# ) -> None:
-#     image_datasets = tuple(image_datasets)
-#     img_ds_options = [(img_ds.name, img_ds) for img_ds in image_datasets]
-
-#     default_img_ds = img_ds_options[0][1]
-#     img_ds_widget = widgets.Dropdown(
-#         options=img_ds_options,
-#         description='Case name:',
-#         value=default_img_ds
-#     )
-
-#     experiment_videos_options = [
-#         (experiment_video.name, experiment_video)
-#         for experiment_video in img_ds_widget.value.values()
-#     ]
-#     experiment_videos_widget = widgets.Dropdown(
-#         options=experiment_videos_options,
-#         description='Experiment video:',
-#         value=experiment_videos_options[0][1]
-#     )
-
-#     def update_videos_list(changes):
-#         experiment_videos_widget.options = [
-#             (experiment_video.name, experiment_video)
-#             for experiment_video in changes['new'].values()
-#         ]
-
-#     img_ds_widget.observe(update_videos_list, 'value')
-
-#     with experiment_videos_widget.value.frames() as f:
-#         index_widget = widgets.IntSlider(
-#             value=0,
-#             min=0,
-#             max=len(f),
-#             description='Frame:'
-#         )
-
-#     def update_max_index(changes):
-#         with changes['new'].frames() as f:
-#             index_widget.max = len(f)
-
-#     experiment_videos_widget.observe(update_max_index, 'value')
-
-#     def show_frames(img_ds, ev, idx):
-#         frame = skimage.transform.downscale_local_mean(ev.frame(idx), (8, 8, 1))
-#         cv2_imshow(frame)
-
-#     if should_interact_frames:
-#         interact(show_frames, img_ds=img_ds_widget, ev=experiment_videos_widget, idx=index_widget)