minor fixes

app/main.py

@@ -15,17 +15,15 @@ logger = logging.getLogger("gradio_log")
 
 
 TAB_LABELS = {
-    "ENG": ["Home", "Simple HTR", "Custom HTR", "Examples"],
-    "SWE": ["Hem", "Enkel HTR", "Anpassad HTR", "Exempel"],
+    "ENG": ["Home", "Simple", "Advanced", "Examples"],
+    "SWE": ["Hem", "Enkel", "Avancerad", "Exempel"],
 }
 
 LANG_CHOICES = ["ENG", "SWE"]
 
 with gr.Blocks(title="HTRflow", theme=theme, css=css) as demo:
     with gr.Row():
-        local_language = gr.BrowserState(
-            default_value="ENG", storage_key="selected_language"
-        )
+        local_language = gr.BrowserState(default_value="ENG", storage_key="selected_language")
         main_language = gr.State(value="ENG")
 
         with gr.Column(scale=1):
@@ -47,17 +45,17 @@ with gr.Blocks(title="HTRflow", theme=theme, css=css) as demo:
         with gr.Tab(label="Home") as tab_home:
             overview.render()
 
-        with gr.Tab(label="Simple HTR") as tab_simple_htr:
+        with gr.Tab(label="Simple") as tab_simple_htr:
             htrflow_pipeline.render()
 
-        with gr.Tab(label="Custom HTR") as tab_custom_htr:
+        with gr.Tab(label="Advanced") as tab_custom_htr:
             adv_htrflow_pipeline.render()
 
         with gr.Tab(label="Examples") as tab_examples:
             examples.render()
 
-        with gr.Tab(label="Data Explorer") as tab_data_explorer:
-            data_explorer.render()
+        # with gr.Tab(label="Data Explorer") as tab_data_explorer:
+        #     data_explorer.render()
 
     @demo.load(
         inputs=[local_language],

app/tabs/adv_htrflow_tab.py

@@ -4,14 +4,19 @@ import gradio as gr
 
 with gr.Blocks() as adv_htrflow_pipeline:
     with gr.Row(variant="panel"):
-        with gr.Column():
-            # TODO: We want to either crop or draw polygon or bbox and send it to the custom model. Or just as the image is.
-            # TODO: For the viewer we should be able to select from the output of the model what for values we want to
-            # TODO: add batch predictions here..
-            # TODO: add load from s3, local, hf daasets( however everything could go through hf_datasets).
-            # TODO: send a crop from the user
+        with gr.Column(scale=2):
+            image_mask2 = gr.ImageEditor(
+                label="Uploaded image", interactive=True, layers=False, eraser=False, brush=False, height=400
+            )
 
-            image_mask = gr.ImageMask(interactive=True)
+            image_mask = gr.Gallery(
+                file_types=["image"],
+                label="Upload images",
+                interactive=True,
+                height=400,
+                object_fit="cover",
+                columns=5,
+            )
 
             with gr.Group():
                 with gr.Row(visible=True) as yaml_pipeline:
@@ -22,21 +27,21 @@ with gr.Blocks() as adv_htrflow_pipeline:
                             label="yaml",
                             interactive=True,
                         )
-                    test = gr.Dropdown(  # TODO: This should be a dropdown to decide input image or mask or s3 or local path
-                        ["Upload", "Draw", "s3", "local", "crop"],
-                        value="Upload",
-                        multiselect=False,
-                        label="Upload method",
-                        container=False,
-                        scale=0,
-                        interactive=True,
-                    )
+                    gr.Checkbox(value=True, label="Batch", container=True, scale=0)
+
+                    # input_files_format_dropdown = gr.Dropdown(
+                    #     ["Upload", "Batch", "Crop"],
+                    #     value="Upload",
+                    #     multiselect=False,
+                    #     label="Upload method",
+                    #     container=False,
+                    #     scale=0,
+                    #     interactive=True,
+                    # )
 
             with gr.Row():
                 run_button = gr.Button("Submit", variant="primary", scale=0)
-                cancel_button = gr.Button(
-                    "stop", variant="stop", scale=0
-                )  # TODO: This should be a cancel button and be hidden until the run button is clicked
+                cancel_button = gr.Button("stop", variant="stop", scale=0)
                 d = gr.DownloadButton(
                     "Download the file", visible=True, scale=0
                 )  # TODO: This should be hidden until the run button is clicked
@@ -48,7 +53,7 @@ with gr.Blocks() as adv_htrflow_pipeline:
                 # TODO: add a upload to hf datasets
                 # TODO: add a hf login button to login to upload datasets
 
-        with gr.Column():
+        with gr.Column(scale=3):
             with gr.Tabs():
                 with gr.Tab("HTR ouput"):
                     gr.CheckboxGroup(
@@ -64,7 +69,7 @@ with gr.Blocks() as adv_htrflow_pipeline:
                     # TODO add https://www.gradio.app/docs/gradio/highlightedtext and graph of run graph
                     pass
 
-        test.select(lambda: gr.update(visible=True), None, image_mask)
+        # input_files_format_dropdown.select(lambda: gr.update(visible=True), None, image_mask)
 
         def foo():
             for i in range(300):

app/tabs/data_explorer_tab copy.py

@@ -0,0 +1,62 @@
+import gradio as gr
+from gradio_modal import Modal
+
+output_image_placehholder = gr.Image(label="Output image", height=400, show_share_button=True)
+
+
+def show_warning(selection: gr.SelectData):
+    gr.Warning(f"Your choice is #{selection.index}, with image: {selection.value['image']['path']}!")
+
+
+with gr.Blocks() as data_explorer:
+    with gr.Row(variant="panel"):
+        with gr.Column(scale=1):
+            output_dataframe_pipeline = gr.Textbox(label="Path", info="path s3 or path, hf-dataset.")
+            with gr.Group():
+                (
+                    gr.Dropdown(
+                        ["ran", "swam", "ate", "slept"],
+                        value=["swam", "slept"],
+                        multiselect=True,
+                        label="Activity",
+                        info="Lorem ipsum dolor sit amet, consectetur adipiscing elit. Sed auctor, nisl eget ",
+                    ),
+                )
+                gr.Slider(2, 20, value=4, label="Count", info="Choose between 2 and 20")
+                gr.Slider(2, 20, value=4, label="Another", info="Choose between 2 and 20")
+            output_dataframe_pipeline = gr.Textbox(label="search", info="search image bla bla..")
+            gr.Button("Search", variant="primary", scale=0)
+        with gr.Column(scale=4):
+            with gr.Tabs():
+                with gr.Tab("Gallery"):
+                    image_gallery = gr.Gallery(
+                        [
+                            "https://unsplash.com/photos/4oaDBgVROGo/download?ixid=M3wxMjA3fDB8MXxhbGx8NHx8fHx8fDJ8fDE3MTA0NjI1MzZ8&force=true&w=640",
+                            "https://unsplash.com/photos/4oaDBgVROGo/download?ixid=M3wxMjA3fDB8MXxhbGx8NHx8fHx8fDJ8fDE3MTA0NjI1MzZ8&force=true&w=640",
+                            "https://unsplash.com/photos/4oaDBgVROGo/download?ixid=M3wxMjA3fDB8MXxhbGx8NHx8fHx8fDJ8fDE3MTA0NjI1MzZ8&force=true&w=640",
+                            "https://unsplash.com/photos/4oaDBgVROGo/download?ixid=M3wxMjA3fDB8MXxhbGx8NHx8fHx8fDJ8fDE3MTA0NjI1MzZ8&force=true&w=640",
+                        ]
+                        * 10,
+                        allow_preview=False,
+                        label="Image Gallery",
+                        preview=False,
+                        columns=[7],
+                        rows=[10],
+                        show_download_button=True,
+                        show_share_button=True,
+                    )
+                with gr.Tab("Embeddings"):
+                    pass
+                    # TODO: add a embedding plot here
+                    # user needs to login hf and get a datasets with embeddings
+
+        with Modal(visible=False) as gallery_modal:
+            with gr.Row():
+                with gr.Column(scale=0):
+                    gr.Markdown("")
+                with gr.Column(scale=4):
+                    gr.Image()
+                with gr.Column(scale=0):
+                    gr.Markdown("")
+
+            image_gallery.select(fn=show_warning, inputs=None).then(lambda: Modal(visible=True), None, gallery_modal)

app/tabs/data_explorer_tab.py

@@ -1,63 +1,28 @@
 import gradio as gr
-import pandas as pd
-from gradio_modal import Modal
 
-output_image_placehholder = gr.Image(label="Output image", height=400, show_share_button=True)
 
-
-def show_warning(selection: gr.SelectData):
-    gr.Warning(f"Your choice is #{selection.index}, with image: {selection.value['image']['path']}!")
+def display_dataset(dataset_repo):
+    return gr.HTML(f"""<iframe
+    src="https://huggingface.co/datasets/{dataset_repo}/embed/viewer/default/train"
+    frameborder="0"
+    width="100%"
+    height="700px"
+></iframe>""")
 
 
 with gr.Blocks() as data_explorer:
     with gr.Row(variant="panel"):
-        with gr.Column(scale=1):
-            output_dataframe_pipeline = gr.Textbox(label="Path", info="path s3 or path, hf-dataset.")
-            with gr.Group():
-                (
-                    gr.Dropdown(
-                        ["ran", "swam", "ate", "slept"],
-                        value=["swam", "slept"],
-                        multiselect=True,
-                        label="Activity",
-                        info="Lorem ipsum dolor sit amet, consectetur adipiscing elit. Sed auctor, nisl eget ",
-                    ),
-                )
-                gr.Slider(2, 20, value=4, label="Count", info="Choose between 2 and 20")
-                gr.Slider(2, 20, value=4, label="Another", info="Choose between 2 and 20")
-            output_dataframe_pipeline = gr.Textbox(label="search", info="search image bla bla..")
-            gr.Button("Search", variant="primary", scale=0)
-        with gr.Column(scale=4):
-            with gr.Tabs():
-                with gr.Tab("Gallery"):
-                    image_gallery = gr.Gallery(
-                        [
-                            "https://unsplash.com/photos/4oaDBgVROGo/download?ixid=M3wxMjA3fDB8MXxhbGx8NHx8fHx8fDJ8fDE3MTA0NjI1MzZ8&force=true&w=640",
-                            "https://unsplash.com/photos/4oaDBgVROGo/download?ixid=M3wxMjA3fDB8MXxhbGx8NHx8fHx8fDJ8fDE3MTA0NjI1MzZ8&force=true&w=640",
-                            "https://unsplash.com/photos/4oaDBgVROGo/download?ixid=M3wxMjA3fDB8MXxhbGx8NHx8fHx8fDJ8fDE3MTA0NjI1MzZ8&force=true&w=640",
-                            "https://unsplash.com/photos/4oaDBgVROGo/download?ixid=M3wxMjA3fDB8MXxhbGx8NHx8fHx8fDJ8fDE3MTA0NjI1MzZ8&force=true&w=640",
-                        ]
-                        * 10,
-                        allow_preview=False,
-                        label="Image Gallery",
-                        preview=False,
-                        columns=[7],
-                        rows=[10],
-                        show_download_button=True,
-                        show_share_button=True,
-                    )
-                with gr.Tab("Embeddings"):
-                    pass
-                    # TODO: add a embedding plot here
-                    # user needs to login hf and get a datasets with embeddings
-
-        with Modal(visible=False) as gallery_modal:
-            with gr.Row():
-                with gr.Column(scale=0):
-                    gr.Markdown("")
-                with gr.Column(scale=4):
-                    gr.Image()
-                with gr.Column(scale=0):
-                    gr.Markdown("")
+        with gr.Column(scale=0, min_width=160):
+            input_datasets_path = gr.Textbox(
+                label="HF datasets", placeholder="Gabriel/linkoping", scale=0, container=False
+            )
+            view_dataset = gr.Button("View dataseet", variant="primary", scale=0)
+            gr.LoginButton("Login to HF", variant="secondary", scale=0)
+        with gr.Column():
+            iframe_output = gr.HTML()
 
-            image_gallery.select(fn=show_warning, inputs=None).then(lambda: Modal(visible=True), None, gallery_modal)
+        view_dataset.click(
+            fn=display_dataset,
+            inputs=input_datasets_path,
+            outputs=[iframe_output],
+        )

app/tabs/examples_tab.py

@@ -12,7 +12,7 @@ output_image_placehholder = gr.Image(
     height=400,
     show_share_button=True,
 )
-markdown_selected_option = gr.Markdown(container=True)
+markdown_selected_option = gr.Markdown(value=" ", container=True)
 
 
 def htr_image_placehholder(txt, method, image):

app/tabs/htrflow_tab.py

@@ -41,6 +41,7 @@ def submit_button_pipeline_fn(method, input_image, yaml_str):
 def get_yaml_button_fn(
     method,
     output_formats,
+    reading_order,
     simple_segment_model=None,
     simple_htr_model=None,
     simple_htr_model_type=None,
@@ -54,21 +55,23 @@ def get_yaml_button_fn(
 ):
     env = Environment(loader=FileSystemLoader("app/templates"))
 
+    if output_formats is None:
+        output_formats = ["txt"]
+
     template_name = "steps_template.yaml.j2"
     try:
         if method == "Simple layout":
             steps = [
                 {
                     "step": "Segmentation",
-                    "model": simple_htr_model_type,
+                    "model": simple_segment_model_type,
                     "model_settings": {"model": simple_segment_model},
                 },
                 {
                     "step": "TextRecognition",
-                    "model": simple_segment_model_type,
+                    "model": simple_htr_model_type,
                     "model_settings": {"model": simple_htr_model},
                 },
-                {"step": "OrderLines"},
             ]
         elif method == "Nested segmentation":
             steps = [
@@ -87,11 +90,12 @@ def get_yaml_button_fn(
                     "model": nested_htr_model_type,
                     "model_settings": {"model": nested_htr_model},
                 },
-                {"step": "OrderLines"},
             ]
         else:
             return "Invalid method or not yet supported."
 
+        steps.append({"step": reading_order})
+
         steps.extend(
             {
                 "step": "Export",
@@ -116,7 +120,7 @@ inital_state_selection_option = "Simple layout"
 
 with gr.Blocks() as htrflow_pipeline:
     with gr.Row(variant="panel"):
-        with gr.Column():
+        with gr.Column(scale=2):
             # gr.Markdown("<h2>Control Panel</h2>")
 
             with gr.Group():
@@ -180,6 +184,13 @@ with gr.Blocks() as htrflow_pipeline:
                         with gr.Column():
                             gr.Textbox("WIP")
                 with gr.Row():
+                    reading_order = gr.Dropdown(
+                        choices=["OrderLines", "ReadingOrderMarginalia"],
+                        value="OrderLines",
+                        multiselect=False,
+                        label="Reading Order",
+                        info="Supported format are: ...",
+                    )
                     output_formats = gr.Dropdown(
                         choices=["txt", "alto", "page"],
                         value="txt",
@@ -193,7 +204,7 @@ with gr.Blocks() as htrflow_pipeline:
                 get_yaml_button = gr.Button("Get Yaml", variant="secondary", scale=0)
                 output_files_pipeline = gr.Files(label="Output files", height=100, visible=False)
 
-        with gr.Column():
+        with gr.Column(scale=2):
             with gr.Tabs():
                 with gr.Tab("Viewer"):
                     with gr.Group():
@@ -224,6 +235,7 @@ with gr.Blocks() as htrflow_pipeline:
         inputs=[
             template_method_radio,
             output_formats,
+            reading_order,
             simple_segment_model,
             simple_htr_model,
             simple_htr_model_type,
@@ -253,6 +265,7 @@ with gr.Blocks() as htrflow_pipeline:
         inputs=[
             template_method_radio,
             output_formats,
+            reading_order,
             simple_segment_model,
             simple_htr_model,
             simple_htr_model_type,

pyproject.toml

@@ -17,9 +17,9 @@ classifiers = [
 requires-python = ">=3.10,<3.13"
 
 dependencies = [
-    # "torch==2.0.1",
+    "torch==2.0.1",
     "htrflow==0.1.3",
-    "gradio>=5.11.0",
+    "gradio[oauth]>=5.11.0",
     "datasets>=3.2.0",
     "pandas>=2.2.3",
     "jinja2>=3.1.4",