v0.10.5 (#26)

- Summary (f6125144e9df803bac37023b6854f03c30d62a31)
- Merge branch 'main' of https://huggingface.co/spaces/ElenaRyumina/OCEANAI (2f050c06a94348ba17660fdf0dd257a7a30f2f9a)
- Summary (8f52d543a1de5a77b59164e6f4f721696fc64299)
- Merge branch 'main' of https://huggingface.co/spaces/ElenaRyumina/OCEANAI (fffa0062f0da04563ce8cd5b8c7e70e3ae783833)
- Summary (1717e06c53138f3eaf52e38f984035d5fbbceac5)
- Merge branch 'main' of https://huggingface.co/spaces/ElenaRyumina/OCEANAI (cd90a1620005039e6998ca3d0ab34fe3bfe1f2dc)
- Summary (98c66eeb6760ebd60d0a4ae4d9d0ac8f896a59df)
- Update config.toml (91d9d7ef4e540930919833c79607c51fc294d60d)
- Merge branch 'main' of https://huggingface.co/spaces/ElenaRyumina/OCEANAI (e6358979bef7df974cdd6c63aca029c57ac62af8)
- Summary (6de61f5802c0e3640047369ad08ecab73b331d34)


Co-authored-by: Dmitry Ryumin <DmitryRyumin@users.noreply.huggingface.co>

README.md

@@ -4,7 +4,7 @@ emoji: 😀🤓😎😉😤
 colorFrom: gray
 colorTo: red
 sdk: gradio
-sdk_version: 5.6.0
+sdk_version: 5.7.1
 app_file: app.py
 pinned: false
 license: mit

app.py

@@ -93,6 +93,6 @@ if __name__ == "__main__":
 
     create_gradio_app().queue(api_open=False).launch(
         share=False,
-        server_name=None,
-        server_port=None,
+        server_name=config_data.AppSettings_SERVER_NAME,
+        server_port=config_data.AppSettings_PORT,
     )

app/data_init.py

@@ -0,0 +1,16 @@
+"""
+File: data_init.py
+Author: Elena Ryumina and Dmitry Ryumin
+Description: Data initialization.
+License: MIT License
+"""
+
+from app.config import config_data
+from app.utils import read_csv_file, extract_profession_weights
+
+
+df_traits_priority_for_professions = read_csv_file(config_data.Links_PROFESSIONS)
+weights_professions, interactive_professions = extract_profession_weights(
+    df_traits_priority_for_professions,
+    config_data.Settings_DROPDOWN_CANDIDATES[0],
+)

app/event_handlers/calculate_practical_tasks.py

@@ -16,11 +16,13 @@ from bs4 import BeautifulSoup
 from app.config import config_data
 from app.video_metadata import video_metadata
 from app.mbti_description import MBTI_DESCRIPTION, MBTI_DATA
+from app.data_init import df_traits_priority_for_professions
 from app.utils import (
     read_csv_file,
     apply_rounding_and_rename_columns,
     preprocess_scores_df,
     get_language_settings,
+    extract_profession_weights,
 )
 from app.components import (
     html_message,
@@ -267,6 +269,7 @@ def event_handler_calculate_practical_task_blocks(
 
                 df_hidden = df.drop(
                     columns=config_data.Settings_SHORT_PROFESSIONAL_SKILLS
+                    + config_data.Settings_DROPDOWN_MBTI_DEL_COLS
                     + config_data.Settings_DROPDOWN_MBTI_DEL_COLS_WEBCAM
                 )
 
@@ -350,117 +353,226 @@ def event_handler_calculate_practical_task_blocks(
 
         return existing_tuple[:-1] + person_metadata + existing_tuple[-1:]
     elif practical_subtasks.lower() == "professional groups":
-        sum_weights = sum(
-            [
-                number_openness,
-                number_conscientiousness,
-                number_extraversion,
-                number_agreeableness,
-                number_non_neuroticism,
-            ]
-        )
-
-        if sum_weights != 100:
-            gr.Warning(config_data.InformationMessages_SUM_WEIGHTS.format(sum_weights))
-
-            return (
-                gr.Row(visible=False),
-                gr.Column(visible=False),
-                dataframe(visible=False),
-                files_create_ui(
-                    None,
-                    "single",
-                    [".csv"],
-                    config_data.OtherMessages_EXPORT_PS,
-                    True,
-                    False,
-                    False,
-                    "csv-container",
-                ),
-                gr.Accordion(visible=False),
-                gr.HTML(visible=False),
-                dataframe(visible=False),
-                gr.Column(visible=False),
-                video_create_ui(visible=False),
-                gr.Column(visible=False),
-                gr.Row(visible=False),
-                gr.Row(visible=False),
-                gr.Image(visible=False),
-                textbox_create_ui(visible=False),
-                gr.Row(visible=False),
-                gr.Image(visible=False),
-                textbox_create_ui(visible=False),
-                gr.Row(visible=False),
-                gr.Row(visible=False),
-                gr.Image(visible=False),
-                textbox_create_ui(visible=False),
-                gr.Row(visible=False),
-                gr.Image(visible=False),
-                textbox_create_ui(visible=False),
-                html_message(
-                    config_data.InformationMessages_SUM_WEIGHTS.format(sum_weights),
-                    False,
-                    True,
-                ),
-            )
-        else:
-            b5._candidate_ranking(
-                df_files=pt_scores.iloc[:, 1:],
-                weigths_openness=number_openness,
-                weigths_conscientiousness=number_conscientiousness,
-                weigths_extraversion=number_extraversion,
-                weigths_agreeableness=number_agreeableness,
-                weigths_non_neuroticism=number_non_neuroticism,
-                out=False,
+        if type_modes == config_data.Settings_TYPE_MODES[0]:
+            sum_weights = sum(
+                [
+                    number_openness,
+                    number_conscientiousness,
+                    number_extraversion,
+                    number_agreeableness,
+                    number_non_neuroticism,
+                ]
             )
 
-            df = apply_rounding_and_rename_columns(b5.df_files_ranking_)
+            if sum_weights != 100:
+                gr.Warning(
+                    config_data.InformationMessages_SUM_WEIGHTS.format(sum_weights)
+                )
 
-            df_hidden = df.drop(columns=config_data.Settings_SHORT_PROFESSIONAL_SKILLS)
+                return (
+                    gr.Row(visible=False),
+                    gr.Column(visible=False),
+                    dataframe(visible=False),
+                    files_create_ui(
+                        None,
+                        "single",
+                        [".csv"],
+                        config_data.OtherMessages_EXPORT_PS,
+                        True,
+                        False,
+                        False,
+                        "csv-container",
+                    ),
+                    gr.Accordion(visible=False),
+                    gr.HTML(visible=False),
+                    dataframe(visible=False),
+                    gr.Column(visible=False),
+                    video_create_ui(visible=False),
+                    gr.Column(visible=False),
+                    gr.Row(visible=False),
+                    gr.Row(visible=False),
+                    gr.Image(visible=False),
+                    textbox_create_ui(visible=False),
+                    gr.Row(visible=False),
+                    gr.Image(visible=False),
+                    textbox_create_ui(visible=False),
+                    gr.Row(visible=False),
+                    gr.Row(visible=False),
+                    gr.Image(visible=False),
+                    textbox_create_ui(visible=False),
+                    gr.Row(visible=False),
+                    gr.Image(visible=False),
+                    textbox_create_ui(visible=False),
+                    html_message(
+                        config_data.InformationMessages_SUM_WEIGHTS.format(sum_weights),
+                        False,
+                        True,
+                    ),
+                )
+            else:
+                b5._candidate_ranking(
+                    df_files=pt_scores.iloc[:, 1:],
+                    weigths_openness=number_openness,
+                    weigths_conscientiousness=number_conscientiousness,
+                    weigths_extraversion=number_extraversion,
+                    weigths_agreeableness=number_agreeableness,
+                    weigths_non_neuroticism=number_non_neuroticism,
+                    out=False,
+                )
 
-            df_hidden.to_csv(config_data.Filenames_POTENTIAL_CANDIDATES)
+                df = apply_rounding_and_rename_columns(b5.df_files_ranking_)
 
-            df_hidden.reset_index(inplace=True)
+                df_hidden = df.drop(
+                    columns=config_data.Settings_SHORT_PROFESSIONAL_SKILLS
+                )
 
-            person_id = (
-                int(df_hidden.iloc[0][config_data.Dataframes_PT_SCORES[0][0]]) - 1
-            )
+                df_hidden.to_csv(config_data.Filenames_POTENTIAL_CANDIDATES)
 
-            person_metadata = create_person_metadata(person_id, files, video_metadata)
+                df_hidden.reset_index(inplace=True)
 
-            existing_tuple = (
-                gr.Row(visible=True),
-                gr.Column(visible=True),
-                dataframe(
-                    headers=df_hidden.columns.tolist(),
-                    values=df_hidden.values.tolist(),
-                    visible=True,
-                ),
-                files_create_ui(
-                    config_data.Filenames_POTENTIAL_CANDIDATES,
-                    "single",
-                    [".csv"],
-                    config_data.OtherMessages_EXPORT_PG,
-                    True,
-                    False,
-                    True,
-                    "csv-container",
-                ),
-                gr.Accordion(visible=False),
-                gr.HTML(visible=False),
-                dataframe(visible=False),
-                gr.Column(visible=True),
-                video_create_ui(
-                    value=files[person_id],
-                    file_name=Path(files[person_id]).name,
-                    label="Best Person ID - " + str(person_id + 1),
-                    visible=True,
-                    elem_classes="video-sorted-container",
-                ),
-                html_message(config_data.InformationMessages_NOTI_IN_DEV, False, False),
-            )
+                person_id = (
+                    int(df_hidden.iloc[0][config_data.Dataframes_PT_SCORES[0][0]]) - 1
+                )
+
+                person_metadata = create_person_metadata(
+                    person_id, files, video_metadata
+                )
+        elif type_modes == config_data.Settings_TYPE_MODES[1]:
+            all_hidden_dfs = []
+
+            for dropdown_candidate in config_data.Settings_DROPDOWN_CANDIDATES[:-1]:
+                weights, _ = extract_profession_weights(
+                    df_traits_priority_for_professions,
+                    dropdown_candidate,
+                )
+
+                sum_weights = sum(weights)
+
+                if sum_weights != 100:
+                    gr.Warning(
+                        config_data.InformationMessages_SUM_WEIGHTS.format(sum_weights)
+                    )
+
+                    return (
+                        gr.Row(visible=False),
+                        gr.Column(visible=False),
+                        dataframe(visible=False),
+                        files_create_ui(
+                            None,
+                            "single",
+                            [".csv"],
+                            config_data.OtherMessages_EXPORT_PS,
+                            True,
+                            False,
+                            False,
+                            "csv-container",
+                        ),
+                        gr.Accordion(visible=False),
+                        gr.HTML(visible=False),
+                        dataframe(visible=False),
+                        gr.Column(visible=False),
+                        video_create_ui(visible=False),
+                        gr.Column(visible=False),
+                        gr.Row(visible=False),
+                        gr.Row(visible=False),
+                        gr.Image(visible=False),
+                        textbox_create_ui(visible=False),
+                        gr.Row(visible=False),
+                        gr.Image(visible=False),
+                        textbox_create_ui(visible=False),
+                        gr.Row(visible=False),
+                        gr.Row(visible=False),
+                        gr.Image(visible=False),
+                        textbox_create_ui(visible=False),
+                        gr.Row(visible=False),
+                        gr.Image(visible=False),
+                        textbox_create_ui(visible=False),
+                        html_message(
+                            config_data.InformationMessages_SUM_WEIGHTS.format(
+                                sum_weights
+                            ),
+                            False,
+                            True,
+                        ),
+                    )
+                else:
+                    b5._candidate_ranking(
+                        df_files=pt_scores.iloc[:, 1:],
+                        weigths_openness=weights[0],
+                        weigths_conscientiousness=weights[1],
+                        weigths_extraversion=weights[2],
+                        weigths_agreeableness=weights[3],
+                        weigths_non_neuroticism=weights[4],
+                        out=False,
+                    )
+
+                    df = apply_rounding_and_rename_columns(b5.df_files_ranking_)
+
+                    df_hidden = df.drop(
+                        columns=config_data.Settings_SHORT_PROFESSIONAL_SKILLS
+                        + config_data.Settings_DROPDOWN_MBTI_DEL_COLS_WEBCAM
+                    )
+
+                    df_hidden.insert(0, "Professional Group", dropdown_candidate)
+
+                    all_hidden_dfs.append(df_hidden)
+
+                df_hidden = pd.concat(all_hidden_dfs, ignore_index=True)
+
+                df_hidden.rename(
+                    columns={
+                        "Candidate score": "Summary Score",
+                    },
+                    inplace=True,
+                )
+
+                df_hidden = df_hidden.sort_values(by="Summary Score", ascending=False)
+
+                df_hidden.reset_index(drop=True, inplace=True)
 
-            return existing_tuple[:-1] + person_metadata + existing_tuple[-1:]
+                df_hidden.to_csv(
+                    config_data.Filenames_POTENTIAL_CANDIDATES, index=False
+                )
+
+                person_id = 0
+
+                person_metadata = create_person_metadata(
+                    person_id, files, video_metadata
+                )
+
+        existing_tuple = (
+            gr.Row(visible=True),
+            gr.Column(visible=True),
+            dataframe(
+                headers=df_hidden.columns.tolist(),
+                values=df_hidden.values.tolist(),
+                visible=True,
+            ),
+            files_create_ui(
+                config_data.Filenames_POTENTIAL_CANDIDATES,
+                "single",
+                [".csv"],
+                config_data.OtherMessages_EXPORT_PG,
+                True,
+                False,
+                True,
+                "csv-container",
+            ),
+            gr.Accordion(visible=False),
+            gr.HTML(visible=False),
+            dataframe(visible=False),
+            gr.Column(visible=True),
+            video_create_ui(
+                value=files[person_id],
+                file_name=Path(files[person_id]).name,
+                label="Best Person ID - " + str(person_id + 1),
+                visible=True,
+                elem_classes="video-sorted-container",
+            ),
+            html_message(config_data.InformationMessages_NOTI_IN_DEV, False, False),
+        )
+
+        return existing_tuple[:-1] + person_metadata + existing_tuple[-1:]
     elif practical_subtasks.lower() == "professional skills":
         df_professional_skills = read_csv_file(config_data.Links_PROFESSIONAL_SKILLS)
 

app/event_handlers/dropdown_candidates.py

@@ -7,16 +7,13 @@ License: MIT License
 
 # Importing necessary components for the Gradio app
 from app.config import config_data
-from app.utils import read_csv_file, extract_profession_weights
+from app.utils import extract_profession_weights
+from app.data_init import df_traits_priority_for_professions
 from app.components import number_create_ui
 
 
 def event_handler_dropdown_candidates(practical_subtasks, dropdown_candidates):
     if practical_subtasks.lower() == "professional groups":
-        df_traits_priority_for_professions = read_csv_file(
-            config_data.Links_PROFESSIONS
-        )
-
         weights, interactive = extract_profession_weights(
             df_traits_priority_for_professions,
             dropdown_candidates,

app/event_handlers/practical_subtasks.py

@@ -9,7 +9,8 @@ import gradio as gr
 
 # Importing necessary components for the Gradio app
 from app.config import config_data
-from app.utils import read_csv_file, extract_profession_weights, get_language_settings
+from app.utils import read_csv_file, get_language_settings
+from app.data_init import weights_professions, interactive_professions
 from app.components import number_create_ui, dropdown_create_ui
 
 
@@ -72,14 +73,6 @@ def event_handler_practical_subtasks(
             number_create_ui(visible=False),
         )
     elif practical_subtasks.lower() == "professional groups":
-        df_traits_priority_for_professions = read_csv_file(
-            config_data.Links_PROFESSIONS
-        )
-        weights_professions, interactive_professions = extract_profession_weights(
-            df_traits_priority_for_professions,
-            config_data.Settings_DROPDOWN_CANDIDATES[0],
-        )
-
         return (
             practical_subtasks_selected,
             gr.Column(visible=visible_subtasks),

app/tabs.py

@@ -13,10 +13,11 @@ from app.description_steps import STEP_1, STEP_2
 from app.mbti_description import MBTI_DESCRIPTION, MBTI_DATA
 from app.app import APP
 from app.authors import AUTHORS
+from app.data_init import weights_professions, interactive_professions
 from app.requirements_app import read_requirements_to_df
 from app.config import config_data
 from app.practical_tasks import supported_practical_tasks
-from app.utils import read_csv_file, extract_profession_weights
+from app.utils import read_csv_file
 from app.components import (
     html_message,
     files_create_ui,
@@ -386,14 +387,6 @@ def app_tab():
                 elem_classes="dropdown-container",
             )
 
-            df_traits_priority_for_professions = read_csv_file(
-                config_data.Links_PROFESSIONS
-            )
-            weights_professions, interactive_professions = extract_profession_weights(
-                df_traits_priority_for_professions,
-                config_data.Settings_DROPDOWN_CANDIDATES[0],
-            )
-
             number_openness = number_create_ui(
                 value=weights_professions[0],
                 minimum=config_data.Values_0_100[0],

config.toml

@@ -1,5 +1,5 @@
 [AppSettings]
-APP_VERSION = "0.10.2"
+APP_VERSION = "0.10.4"
 SERVER_NAME = "127.0.0.1"
 PORT = 7860
 CSS_PATH = "app.css"
@@ -148,7 +148,7 @@ DROPDOWN_MBTI = [
     "The Commander (ENTJ): Construction Supervisor, Health Services Administrator, Financial Accountant, Auditor, Lawyer, School Principal, Chemical Engineer, Database Manager, etc.",
 ]
 DROPDOWN_MBTI_DEL_COLS = ["EI", "SN", "TF", "JP", "Match"]
-DROPDOWN_MBTI_DEL_COLS_WEBCAM = ["EI", "SN", "TF", "JP", "Match", "Path"]
+DROPDOWN_MBTI_DEL_COLS_WEBCAM = ["Path"]
 SHOW_VIDEO_METADATA = true
 SUPPORTED_VIDEO_EXT = ["mp4", "mov", "avi", "flv"]
 TYPE_MODES = ["Files", "Web"]

requirements.txt

@@ -1,4 +1,4 @@
-gradio==5.6.0
+gradio==5.7.1
 PyYAML==6.0.2
 toml==0.10.2
 oceanai==1.0.0a46