Update app.py

app.py

@@ -5,46 +5,38 @@ import gradio as gr
 import pandas as pd
 from huggingface_hub import HfApi, repocard
 
-def is_duplicated(space_id:str)->None:
-    card = repocard.RepoCard.load(space_id, repo_type="space")
-    return getattr(card.data, "duplicated_from", None) is not None
-
-
-
-def make_clickable_model(model_name, link=None):
-    if link is None:
-        link = "https://huggingface.co/" + "spaces/" + model_name
-    return f'<a target="_blank" href="{link}">{model_name.split("/")[-1]}</a>'
-
-def get_space_ids(category):
+def make_clickable_space(name, repo_type):
+    if repo_type == "spaces":
+        link = "https://huggingface.co/" + "spaces/" + name
+    elif repo_type == "models":
+        link = "https://huggingface.co/"  + name
+    else:
+        link = "https://huggingface.co/" + "datasets/" + name
+    return f'<a target="_blank" href="{link}">{name.split("/")[-1]}</a>'
+
+def get_repo_ids(repo_type):
     api = HfApi()
-    spaces = api.list_spaces(filter=["keras-dreambooth", category])
-    print(spaces)
-    space_ids = [x for x in spaces]
-    return space_ids
-
-
-def make_clickable_user(user_id):
-    link = "https://huggingface.co/" + user_id
-    return f'<a  target="_blank" href="{link}">{user_id}</a>'
-
-def get_submissions(category):
-    submissions = get_space_ids(category)
-    leaderboard_models = []
+    if repo_type == "spaces":
+        repos = api.list_spaces(filter=["hackathon-somos-nlp-2023"])
+    elif repo_type == "models":
+        repos = api.list_models(filter=["hackathon-somos-nlp-2023"])
+    else:
+        repos = api.list_datasets(filter=["hackathon-somos-nlp-2023"])
+    return repos
+
+def get_submissions(repo_type):
+    submissions = get_repo_ids(repo_type)
+    leaderboard = []
 
     for submission in submissions:
-        # user, model, likes
-        if not is_duplicated(submission.id):
-            user_id = submission.id.split("/")[0]
-            leaderboard_models.append(
-                (
-                    make_clickable_user(user_id),
-                    make_clickable_model(submission.id),
-                    submission.likes,
-                )
+        leaderboard.append(
+            (
+                make_clickable_model(submission.id),
+                submission.likes,
             )
+        )
 
-    df = pd.DataFrame(data=leaderboard_models, columns=["User", "Space", "Likes"])
+    df = pd.DataFrame(data=leaderboard, columns=[Repo", "Likes"])
     df.sort_values(by=["Likes"], ascending=False, inplace=True)
     df.insert(0, "Rank", list(range(1, len(df) + 1)))
     return df
@@ -53,62 +45,45 @@ block = gr.Blocks()
 
 with block:
     gr.Markdown(
-        """# Keras DreamBooth Leaderboard
-    
-    Welcome to the leaderboard for the Keras DreamBooth Event! This is a community event where participants **personalise a Stable Diffusion model** by fine-tuning it with a powerful technique called [_DreamBooth_](https://arxiv.org/abs/2208.12242). This technique allows one to implant a subject into the output domain of the model such that it can be synthesized with a _unique identifier_ in the prompt. 
-    
-    This competition is composed of 4 _themes_, where each theme will collect models belong to one of the categories shown in the tabs below. We'll be **giving out prizes to the top 3 most liked models per theme**, and you're encouraged to submit as many models as you want!
-    
+        """# Hackathon Somos NLP 2023 Leaderboard
     """
     )
     with gr.Tabs():
-        with gr.TabItem("Nature 🐨 🌳 "):
+        with gr.TabItem("Spaces (ML apps) 🐨 🌳 "):
             with gr.Row():
-                nature_data = gr.components.Dataframe(
-                    type="pandas", datatype=["number", "markdown", "markdown", "number"]
+                models_data = gr.components.Dataframe(
+                    type="pandas", datatype=["number", "markdown", "number"]
                 )
             with gr.Row():
                 data_run = gr.Button("Refresh")
                 data_run.click(
-                    get_submissions, inputs=gr.Variable("nature"), outputs=nature_data
+                    get_submissions, inputs=gr.Variable("spaces"), outputs=nature_data
                 )
-        with gr.TabItem("Science Fiction & Fantasy 🧙‍♀️ 🧛‍♀️ 🤖  "):
+        with gr.TabItem("Models"):
             with gr.Row():
                 scifi_data = gr.components.Dataframe(
-                    type="pandas", datatype=["number", "markdown", "markdown", "number"]
+                    type="pandas", datatype=["number", "markdown", "number"]
                 )
             with gr.Row():
                 data_run = gr.Button("Refresh")
                 data_run.click(
-                    get_submissions, inputs=gr.Variable("scifi"), outputs=scifi_data
+                    get_submissions, inputs=gr.Variable("models"), outputs=scifi_data
                 )
-        with gr.TabItem("Consentful 🖼️ 🎨 "):
+        with gr.TabItem("Datasets"):
             with gr.Row():
                 consentful_data = gr.components.Dataframe(
-                    type="pandas", datatype=["number", "markdown", "markdown", "number"]
+                    type="pandas", datatype=["number", "markdown", "number"]
                 )
             with gr.Row():
                 data_run = gr.Button("Refresh")
                 data_run.click(
-                    get_submissions, inputs=gr.Variable("consentful"), outputs=consentful_data
-                )
-        with gr.TabItem("Wild Card 🃏"):
-            with gr.Row():
-                wildcard_data = gr.components.Dataframe(
-                    type="pandas", datatype=["number", "markdown", "markdown", "number"]
-                )
-            with gr.Row():
-                data_run = gr.Button("Refresh")
-                data_run.click(
-                    get_submissions,
-                    inputs=gr.Variable("wildcard"),
-                    outputs=wildcard_data,
+                    get_submissions, inputs=gr.Variable("datasets"), outputs=consentful_data
                 )
 
-    block.load(get_submissions, inputs=gr.Variable("nature"), outputs=nature_data)
-    block.load(get_submissions, inputs=gr.Variable("scifi"), outputs=scifi_data)
-    block.load(get_submissions, inputs=gr.Variable("consentful"), outputs=consentful_data)
-    block.load(get_submissions, inputs=gr.Variable("wildcard"), outputs=wildcard_data)
+
+    block.load(get_submissions, inputs=gr.Variable("spaces"), outputs=spaces_data)
+    block.load(get_submissions, inputs=gr.Variable("models"), outputs=models_data)
+    block.load(get_submissions, inputs=gr.Variable("datasets"), outputs=datasets_data)
 
 
 block.launch()