Adding FAQs, moving submission and utils code around

about.py

@@ -21,4 +21,29 @@ CACHE_PATH=os.getenv("HF_HOME", ".")
 API = HfApi(token=TOKEN)
 organization="ginkgo-datapoints"
 submissions_repo = f'{organization}/abdev-bench-submissions'
-results_repo = f'{organization}/abdev-bench-results'
+results_repo = f'{organization}/abdev-bench-results'
+
+ABOUT_TEXT = """
+## About this challenge
+
+We're inviting the ML/bio community to predict developability properties for 244 antibodies from the [GDPa1 dataset](https://huggingface.co/datasets/ginkgo-datapoints/GDPa1).
+
+**What is antibody developability?**
+
+Antibodies have to be manufacturable, stable in high concentrations, and have low off-target effects. 
+Properties such as these can often hinder the progression of an antibody to the clinic, and are collectively referred to as 'developability'.
+Here we show 5 of these properties and invite the community to submit and develop better predictors, which will be tested out on a heldout private set to assess model generalization.
+
+**How to submit?**
+
+TODO
+
+**How to evaluate?**
+
+TODO
+
+FAQs: A list of frequently asked questions.
+"""
+FAQS = {
+    "Example FAQ with dropdown": """Full answer to this question""",
+}

app.py

@@ -5,57 +5,10 @@ import pandas as pd
 import gradio as gr
 from gradio_leaderboard import Leaderboard
 
-from utils import read_submission_from_hub, write_results
-from about import ASSAY_LIST, ASSAY_RENAME, ASSAY_EMOJIS, submissions_repo, API, results_repo
-from typing import BinaryIO, Literal
-from datetime import datetime
-import tempfile
-from datasets import load_dataset
-import io
-
-def make_submission(
-    submitted_file: BinaryIO,
-    user_state):
-
-    if user_state is None:
-        raise gr.Error("You must submit your username to submit a file.")
-    
-    file_path = submitted_file.name
-
-    if not file_path:
-        raise gr.Error("Uploaded file object does not have a valid file path.")
-    
-    path_obj = Path(file_path)
-    timestamp = datetime.utcnow().isoformat()
-
-    with (path_obj.open("rb") as f_in):
-        file_content = f_in.read().decode("utf-8")
-
-        # write to dataset
-        filename = f"{user_state}/{timestamp.replace(':', '-')}_{user_state}.json"
-        record = {
-            "submission_filename": filename,
-            "submission_time": timestamp,
-            "csv_content": file_content,
-            "evaluated": False,
-            "user": user_state,
-        }
-        with tempfile.NamedTemporaryFile(mode="w", suffix=".json", delete=False) as tmp:
-            json.dump(record, tmp, indent=2)
-            tmp.flush()
-            tmp_name = tmp.name
-
-        API.upload_file(
-            path_or_fileobj=tmp_name, 
-            path_in_repo=filename,
-            repo_id=submissions_repo,
-            repo_type="dataset",
-            commit_message=f"Add submission for {user_state} at {timestamp}"
-        )
-        Path(tmp_name).unlink()
-
-    return "✅ Your submission has been received! Sit tight and your scores will appear on the leaderboard shortly."
-
+from utils import fetch_hf_results, show_output_box
+from about import ASSAY_LIST, ASSAY_RENAME, ASSAY_EMOJIS, submissions_repo, API
+from submit import make_submission
+from about import ABOUT_TEXT, FAQS
     
 def get_leaderboard_table(df_results: pd.DataFrame, assay: str | None = None):
     # ds = load_dataset(results_repo, split='train', download_mode="force_redownload")
@@ -92,16 +45,6 @@ def get_leaderboard_object(df_results: pd.DataFrame, assay: str | None = None):
         render=True
     )
 
-def show_output_box(message):
-    return gr.update(value=message, visible=True)
-
-def fetch_hf_results():
-    ds = load_dataset(results_repo, split='no_low_spearman', download_mode="force_redownload")
-    df = pd.DataFrame(ds).drop_duplicates(subset=["model", "assay"])
-    df["property"] = df["assay"].map(ASSAY_RENAME)
-    print(df.head())
-    return df
-
 with gr.Blocks() as demo:
     gr.Markdown("""
         ## Welcome to the Ginkgo Antibody Developability Benchmark!
@@ -148,71 +91,53 @@ with gr.Blocks() as demo:
                 elem_classes=["resized-image"],
                 show_download_button=False,
             )
+            gr.Markdown(ABOUT_TEXT)
+            for question, answer in FAQS.items():
+                gr.Accordion(question, answer)
+
+        with gr.TabItem("✉️ Submit", elem_id="boundary-benchmark-tab-table"):
             gr.Markdown(
                 """
-                ## About this challenge
-                
-                We're inviting the ML/bio community to predict developability properties for 244 antibodies from the [GDPa1 dataset](https://huggingface.co/datasets/ginkgo-datapoints/GDPa1).
-                
-                **What is antibody developability?**
-                
-                Antibodies have to be manufacturable, stable in high concentrations, and have low off-target effects. 
-                Properties such as these can often hinder the progression of an antibody to the clinic, and are collectively referred to as 'developability'.
-                Here we show 5 of these properties and invite the community to submit and develop better predictors, which will be tested out on a heldout private set to assess model generalization.
-                
-                **How to submit?**
-                
-                TODO
-                
-                **How to evaluate?**
-                
-                TODO
+            # Antibody Developability Submission
+            Upload a CSV to get a score!
             """
             )
-
-        with gr.TabItem("✉️ Submit", elem_id="boundary-benchmark-tab-table"):
-                gr.Markdown(
-                    """
-                # Antibody Developability Submission
-                Upload a CSV to get a score!
-                """
-                )
-                filename = gr.State(value=None) 
-                eval_state = gr.State(value=None) 
-                user_state = gr.State(value=None)
-
-                # gr.LoginButton()
-
-                with gr.Row():
-                    with gr.Column():
-                        username_input = gr.Textbox(
-                            label="Username", 
-                            placeholder="Enter your Hugging Face username",
-                            info="This will be displayed on the leaderboard."
-                        )
-                    with gr.Column():
-                        boundary_file = gr.File(label="Submission CSV")
-
-                username_input.change(
-                    fn=lambda x: x if x.strip() else None,
-                    inputs=username_input,
-                    outputs=user_state
-                )                       
-
-                submit_btn = gr.Button("Evaluate")
-                message = gr.Textbox(label="Status", lines=1, visible=False)
-                # help message
-                gr.Markdown("If you have issues with submission or using the leaderboard, please start a discussion in the Community tab of this Space.")
-                
-                submit_btn.click(
-                    make_submission,
-                    inputs=[boundary_file, user_state],
-                    outputs=[message],
-                ).then(
-                    fn=show_output_box,
-                    inputs=[message],
-                    outputs=[message],
-                )
+            filename = gr.State(value=None) 
+            eval_state = gr.State(value=None) 
+            user_state = gr.State(value=None)
+
+            # gr.LoginButton()
+
+            with gr.Row():
+                with gr.Column():
+                    username_input = gr.Textbox(
+                        label="Username", 
+                        placeholder="Enter your Hugging Face username",
+                        info="This will be displayed on the leaderboard."
+                    )
+                with gr.Column():
+                    boundary_file = gr.File(label="Submission CSV")
+
+            username_input.change(
+                fn=lambda x: x if x.strip() else None,
+                inputs=username_input,
+                outputs=user_state
+            )                       
+
+            submit_btn = gr.Button("Evaluate")
+            message = gr.Textbox(label="Status", lines=1, visible=False)
+            # help message
+            gr.Markdown("If you have issues with submission or using the leaderboard, please start a discussion in the Community tab of this Space.")
+            
+            submit_btn.click(
+                make_submission,
+                inputs=[boundary_file, user_state],
+                outputs=[message],
+            ).then(
+                fn=show_output_box,
+                inputs=[message],
+                outputs=[message],
+            )
 
 
 if __name__ == "__main__":

submit.py

@@ -1,58 +1,37 @@
-import pathlib
 from pathlib import Path
 import tempfile
-from typing import BinaryIO, Literal
+from typing import BinaryIO
 import json
-import pandas as pd
 
 import gradio as gr
-from datasets import load_dataset, Dataset
-from huggingface_hub import upload_file, hf_hub_download
-from gradio_leaderboard import ColumnFilter, Leaderboard, SelectColumns
-from evaluation import evaluate_problem
 from datetime import datetime
-import os
-from huggingface_hub import HfApi
 
-from about import PROBLEM_TYPES, TOKEN, CACHE_PATH, API, submissions_repo, results_repo
+from about import API, submissions_repo
+
+def make_submission(
+    submitted_file: BinaryIO,
+    user_state):
 
-def submit_boundary(
-    problem_type: Literal["geometrical", "simple_to_build", "mhd_stable"],
-    boundary_file: BinaryIO,
-    user_state
-) -> str:
-    
-    # profile: gr.OAuthProfile | None
-    # user_state = profile.username
-    
-    # error handling
-    # if profile.username is None:
     if user_state is None:
         raise gr.Error("You must submit your username to submit a file.")
     
-    file_path = boundary_file.name
+    file_path = submitted_file.name
 
     if not file_path:
         raise gr.Error("Uploaded file object does not have a valid file path.")
     
-    path_obj = pathlib.Path(file_path)
+    path_obj = Path(file_path)
     timestamp = datetime.utcnow().isoformat()
 
-    with (
-        path_obj.open("rb") as f_in,
-        tempfile.NamedTemporaryFile(delete=False, suffix=".json") as tmp_boundary,
-    ):
-        file_content = f_in.read()
-        tmp_boundary.write(file_content)
-        tmp_boundary_path = pathlib.Path(tmp_boundary.name)
+    with (path_obj.open("rb") as f_in):
+        file_content = f_in.read().decode("utf-8")
 
         # write to dataset
-        filename = f"{problem_type}/{timestamp.replace(':', '-')}_{problem_type}.json"
+        filename = f"{user_state}/{timestamp.replace(':', '-')}_{user_state}.json"
         record = {
             "submission_filename": filename,
             "submission_time": timestamp,
-            "problem_type": problem_type,
-            "boundary_json": file_content.decode("utf-8"),
+            "csv_content": file_content,
             "evaluated": False,
             "user": user_state,
         }
@@ -66,10 +45,8 @@ def submit_boundary(
             path_in_repo=filename,
             repo_id=submissions_repo,
             repo_type="dataset",
-            commit_message=f"Add submission for {problem_type} at {timestamp}"
+            commit_message=f"Add submission for {user_state} at {timestamp}"
         )
-        pathlib.Path(tmp_name).unlink()
-
-    tmp_boundary_path.unlink()
+        Path(tmp_name).unlink()
 
-    return "✅ Your submission has been received! Sit tight and your scores will appear on the leaderboard shortly.", filename
+    return "✅ Your submission has been received! Sit tight and your scores will appear on the leaderboard shortly."

utils.py

@@ -3,9 +3,11 @@ import tempfile
 import json
 
 import gradio as gr
+import pandas as pd
+from datasets import load_dataset
 from huggingface_hub import hf_hub_download
 
-from about import API, submissions_repo, results_repo
+from about import API, submissions_repo, results_repo, ASSAY_RENAME
 
 # def make_user_clickable(name):
 #     link =f'https://huggingface.co/{name}'
@@ -15,6 +17,16 @@ from about import API, submissions_repo, results_repo
 #     link =f'https://huggingface.co/datasets/proxima-fusion/constellaration-bench-results/blob/main/{filename}'
 #     return f'<a target="_blank" href="{link}" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">link</a>'
 
+def show_output_box(message):
+    return gr.update(value=message, visible=True)
+
+def fetch_hf_results():
+    ds = load_dataset(results_repo, split='no_low_spearman', download_mode="force_redownload")
+    df = pd.DataFrame(ds).drop_duplicates(subset=["model", "assay"])
+    df["property"] = df["assay"].map(ASSAY_RENAME)
+    print(df.head())
+    return df
+
 def read_result_from_hub(filename):
     local_path = hf_hub_download(
         repo_id=results_repo,