mv constants to constants.py

about.py

@@ -1,34 +1,3 @@
-import os
-from huggingface_hub import HfApi
-
-ASSAY_LIST = ["AC-SINS_pH7.4", "PR_CHO", "HIC", "Tm2", "Titer"]
-ASSAY_RENAME = {
-    "AC-SINS_pH7.4": "Self-association",
-    "PR_CHO": "Polyreactivity",
-    "HIC": "Hydrophobicity",
-    "Tm2": "Thermostability",
-    "Titer": "Titer",
-}
-ASSAY_DESCRIPTION = {
-    "AC-SINS_pH7.4": "Self association by AC-SINS at pH 7.4",
-    "PR_CHO": "Polyreactivity by bead-based method against CHO SMP and ovalbumin",
-    "HIC": "Hydrophobicity by HIC",
-    "Tm2": "Thermostability by nanoDSF",
-    "Titer": "Titer by Valita",
-}
-ASSAY_EMOJIS = {
-    "AC-SINS_pH7.4": "🧲",
-    "PR_CHO": "🎯",
-    "HIC": "💧",
-    "Tm2": "🌡️",
-    "Titer": "🧪",
-}
-TOKEN = os.environ.get("HF_TOKEN")
-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'
 
 ABOUT_TEXT = """
 ## About this challenge

app.py

@@ -4,7 +4,8 @@ import gradio as gr
 from gradio_leaderboard import Leaderboard
 
 from utils import fetch_hf_results, show_output_box
-from about import ASSAY_LIST, ASSAY_RENAME, ASSAY_EMOJIS, ASSAY_DESCRIPTION, ABOUT_TEXT, FAQS
+from constants import ASSAY_LIST, ASSAY_RENAME, ASSAY_EMOJIS, ASSAY_DESCRIPTION
+from about import ABOUT_TEXT, FAQS
 from submit import make_submission
     
 def format_leaderboard_table(df_results: pd.DataFrame, assay: str | None = None):

constants.py

@@ -0,0 +1,49 @@
+"""
+Constants for the Antibody Developability Benchmark
+"""
+
+import os
+from huggingface_hub import HfApi
+
+
+ASSAY_LIST = ["AC-SINS_pH7.4", "PR_CHO", "HIC", "Tm2", "Titer"]
+ASSAY_RENAME = {
+    "AC-SINS_pH7.4": "Self-association",
+    "PR_CHO": "Polyreactivity",
+    "HIC": "Hydrophobicity",
+    "Tm2": "Thermostability",
+    "Titer": "Titer",
+}
+ASSAY_DESCRIPTION = {
+    "AC-SINS_pH7.4": "Self association by AC-SINS at pH 7.4",
+    "PR_CHO": "Polyreactivity by bead-based method against CHO SMP and ovalbumin",
+    "HIC": "Hydrophobicity by HIC",
+    "Tm2": "Thermostability by nanoDSF",
+    "Titer": "Titer by Valita",
+}
+ASSAY_EMOJIS = {
+    "AC-SINS_pH7.4": "🧲",
+    "PR_CHO": "🎯",
+    "HIC": "💧",
+    "Tm2": "🌡️",
+    "Titer": "🧪",
+}
+
+# Input CSV file requirements
+MINIMAL_NUMBER_OF_ROWS: int = 50
+REQUIRED_COLUMNS: list[str] = [
+    "antibody_id",
+    "antibody_name",
+    "vh_protein_sequence",
+    "vl_protein_sequence",
+]  + ASSAY_LIST
+
+# Huggingface API
+TOKEN = os.environ.get("HF_TOKEN")
+CACHE_PATH=os.getenv("HF_HOME", ".")
+API = HfApi(token=TOKEN)
+
+# Huggingface repos
+ORGANIZATION="ginkgo-datapoints"
+SUBMISSIONS_REPO = f'{ORGANIZATION}/abdev-bench-submissions'
+RESULTS_REPO = f'{ORGANIZATION}/abdev-bench-results'

utils.py

@@ -7,7 +7,7 @@ import pandas as pd
 from datasets import load_dataset
 from huggingface_hub import hf_hub_download
 
-from about import API, submissions_repo, results_repo, ASSAY_RENAME
+from constants import API, SUBMISSIONS_REPO, RESULTS_REPO, ASSAY_RENAME
 
 # def make_user_clickable(name):
 #     link =f'https://huggingface.co/{name}'
@@ -21,7 +21,7 @@ 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")
+    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())
@@ -29,7 +29,7 @@ def fetch_hf_results():
 
 def read_result_from_hub(filename):
     local_path = hf_hub_download(
-        repo_id=results_repo,
+        repo_id=RESULTS_REPO,
         repo_type="dataset",
         filename=filename,
     )
@@ -37,7 +37,7 @@ def read_result_from_hub(filename):
 
 def read_submission_from_hub(filename):
     local_path = hf_hub_download(
-        repo_id=submissions_repo,
+        repo_id=SUBMISSIONS_REPO,
         repo_type="dataset",
         filename=filename,
     )
@@ -65,7 +65,7 @@ def write_results(record, result):
     API.upload_file(
             path_or_fileobj=tmp_name, 
             path_in_repo=record['result_filename'],
-            repo_id=results_repo,
+            repo_id=RESULTS_REPO,
             repo_type="dataset",
             commit_message=f"Add result data for {record['result_filename']}"
         )