fixes

app.py

@@ -17,8 +17,8 @@ api = HfApi()
 paper_list = PaperList()
 
 path = api.hf_hub_download(repo_id=REPO_ID, filename=FILENAME, repo_type="dataset")
-df = pd.read_csv(path)
-paper_id_to_index = {str(row["id"]): i for i, row in df.iterrows()}
+actual_df = pd.read_csv(path)
+paper_id_to_index = {str(row["id"]): i for i, row in actual_df.iterrows()}
 
 
 with gr.Blocks() as demo_search:
@@ -65,17 +65,23 @@ def load_data(paper_id: str) -> tuple[str, str, str, str, str, str, str, str, st
         index = paper_id_to_index[paper_id]
     except KeyError:
         raise gr.Error(f"Paper ID {paper_id} not found.")
+    
+    paper = actual_df.iloc[index]
 
-    paper = raw_data[index]
     return (
         paper["id"],
         paper["title"],
-        "\n".join(paper["authors"]),
+        paper["authors"],
         paper["arxiv_id"],
-        "\n".join(paper["GitHub"]),
-        "\n".join(paper["Space"]),
-        "\n".join(paper["Model"]),
-        "\n".join(paper["Dataset"]),
+         "\n".join([PaperList.create_link("GitHub", url) for url in paper["GitHub"]] if paper["GitHub"]!="[]" else " "),
+        "\n".join([PaperList.create_link(repo_id, f"https://huggingface.co/spaces/{repo_id}")
+                    for repo_id in paper["Space"]
+                    ] if paper["Space"] != "[]" else [" "]),
+        "\n".join([PaperList.create_link(repo_id, f"https://huggingface.co/{repo_id}") for repo_id in paper["Model"]]
+                    if paper["Model"] != "[]" else [" "]),
+        "\n".join([PaperList.create_link(repo_id, f"https://huggingface.co/datasets/{repo_id}") for repo_id in paper["Dataset"]
+                    ] if paper["Dataset"] != "[]" else [" "]
+                )
     )
 
 
@@ -105,12 +111,12 @@ def create_pr(
 
     data = copy.deepcopy(df)
     data[index]["title"] = title.strip()
-    data[index]["authors"] = split_and_strip(authors)
+    data[index]["authors"] = authors
     data[index]["arxiv_id"] = arxiv_id.strip()
-    data[index]["GitHub"] = split_and_strip(github_links)
-    data[index]["Space"] = split_and_strip(space_ids)
-    data[index]["Model"] = split_and_strip(model_ids)
-    data[index]["Dataset"] = split_and_strip(dataset_ids)
+    data[index]["GitHub"] = github_links
+    data[index]["Space"] = space_ids
+    data[index]["Model"] = model_ids
+    data[index]["Dataset"] = dataset_ids
     with tempfile.NamedTemporaryFile(mode="w", delete=False) as f:
         data.to_csv(f)
 
@@ -201,4 +207,4 @@ with gr.Blocks(css="style.css") as demo:
             demo_edit.render()
 
 if __name__ == "__main__":
-    demo.queue(api_open=False).launch(show_api=False)
+    demo.queue(api_open=False).launch(show_api=False, debug=True)

papers.py

@@ -3,6 +3,7 @@ import operator
 import datasets
 import pandas as pd
 from huggingface_hub import HfApi
+from math import isnan
 
 api = HfApi()
 
@@ -26,9 +27,8 @@ class PaperList:
     @staticmethod
     def get_df() -> pd.DataFrame:
         df = datasets.load_dataset("CVPR2024/CVPR2024-papers", split="train").to_pandas()
-        df["authors_str"] = df["authors"].apply(lambda x: ", ".join(x))
         df["paper_page"] = df["arxiv_id"].apply(
-            lambda arxiv_id: f"https://huggingface.co/papers/{arxiv_id}" if arxiv_id else ""
+            lambda arxiv_id: f"https://huggingface.co/papers/{arxiv_id}" if not isnan(arxiv_id) else ""
         )
         return df
 
@@ -43,23 +43,22 @@ class PaperList:
             new_row = {
                 "ID": row["id"],
                 "Title": row["title"],
-                "Authors": ", ".join(row["authors"]),
-                "Paper page": PaperList.create_link(row["arxiv_id"], row["paper_page"]),
-                "GitHub": "\n".join([PaperList.create_link("GitHub", url) for url in row["GitHub"]]),
+                "Authors": row["authors"],
+                "Paper page": PaperList.create_link(row["arxiv_id"], row["paper_page"]) if not isnan(row["arxiv_id"]) else  " ",
+                "GitHub": "\n".join([PaperList.create_link("GitHub", url) for url in row["GitHub"]] if row["GitHub"]!="[]" else " "),
                 "Spaces": "\n".join(
                     [
                         PaperList.create_link(repo_id, f"https://huggingface.co/spaces/{repo_id}")
                         for repo_id in row["Space"]
-                    ]
-                ),
+                    ] if row["Space"] != "[]" else [" "]),
                 "Models": "\n".join(
                     [PaperList.create_link(repo_id, f"https://huggingface.co/{repo_id}") for repo_id in row["Model"]]
-                ),
+                    if row["Model"] != "[]" else [" "]) ,
                 "Datasets": "\n".join(
                     [
                         PaperList.create_link(repo_id, f"https://huggingface.co/datasets/{repo_id}")
                         for repo_id in row["Dataset"]
-                    ]
+                    ] if row["Dataset"] != "[]" else [" "]
                 ),
             }
             rows.append(new_row)
@@ -80,5 +79,5 @@ class PaperList:
     ) -> pd.DataFrame:
         df = self.df_raw.copy()
         df = df[df["title"].str.contains(title_search_query, case=False)]
-        df = df[df["authors_str"].str.contains(author_search_query, case=False)]
+        df = df[df["authors"].str.contains(author_search_query, case=False)]
         return self.prettify(df)