Update gradio version and fix error logging

README.md

@@ -4,7 +4,7 @@ emoji: 📚
 colorFrom: green
 colorTo: blue
 sdk: gradio
-sdk_version: 3.41.0
+sdk_version: 4.38.1
 app_file: app.py
 pinned: false
 license: mit

app.py

@@ -52,6 +52,7 @@ our funders that it is being used. Thank you!
 # """
 
 article = """
+Pairs file should be a comma-separated or tab-separated (.csv/.tsv) file with two columns, "protein1" and "protein2", where each row contains the names of two proteins. The sequences should be a FASTA file with the corresponding protein names as the headers.
 
 Note that running here with the "TT3D" model does not run structure prediction on the sequences, but rather uses the [ProstT5](https://github.com/mheinzinger/ProstT5) language model to
 translate amino acid to 3di sequences. This is much faster than running structure prediction, but the results may not be as accurate.
@@ -95,7 +96,8 @@ def predict(model_name, pairs_file, sequence_file, progress = gr.Progress()):
         # gr.Info("Loading files...")
         try:
             seqs = SeqIO.to_dict(SeqIO.parse(sequence_file.name, "fasta"))
-        except ValueError as _:
+        except ValueError as e:
+            print(e)
             raise gr.Error("Invalid FASTA file - duplicate entry")
 
         if Path(pairs_file.name).suffix == ".csv":
@@ -104,7 +106,8 @@ def predict(model_name, pairs_file, sequence_file, progress = gr.Progress()):
             pairs = pd.read_csv(pairs_file.name, sep="\t")
         try:
             pairs.columns = ["protein1", "protein2"]
-        except ValueError as _:
+        except ValueError as e:
+            print(e)
             raise gr.Error("Invalid pairs file - must have two columns 'protein1' and 'protein2'")
 
         do_foldseek = False
@@ -134,6 +137,7 @@ def predict(model_name, pairs_file, sequence_file, progress = gr.Progress()):
             #     print(foldseek_embeddings[k])
             #     print(foldseek_embeddings[k].shape)
 
+        print("Starting predictions")
         progress(0, desc="Starting...")
         results = []
         for i in progress.tqdm(range(len(pairs))):
@@ -165,7 +169,8 @@ def predict(model_name, pairs_file, sequence_file, progress = gr.Progress()):
         return results, file_path
     
     except Exception as e:
-        gr.Error(e)
+        print(e)
+        raise gr.Error(e)
         return None, None
 
 demo = gr.Interface(
@@ -177,13 +182,13 @@ demo = gr.Interface(
     ],
     outputs = [
         gr.DataFrame(label='Results', headers=['Protein 1', 'Protein 2', 'Interaction']),
-        gr.File(label="Download results", type="file")
+        gr.File(label="Download results", type="filepath")
     ],
-    # title = title,
-    # description = description,
+    title = title,
+    description = description,
     article = article,
     theme = theme,
 )
 
 if __name__ == "__main__":
-    demo.queue(max_size=20).launch()
+    demo.queue(max_size=20).launch()