Update app.py

app.py

@@ -1,3 +1,4 @@
+import re
 import gradio as gr
 from transformers import AutoTokenizer, AutoModelForTokenClassification, pipeline
 
@@ -16,10 +17,18 @@ def extract_coordinates(text):
     # Collect coordinate entities
     coords_text = [ent['word'] for ent in entities if ent['entity_group'] == 'Latlong']
     
-    if coords_text:
-        return "\n".join(coords_text)
-    else:
-        return "No coordinates found"
+    parsed_coords = []
+    for coord_str in coords_text:
+        matches = dms_pattern.findall(coord_str)
+        if len(matches) >= 2:  # Expecting lat + lon
+            lat_match, lon_match = matches[0], matches[1]
+            lat = dms_to_dd(*lat_match)
+            lon = dms_to_dd(*lon_match)
+            parsed_coords.append([lat, lon])
+
+    result_text = "\n".join(coords_text) if coords_text else "No coordinates found"
+    return result_text, parsed_coords if parsed_coords else None
+
 
 
 def norm_coordinates(text):
@@ -34,6 +43,19 @@ def norm_coordinates(text):
 
     return predicted_coordinates_from_pipeline[0]['generated_text']
 
+# Function to convert DMS to decimal
+def dms_to_dd(degrees, minutes, seconds, direction):
+    dd = float(degrees) + float(minutes)/60 + float(seconds)/3600
+    if direction in ["S", "W"]:
+        dd *= -1
+    return dd
+
+# Regex for DMS coordinates (example: 48°51′29.9″N or 2°17′40.2″E)
+dms_pattern = re.compile(
+    r"(\d{1,3})[°\s]\s*(\d{1,2})[′']\s*(\d{1,2}(?:\.\d+)?)[″\"]?\s*([NSEW])"
+)
+
+
 examples = [
     "* AACH ou ACH, s. f. petite ville d'Allemagne dans le cercle de Souabe, près de la source de l'Aach. Long. 26. 57. lat. 47. 55.",
     "* ARCALU (Principauté d') petit état des Tartares-Monguls, sur la riviere d'Hoamko, où commence la grande muraille de la Chine, sous le 122e degré de longitude & le 42e de latitude septentrionale.",
@@ -63,7 +85,9 @@ with gr.Blocks() as demo:
             run_btn.click(fn=extract_coordinates, inputs=inp, outputs=out_text)
 
             out_text = gr.Textbox(label="Extracted and normalized DMS coordinates (fine-tuned mT5)")
-            run_btn.click(fn=norm_coordinates, inputs=inp, outputs=out_text)
+            out_map = gr.Map(label="Coordinates on Map")
+
+            run_btn.click(fn=norm_coordinates, inputs=inp, outputs=[out_text,  out_map])
 
     with gr.Row():