Upload folder using huggingface_hub

.gradio/certificate.pem

@@ -0,0 +1,31 @@
+-----BEGIN CERTIFICATE-----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=
+-----END CERTIFICATE-----

data/simplified_dict.txt

@@ -14975,6 +14975,7 @@
 	"िशवालय": "शवलय",
 	"िसददेश्वर अाधारभुत िवद्यालय": "सददश्वरअधरभतवद्यलय",
 	"ोखरपौवा": "खरपव",
+	"रुकुम":"रकम",
 	"ॐ ओम शान्ति ध्यान केन्द्र": "ॐओमशन्तध्यनकन्द्र",
 	"ॐ नम पाण्डवेश्वर महादेव बचत तथा ऋण सहकारी सस्था लिमिटेड": "ॐनमपण्डवश्वरमहदवबचततथऋणसहकरसस्थलमटड",
 	"ॐ नमः शिवाय फर्मा": "ॐनमशवयफर्म",
@@ -14995,5 +14996,5 @@
 	"४ नं वडा कार्यालय महेन्द्रधार": "४नवडकर्यलयमहन्द्रधर",
 	"६६१/१५": "६६११५",
 	"७ अ साईड मैडान": "७असईडमडन",
-	"८डी म्याजिक थ्रीयटर": "८डम्यजकथ्रयटर",
+	"८डी म्याजिक थ्रीयटर": "८डम्यजकथ्रयटर"
 }

data/vocab.txt

@@ -252549,3 +252549,4 @@
 देख्नुभएका
 बदल्छन्
 मस्यौदाभित्र
+रुकुम (पश्चिम भाग)

requirements.txt

@@ -60,8 +60,6 @@ hyperlink==21.0.0
 idna==2.8
 importlib-resources==1.4.0
 incremental==24.7.2
-ipython==9.2.0
-ipython_pygments_lexers==1.1.1
 itemadapter==0.11.0
 itemloaders==1.3.2
 itsdangerous==2.2.0

runed_gradio.py

@@ -1,9 +1,15 @@
 import gradio as gr
 import re
+import json
 from symspellpy import SymSpell, Verbosity
 from nepali_stemmer.stemmer import NepStemmer
 from itertools import product
-from typing import List, Tuple, Dict, Set
+from typing import List, Dict, Set
+
+# ------------------- Paths -------------------
+simplified_only_path = "./data/simplified_only_names2.txt"
+simplified_dict_path = "./data/simplified_dict.txt"
+vocab_path = "./data/vocab.txt"
 
 # ------------------- Utilities -------------------
 
@@ -17,17 +23,50 @@ def load_vocab(filepath: str) -> Set[str]:
     with open(filepath, "r", encoding="utf-8") as f:
         return {line.strip() for line in f if line.strip()}
 
+def save_to_vocab(word: str, filepath: str = vocab_path) -> str:
+    word = word.strip()
+    if not word:
+        return "Invalid input. No word added."
+    vocab = load_vocab(filepath)
+    if word in vocab:
+        return f"'{word}' already exists in the vocab."
+    with open(filepath, "a", encoding="utf-8") as f:
+        f.write(word + "\n")
+    return f"'{word}' added to vocab."
+
+def load_simplified_keys(filepath: str) -> Set[str]:
+    keys = set()
+    with open(filepath, "r", encoding="utf-8") as f:
+        for line in f:
+            word = line.strip()
+            if word:
+                keys.add(word)
+    return keys
+
 def load_simplified_map(filepath: str) -> Dict[str, str]:
-    simplified_map = {}
     with open(filepath, "r", encoding="utf-8") as f:
+        data = json.load(f)
+    return {v: k for k, v in data.items()}
+
+def list_locations(simplified_keys_file: str = simplified_only_path,
+                   simplified_map_file: str = simplified_dict_path) -> str:
+    simplified_map = load_simplified_map(simplified_map_file)
+
+    keys = []
+    with open(simplified_keys_file, "r", encoding="utf-8") as f:
         for line in f:
-            if ":" not in line:
+            line = line.strip()
+            if not line:
                 continue
-            parts = line.strip().strip(",").replace('"', '').split(":")
-            if len(parts) == 2:
-                orig, simp = parts[0].strip(), parts[1].strip()
-                simplified_map[simp] = orig
-    return simplified_map
+            key = line.split("$")[0] 
+            keys.append(key)
+
+    output_lines = []
+    for key in keys:
+        original_name = simplified_map.get(key, "Unknown")
+        output_lines.append(f"{key}  ->  {original_name}")
+
+    return "\n".join(output_lines)
 
 def init_spellchecker(dict_path: str, max_edit_distance: int, prefix_length: int) -> SymSpell:
     sym_spell = SymSpell(max_dictionary_edit_distance=max_edit_distance, prefix_length=prefix_length)
@@ -44,11 +83,6 @@ def correct_sentence(
     top_k: int
 ) -> List[str]:
 
-    simplified_only_path = "./data/simplified_only_names2.txt"
-    simplified_dict_path = "./data/simplified_dict.txt"
-    vocab_path = "./data/vocab.txt"
-
-    # Load components
     sym_spell = init_spellchecker(simplified_only_path, max_edit_distance, prefix_length)
     simplified_map = load_simplified_map(simplified_dict_path)
     vocab = load_vocab(vocab_path)
@@ -92,24 +126,54 @@ def correct_sentence(
 
 # ------------------- Gradio UI -------------------
 
-examples = [
-    ["भतपरको  जिज्ञासु वातावरणले धेरै पर्यटकलाई आकर्षित गर्छ।", 2, 3, 3],
-    ["ललतपुर प्राचीन मूर्तिकला र वास्तुकलाको केन्द्र हो।", 2, 3, 3],
-]
-
-iface = gr.Interface(
-    fn=correct_sentence,
-    inputs=[
-        gr.Textbox(label="Input Nepali Sentence", lines=2, placeholder="नेपालको समृद्ध इतिहास..."),
-        gr.Slider(0, 4, value=2, step=1, label="Max Edit Distance"),
-        gr.Slider(1, 5, value=3, step=1, label="Prefix Length"),
-        gr.Slider(1, 5, value=3, step=1, label="Top-K Suggestions per Word")
-    ],
-    outputs=gr.Textbox(label="Corrected Sentence Variants"),
-    title="Nepali Spell Correction App",
-    description="Generates corrected sentence variants using SymSpell and a stemmer.",
-    examples=examples
-)
+with gr.Blocks(title="Nepali Spell Correction Tool") as demo:
+    gr.Markdown(
+        """
+        # Nepali Spell Correction Tool
+        Automatically correct Nepali sentences using spelling correction and stemming techniques.
+        """
+    )
+
+    with gr.Row():
+        with gr.Column(scale=3):
+            gr.Markdown("### Sentence Correction")
+            sentence_input = gr.Textbox(label="Input Sentence", placeholder="Enter a Nepali sentence...", lines=2)
+            max_dist = gr.Slider(0, 4, value=2, step=1, label="Max Edit Distance")
+            prefix_len = gr.Slider(1, 5, value=3, step=1, label="Prefix Length")
+            top_k = gr.Slider(1, 5, value=3, step=1, label="Top-K Suggestions")
+
+            submit_btn = gr.Button("Correct Sentence")
+
+            gr.Markdown("### Location Vocabulary")
+
+            with gr.Accordion("View and Manage Locations", open=False):
+                loc_out = gr.Textbox(label="Available Locations", lines=8, interactive=False)
+                view_btn = gr.Button("Show Locations")
+
+                # Uncomment below to add vocab management features
+                # new_loc = gr.Textbox(label="Add New Place", placeholder="e.g., काठमाडौँ")
+                # add_btn = gr.Button("Add Location")
+                # add_msg = gr.Textbox(label="Status", interactive=False)
+
+        with gr.Column(scale=2):
+            corrected_out = gr.Textbox(label="Corrected Variants", lines=8)
+
+    # Bind buttons to functions
+    submit_btn.click(
+        correct_sentence,
+        inputs=[sentence_input, max_dist, prefix_len, top_k],
+        outputs=corrected_out
+    )
+
+    view_btn.click(
+        list_locations,
+        inputs=[],
+        outputs=loc_out
+    )
+
+    # add_btn.click(save_to_vocab, inputs=new_loc, outputs=add_msg)
+
+# ------------------- Launch App -------------------
 
 if __name__ == "__main__":
-    iface.launch(share=True)
+    demo.launch()