Upload 2 files

app.py

@@ -0,0 +1,76 @@
+import gradio as gr
+from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
+import torch
+
+REPO_ID = "Omarrran/koshur-diacritizer-byt5-small"
+
+print("Loading model...")
+tokenizer = AutoTokenizer.from_pretrained(REPO_ID)
+model = AutoModelForSeq2SeqLM.from_pretrained(REPO_ID)
+model.eval()
+print("Model loaded.")
+
+
+def diacritize(text: str, max_tokens: int) -> str:
+    if not text or not text.strip():
+        return ""
+    inputs = tokenizer(text.strip(), return_tensors="pt", padding=True)
+    with torch.no_grad():
+        out = model.generate(**inputs, max_new_tokens=int(max_tokens))
+    return tokenizer.decode(out[0], skip_special_tokens=True)
+
+
+examples = [
+    ["کاشر زبان", 256],
+    ["میانی ہند", 256],
+    ["سریںنگر شہر بوہت خوبصورت چھ", 256],
+    ["اس کتاب منز بوہت ژھور معلومات چھ", 256],
+    ["کشیر گرمی منز سبز تہ خوبصورت اوسان چھ", 256],
+    ["زہ پرون شہر گوم", 256],
+    ["امی گر کیاہ پیٹھ بنایو", 256],
+]
+
+description = """
+## Koshur Diacritizer — ByT5-Small
+
+This model restores **diacritical marks** (اِعراب) to undiacritized Kashmiri (کٲشُر) text written in Perso-Arabic script.
+
+**Model:** [`Omarrran/koshur-diacritizer-byt5-small`](https://huggingface.co/Omarrran/koshur-diacritizer-byt5-small)
+
+Enter raw Kashmiri text below or click an example to try instantly.
+"""
+
+demo = gr.Interface(
+    fn=diacritize,
+    inputs=[
+        gr.Textbox(
+            label="Input Text (undiacritized Kashmiri)",
+            placeholder="یہاں کٲشُر متن لِکھِو…",
+            lines=3,
+            rtl=True,
+        ),
+        gr.Slider(
+            minimum=64,
+            maximum=512,
+            value=256,
+            step=32,
+            label="Max New Tokens",
+        ),
+    ],
+    outputs=gr.Textbox(
+        label="Diacritized Output",
+        lines=3,
+        rtl=True,
+        show_copy_button=True,
+    ),
+    examples=examples,
+    title="کٲشُر ڈایاکرِٹایزر | Koshur Diacritizer",
+    description=description,
+    article="Built by [Omar Haq Nawaz Malik](https://huggingface.co/Omarrran) as part of the Kashmiri language AI infrastructure initiative.",
+    theme=gr.themes.Soft(),
+    cache_examples=False,
+    flagging_mode="never",
+)
+
+if __name__ == "__main__":
+    demo.launch()

requirements.txt

@@ -0,0 +1,5 @@
+transformers
+torch
+gradio
+sentencepiece
+protobuf