Update app.py

app.py

@@ -60,8 +60,21 @@ class Translators:
 
     def quickmt(self):
         from quickmt import Translator
+        from quickmt.hub import hf_download, hf_list
+        from pathlib import Path
+        input_model = f"quickmt-{self.sl}-{self.tl}"
+        model_path = Path("/code/models") / input_model
+        choices = [i.split("/")[1] for i in hf_list()]
+        print(choices)
+        if not model_path.exists():
+            hf_download(
+            model_name = f"quickmt/{input_model}",
+            output_dir=Path("/code/models") / input_model,
+        )
+        print(f"Loading model {input_model}")
         # Auto-detects GPU, set to "cpu" to force CPU inference
-        translation = Translator(f"./quickmt-{self.sl}-{self.tl}/", device="auto")       
+        translation = Translator(str(model_path), device="auto", inter_threads=2)       
+        # translation = Translator(f"./quickmt-{self.sl}-{self.tl}/", device="auto", inter_threads=2)       
         # Translate - set beam size to 1 for faster speed (but lower quality)
         return translation({self.input_text}, beam_size=5)