Set max_length=500 when calling generate()

app.py

@@ -12,7 +12,7 @@ def translate(text_src):
     embeddings = src_tokenizer(text_src, return_attention_mask=False, return_token_type_ids=False, return_tensors='pt')
     embeddings = {k: v for k, v in embeddings.items()}
     # using default generation method: GreedySearch, No LogitsProcessor
-    output = model.generate(**embeddings)[0, 1:-1]
+    output = model.generate(**embeddings, max_length=500)[0, 1:-1]
     text_trg = trg_tokenizer.decode(output.cpu())
     return text_trg