Update

app.py

@@ -45,7 +45,7 @@ def generate_summary(question, model):
     input_ids = inputs.input_ids.to(model.device)
     attention_mask = inputs.attention_mask.to(model.device)
     outputs = model.generate(
-        input_ids, attention_mask=attention_mask, max_new_tokens=512)
+        input_ids, attention_mask=attention_mask, max_new_tokens=4096, do_sample=True, num_beams=4, top_k=50, early_stopping=True, no_repeat_ngram_size=2, )
     output_str = tokenizer.batch_decode(outputs, skip_special_tokens=True)
     return outputs, output_str
 
@@ -144,14 +144,14 @@ def chat():
 def post_process(output):
     
     output = textwrap.fill(textwrap.dedent(output).strip(), width=120)
-    lines = output.split("\n")
+    lines = output.split("\.")
     for line in lines:
         for word in word_remove_sentence:
             if word in line.lower():
                 lines.remove(line)
                 break
     
-    output = "\n".join(lines)
+    output = "\.".join(lines)
     for item in map_words.keys():
         output = re.sub(item, map_words[item], output, re.I)