app

app.py

@@ -1,7 +1,38 @@
 import gradio as gr
 
-def greet(name):
-    return "Hello " + name + "!!"
 
-iface = gr.Interface(fn=greet, inputs="text", outputs="text")
+
+
+import torch
+from transformers import AutoTokenizer
+from transformers import T5Tokenizer, T5ForConditionalGeneration
+
+# tokenizer = T5Tokenizer.from_pretrained("ClueAI/PromptCLUE-base")
+# model = T5ForConditionalGeneration.from_pretrained("ClueAI/PromptCLUE-base")
+tokenizer = T5Tokenizer.from_pretrained("ClueAI/PromptCLUE-base-v1-5")
+model = T5ForConditionalGeneration.from_pretrained("ClueAI/PromptCLUE-base-v1-5")
+
+device = torch.device('cpu')
+model.to(device)
+def preprocess(text):
+  return text.replace("\n", "_")
+
+def postprocess(text):
+  return text.replace("_", "\n")
+
+def answer(text, sample=False, top_p=0.6):
+  '''sample：是否抽样。生成任务，可以设置为True;
+  top_p：0-1之间，生成的内容越多样'''
+  text = preprocess(text)
+  encoding = tokenizer(text=[text], truncation=True, padding=True, max_length=768, return_tensors="pt").to(device) 
+  if not sample:
+    out = model.generate(**encoding, return_dict_in_generate=True, output_scores=False, max_new_tokens=128, num_beams=4, length_penalty=0.6)
+  else:
+    out = model.generate(**encoding, return_dict_in_generate=True, output_scores=False, max_new_tokens=128, do_sample=True, top_p=top_p)
+  out_text = tokenizer.batch_decode(out["sequences"], skip_special_tokens=True)
+  return postprocess(out_text[0])
+
+
+
+iface = gr.Interface(fn=answer, inputs="text", outputs="text")
 iface.launch()

requirements.txt

@@ -0,0 +1,4 @@
+transformers
+torch
+gradio
+sentencepiece