Update app.py

app.py

@@ -1,40 +1,10 @@
 import gradio as gr
-from transformers import pipeline
+from transformers import AutoModelForCausalLM, AutoTokenizer
+import torch
 
-#pipe = pipeline("text-generation")
-#gr.Interface.from_pipeline(pipe).launch()
+tokenizer = AutoTokenizer.from_pretrained("jslin09/bloom-560m-finetuned-fraud")
+model = AutoModelForCausalLM.from_pretrained("jslin09/bloom-560m-finetuned-fraud")
 
-API_URL = "https://api-inference.huggingface.co/models/jslin09/bloom-560m-finetuned-fraud"
-headers = {"Authorization": "Bearer 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"} # Read only
-
-description = "Legal Document Drafting with BLOOM"
-api_key="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"
-examples=[
-          ["闕很大明知金融帳戶之存摺、提款卡及密碼係供自己使用之重要理財工具，"],
-          ["森上梅前明知其無資力支付酒店消費，亦無付款意願，竟意圖為自己不法之所有，"]
-]
-
-iface = gr.Interface.load(
-    "huggingface/jslin09/bloom-560m-finetuned-fraud",
-    title="Drafting",
-    inputs=[
-        gr.Textbox(lines=10, label="Prompt", value="闕很大明知金融帳戶之存摺、提款卡及密碼係供自己使用之重要理財工具，"),  # prompt
-        gr.Slider(10, 200, step=10, value=100),  # token_count
-        gr.Slider(0.2, 2.0, step=0.1, value=1.0),  # temperature
-        gr.Slider(0.0, 1.0, step=0.05, value=0.8),  # top_p
-        gr.Slider(0.0, 1.0, step=0.1, value=0.1),  # presencePenalty
-        gr.Slider(0.0, 1.0, step=0.1, value=0.1),  # countPenalty
-    ],
-    outputs=gr.Textbox(label="生成的草稿", lines=28),
-    description=description,
-    examples=examples,
-    api_key=api_key,
-).queue()
-
-demo = gr.TabbedInterface(
-    [iface], ["分頁標籤"],
-    title="Legal Document Drafting",
-)
-
-demo.queue()
-demo.launch(share=False)
+def predict(input, history=[]):
+    # tokenize the new input sentence
+    new_user_input_ids = tokenizer.encode(input + tokenizer.eos_token, return_tensors='pt')