EITD commited on
Commit
a408e8f
·
1 Parent(s): 2a078b6
Files changed (1) hide show
  1. app.py +28 -28
app.py CHANGED
@@ -1,6 +1,6 @@
1
  from peft import AutoPeftModelForCausalLM
2
  from transformers import AutoTokenizer, TextStreamer
3
- # import gradio as gr
4
  """
5
  For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
6
  """
@@ -12,19 +12,19 @@ model = AutoPeftModelForCausalLM.from_pretrained(
12
  )
13
  tokenizer = AutoTokenizer.from_pretrained("EITD/lora_model_1")
14
 
15
- messages = [{"role": "user", "content": "Continue the Fibonacci sequence: 1, 1, 2, 3, 5, 8,"},]
16
 
17
- inputs = tokenizer.apply_chat_template(
18
- messages,
19
- tokenize = True,
20
- add_generation_prompt = True, # Must add for generation
21
- return_tensors = "pt",
22
- )
23
 
24
- outputs = model.generate(input_ids = inputs, max_new_tokens = 64, use_cache = True,
25
- temperature = 1.5, min_p = 0.1)
26
 
27
- print(tokenizer.batch_decode(outputs))
28
 
29
  def respond(
30
  message,
@@ -78,22 +78,22 @@ def respond(
78
  """
79
  For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
80
  """
81
- # demo = gr.ChatInterface(
82
- # respond,
83
- # additional_inputs=[
84
- # gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
85
- # gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
86
- # gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
87
- # gr.Slider(
88
- # minimum=0.1,
89
- # maximum=1.0,
90
- # value=0.95,
91
- # step=0.05,
92
- # label="Top-p (nucleus sampling)",
93
- # ),
94
- # ],
95
- # )
96
 
97
 
98
- # if __name__ == "__main__":
99
- # demo.launch()
 
1
  from peft import AutoPeftModelForCausalLM
2
  from transformers import AutoTokenizer, TextStreamer
3
+ import gradio as gr
4
  """
5
  For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
6
  """
 
12
  )
13
  tokenizer = AutoTokenizer.from_pretrained("EITD/lora_model_1")
14
 
15
+ # messages = [{"role": "user", "content": "Continue the Fibonacci sequence: 1, 1, 2, 3, 5, 8,"},]
16
 
17
+ # inputs = tokenizer.apply_chat_template(
18
+ # messages,
19
+ # tokenize = True,
20
+ # add_generation_prompt = True, # Must add for generation
21
+ # return_tensors = "pt",
22
+ # )
23
 
24
+ # outputs = model.generate(input_ids = inputs, max_new_tokens = 64, use_cache = True,
25
+ # temperature = 1.5, min_p = 0.1)
26
 
27
+ # print(tokenizer.batch_decode(outputs))
28
 
29
  def respond(
30
  message,
 
78
  """
79
  For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
80
  """
81
+ demo = gr.ChatInterface(
82
+ respond,
83
+ additional_inputs=[
84
+ gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
85
+ gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
86
+ gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
87
+ gr.Slider(
88
+ minimum=0.1,
89
+ maximum=1.0,
90
+ value=0.95,
91
+ step=0.05,
92
+ label="Top-p (nucleus sampling)",
93
+ ),
94
+ ],
95
+ )
96
 
97
 
98
+ if __name__ == "__main__":
99
+ demo.launch()