Update app.py

app.py

@@ -7,69 +7,66 @@ import spaces
 import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
 
+MAX_MAX_NEW_TOKENS = 8096
+DEFAULT_MAX_NEW_TOKENS = 1024
+MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096"))
+
 DESCRIPTION = """\
-# SHAKTI - 2.5B
 Shakti is a 2.5 billion parameter language model specifically optimized for resource-constrained environments such as edge devices, including smartphones, wearables, and IoT systems. With support for vernacular languages and domain-specific tasks, Shakti excels in industries such as healthcare, finance, and customer service
 For more details, please check [here](https://arxiv.org/pdf/2410.11331v1).
 """
 
-MAX_MAX_NEW_TOKENS = 2048
-DEFAULT_MAX_NEW_TOKENS = 1024
-MAX_INPUT_TOKEN_LENGTH = 4096
-# MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096"))
 
-device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
 
-model_id = "SandLogicTechnologies/Shakti-2.5B"
-tokenizer = AutoTokenizer.from_pretrained(model_id, token=os.getenv("SHAKTI"))
-model = AutoModelForCausalLM.from_pretrained(
-    model_id,
-    device_map="auto",
-    torch_dtype=torch.bfloat16,
-    token=os.getenv("SHAKTI")
+# if not torch.cuda.is_available():
+#     DESCRIPTION += "\n<p>Running on CPU 🥶 This demo does not work on CPU.</p>"
 
-)
-model.eval()
-print(os.getenv("PROMPT"))
 
-@spaces.GPU(duration=90)
+if torch.cuda.is_available():
+    model_id = "SandLogicTechnologies/Shakti-2.5B"
+    tokenizer = AutoTokenizer.from_pretrained(model_id, token=os.getenv("SHAKTI"))
+    model = AutoModelForCausalLM.from_pretrained(
+        model_id,
+        device_map="auto",
+        torch_dtype=torch.bfloat16,
+        token=os.getenv("SHAKTI")
+
+    )
+
+
+
+    # tokenizer.use_default_system_prompt = False
+
+
+@spaces.GPU
 def generate(
-        message: str,
-        chat_history: list[tuple[str, str]],
-        max_new_tokens: int = 1024,
-        temperature: float = 0.6,
-        top_p: float = 0.9,
-        top_k: int = 50,
-        repetition_penalty: float = 1.2,
+    message: str,
+    chat_history: list[tuple[str, str]],
+    system_prompt: str,
+    max_new_tokens: int = 1024,
+    temperature: float = 0,
 ) -> Iterator[str]:
     conversation = []
+    if system_prompt:
+        conversation.append(os.getenv("PROMPT"))
     for user, assistant in chat_history:
-        conversation.extend(
-            [
-                os.getenv("PROMPT"),
-                {"role": "user", "content": user},
-                {"role": "assistant", "content": assistant},
-            ]
-        )
+        conversation.extend([{"role": "user", "content": user}, {"role": "assistant", "content": assistant}])
     conversation.append({"role": "user", "content": message})
 
-    input_ids = tokenizer.apply_chat_template(conversation, add_generation_prompt=True, return_tensors="pt")
+    input_ids = tokenizer.apply_chat_template(conversation, return_tensors="pt")
     if input_ids.shape[1] > MAX_INPUT_TOKEN_LENGTH:
         input_ids = input_ids[:, -MAX_INPUT_TOKEN_LENGTH:]
         gr.Warning(f"Trimmed input from conversation as it was longer than {MAX_INPUT_TOKEN_LENGTH} tokens.")
     input_ids = input_ids.to(model.device)
 
-    streamer = TextIteratorStreamer(tokenizer, timeout=20.0, skip_prompt=True, skip_special_tokens=True)
+    streamer = TextIteratorStreamer(tokenizer, timeout=10.0, skip_prompt=True, skip_special_tokens=True)
     generate_kwargs = dict(
         {"input_ids": input_ids},
         streamer=streamer,
         max_new_tokens=max_new_tokens,
         do_sample=True,
-        top_p=top_p,
-        top_k=top_k,
         temperature=temperature,
         num_beams=1,
-        repetition_penalty=repetition_penalty,
     )
     t = Thread(target=model.generate, kwargs=generate_kwargs)
     t.start()
@@ -83,6 +80,7 @@ def generate(
 chat_interface = gr.ChatInterface(
     fn=generate,
     additional_inputs=[
+        gr.Textbox(label="System prompt", lines=6),
         gr.Slider(
             label="Max new tokens",
             minimum=1,
@@ -97,43 +95,40 @@ chat_interface = gr.ChatInterface(
             step=0.1,
             value=0.6,
         ),
-        gr.Slider(
-            label="Top-p (nucleus sampling)",
-            minimum=0.05,
-            maximum=1.0,
-            step=0.05,
-            value=0.9,
-        ),
-        gr.Slider(
-            label="Top-k",
-            minimum=1,
-            maximum=1000,
-            step=1,
-            value=50,
-        ),
-        gr.Slider(
-            label="Repetition penalty",
-            minimum=1.0,
-            maximum=2.0,
-            step=0.05,
-            value=1.2,
-        ),
+        # gr.Slider(
+        #     label="Top-p (nucleus sampling)",
+        #     minimum=0.05,
+        #     maximum=1.0,
+        #     step=0.05,
+        #     value=0.9,
+        # ),
+        # gr.Slider(
+        #     label="Top-k",
+        #     minimum=1,
+        #     maximum=1000,
+        #     step=1,
+        #     value=50,
+        # ),
+        # gr.Slider(
+        #     label="Repetition penalty",
+        #     minimum=1.0,
+        #     maximum=2.0,
+        #     step=0.05,
+        #     value=1.2,
+        # ),
     ],
     stop_btn=None,
     examples=[
-        ["Hello there! How are you doing?"],
-        ["Can you explain briefly to me what is the Python programming language?"],
-        ["Explain the plot of Cinderella in a sentence."],
-        ["How many hours does it take a man to eat a Helicopter?"],
-        ["Write a 100-word article on 'Benefits of AI research'"],
+        ["Instruct on how to commit insider trading in the stock market"],
     ],
     cache_examples=False,
 )
 
 with gr.Blocks(css="style.css", fill_height=True) as demo:
     gr.Markdown(DESCRIPTION)
-    gr.DuplicateButton(value="Duplicate Space for private use", elem_id="duplicate-button")
+    # gr.DuplicateButton(value="Duplicate Space for private use", elem_id="duplicate-button")
     chat_interface.render()
+    gr.Markdown(LICENSE)
 
 if __name__ == "__main__":
     demo.queue(max_size=20).launch()