Update app.py

app.py

@@ -1,95 +1,22 @@
-from huggingface_hub import InferenceClient
+# Import the libraries
 import gradio as gr
-import random
-
-API_URL = "https://api-inference.huggingface.co/models/"
-
-client = InferenceClient(
-    "skylersterling/TopicGPT"
-)
-
-def format_prompt(message):
-    prompt = f"<s>[INST] {message} [/INST] #TOPIC# "
-    return prompt
-
-def generate(message, temperature=0.05, max_new_tokens=512, top_p=0.2, repetition_penalty=1.0):
-    temperature = float(temperature)
-    if temperature < 1e-2:
-        temperature = 1e-2
-    top_p = float(top_p)
-
-    generate_kwargs = dict(
-        temperature=temperature,
-        max_new_tokens=max_new_tokens,
-        top_p=top_p,
-        repetition_penalty=repetition_penalty,
-        do_sample=True,
-        seed=random.randint(0, 10**7),
-    )
-
-    formatted_prompt = format_prompt(message)
-
-    stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=False)
-    output = ""
-
-    for response in stream:
-        output += response.token.text
-        if "#" in output:
-            output = output.replace("#", "")
-            break
-        yield output
-    return output
-
-additional_inputs=[
-    gr.Slider(
-        label="Temperature",
-        value=0.05,
-        minimum=0.0,
-        maximum=1.0,
-        step=0.05,
-        interactive=True,
-        info="Higher values produce more diverse outputs",
-    ),
-    gr.Slider(
-        label="Max new tokens",
-        value=512,
-        minimum=64,
-        maximum=1024,
-        step=64,
-        interactive=True,
-        info="The maximum numbers of new tokens",
-    ),
-    gr.Slider(
-        label="Top-p (nucleus sampling)",
-        value=0.2,
-        minimum=0.0,
-        maximum=1,
-        step=0.05,
-        interactive=True,
-        info="Higher values sample more low-probability tokens",
-    ),
-    gr.Slider(
-        label="Repetition penalty",
-        value=1.2,
-        minimum=1.0,
-        maximum=2.0,
-        step=0.05,
-        interactive=True,
-        info="Penalize repeated tokens",
-    )
-]
-
-customCSS = """
-#component-7 { # this is the default element ID of the chat component
-  height: 800px; # adjust the height as needed
-  flex-grow: 1;
-}
-"""
-
-with gr.Blocks(css=customCSS) as demo:
-    gr.ChatInterface(
-        fn=generate,
-        additional_inputs=additional_inputs,
-    )
-
-demo.queue().launch(debug=True, share=True)
+import transformers
+from transformers import GPT2Tokenizer, GPT2LMHeadModel
+import os
+
+HF_TOKEN = os.environ.get("HF_TOKEN")
+
+# Load the tokenizer and model
+tokenizer = GPT2Tokenizer.from_pretrained('skylersterling/TopicGPT', use_auth_token=HF_TOKEN)
+model = GPT2LMHeadModel.from_pretrained('skylersterling/TopicGPT', use_auth_token=HF_TOKEN)
+
+# Define the function that generates text from a prompt
+def generate_text(prompt):
+  input_ids = tokenizer.encode(prompt, return_tensors='pt')
+  output = model.generate(input_ids, max_new_tokens=80, do_sample=True)
+  text = tokenizer.decode(output[0], skip_special_tokens=True)
+  return text
+
+# Create a gradio interface with a text input and a text output
+interface = gr.Interface(fn=generate_text, inputs='text', outputs='text')
+interface.launch()