Model Card for Model ID

Requirements

!pip install gradio
!pip install -U xformers --index-url https://download.pytorch.org/whl/cu121
!pip install "unsloth[kaggle-new] @ git+https://github.com/unslothai/unsloth.git"

import os
os.environ["WANDB_DISABLED"] = "true"

Gradio App


import gradio as gr
from transformers import AutoTokenizer
from peft import AutoPeftModelForCausalLM
import torch
import anthropic

# Assuming the model and tokenizer for Mistral are correctly set up as per your provided code.
# Let's also assume you have a way to call the Anthropic model, perhaps via an API or another library.
load_in_4bit = True
model = AutoPeftModelForCausalLM.from_pretrained(
        "DisgustingOzil/Mistral_summarizer",
        load_in_4bit=load_in_4bit,
        torch_dtype=torch.float16,
    ).to("cuda")
tokenizer = AutoTokenizer.from_pretrained("DisgustingOzil/Mistral_summarizer")
def summarize_with_mistral(text):


    summary_prompt = f"""Below is a text that needs to be summarized. Based on the input, write a good summary which summarize all main points.

### Text:
{text}

### Summary:
"""  # The summary part is left empty for generation

    inputs = tokenizer([summary_prompt], return_tensors="pt").to("cuda")
    outputs = model.generate(**inputs, max_new_tokens=150, use_cache=True)
    summary = tokenizer.batch_decode(outputs, skip_special_tokens=True)
    summary_start_index = summary[0].find("### Summary:")
    summary_text = summary[0][summary_start_index:].replace("### Summary:", "").strip()
    return summary_text
summary_1=""
def summarize_with_anthropic(text):
      API_KEY="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"

      client = anthropic.Anthropic(
          # defaults to os.environ.get("ANTHROPIC_API_KEY")
          api_key=API_KEY,
      )
      message = client.messages.create(
          model="claude-3-haiku-20240307",
          max_tokens=3214,
          temperature=0,
          system="Create Good summary explaining all key points in detail, easy and understandable way",
          messages=[
              {
                  "role": "user",
                  "content": [
                      {
                          "type": "text",
                          "text": text
                      }
                  ]
              }
          ]
      )
    # Placeholder function to represent summarization with an Anthropic model.
    # This should be replaced with actual API calls or function calls to the Anthropic model.
      # summary_1=message.content[0]
      summary=message.content[0]
      return summary.text

def summarize_text(text, model_choice):
    if model_choice == "Mistral 7b":
        return summarize_with_mistral(text)
    elif model_choice == "Claude-3-Haiku":
        return summarize_with_anthropic(text)
    else:
        return "Invalid model choice."

# Define the Gradio interface with a dropdown for model selection
iface = gr.Interface(
    fn=summarize_text,
    inputs=[gr.Textbox(lines=10, label="Input Text"), gr.Dropdown(choices=["Mistral 7b", "Claude-3-Haiku"], label="Model Choice")],
    outputs=gr.Textbox(label="Summary"),
    title="Text Summarization",
    description="Enter text to summarize based on Maxwell's equations and related concepts. Select a model for summarization."
)

# Launch the app
if __name__ == "__main__":
    iface.launch(debug=True)