email-summarizer

Model Card for Model ID

This Model takes in an email and provides a quick summary.

Model Details

How to use on google colab:

  #Install dependencies
  !pip install transformers
  
  #Load the model and tokenizer
  from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
  
  tokenizer = AutoTokenizer.from_pretrained("IrisWiris/email-summarizer")
  model = AutoModelForSeq2SeqLM.from_pretrained("IrisWiris/email-summarizer")
  
  #Summarize an email
  email = 
  """
    paste the email
  """
  inputs = tokenizer(email, return_tensors="pt", truncation=True)
  outputs = model.generate(inputs.input_ids, max_length=100)
  summary = tokenizer.decode(outputs[0], skip_special_tokens=True)
  print(summary)

example:

  email = """
    Good morning,
  
      I hope this message finds you well. Unfortunately, I am unwell and will not be able to hold today's lecture or any classes for the remainder of this week. As a result, all classes are canceled.
      However, I uploaded a document that you should review before classes resume next week as scheduled. 
      Please ensure you go through the material, as it will be important for our upcoming discussions.
      
      Thank you for your understanding, and I look forward to seeing you next week.
  
    Best regards,
    James
  """
  inputs = tokenizer(email, return_tensors="pt", truncation=True)
  outputs = model.generate(inputs.input_ids, max_length=100)
  summary = tokenizer.decode(outputs[0], skip_special_tokens=True)
  print(summary)


  Output:
  James is unwell and will not hold today's lecture or any classes for the remainder of this week. He uploaded a document that will review the material before classes resume next week as scheduled.

Model Description

As a university student, I get a lot of emails. Unforunatly I have to read each and every one, because among a lot of unessary information there exist a few important notes. To make the email reading process less tedious, I have trained this model to summarize an email input.

• Developed by: IrisWiris
• Model type: Generative AI

Training Data

Used argilla/FinePersonas-Conversations-Email-Summaries: https://huggingface.co/datasets/argilla/FinePersonas-Conversations-Email-Summaries

Citation [optional]

All code used was taught to me by Shiftkey Labs' GenAI Course: https://shiftkeylabs.ca/