DialoGPT2 Instruction Following

This is the fine-tuned version of the microsoft/dialogpt-small on the instruction following task. The dataset used was the hakurei/open-instruct-v1 dataset.

Find the training notebook here on Kaggle.

Using the model

Using model.generate()

To use the model, first call the checkpoints and initialize the model

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("smji/dialogpt2-instruct-following")
model = AutoModelForCausalLM.from_pretrained("smji/dialogpt2-instruct-following")

And then move onto generating the text

def generate_text(prompt):
    inputs = tokenizer.encode(prompt, return_tensors='pt').to(device)
    outputs = model.generate(inputs, max_length=512, pad_token_id=tokenizer.eos_token_id)
    generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)

    return generated_text[:generated_text.rfind('.')+1]

generate_text("How can I bake a cake?")

Using the pipeline

Or, you can also use the pipeline

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="smji/dialogpt2-instruct-following")

pipe("How can I bake a cake?", max_length=512)

Done by S M Jishanul Islam

Downloads last month
18
Safetensors
Model size
124M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Dataset used to train smji/dialogpt2-instruct-following