Code Generation using GPT2-Large
This is a GPT2-large model that's further fine-tuned on the Codeparrot dataset with a custom metric focused on code generation.
The Tokenizer is initialized from the GPT2-large and further trained on the same dataset to better align the tokenization for generating code.
Model description
This Model has the same architecture and Parameters as the GPT2-large model. Please refer to this link to know more about the model details.
Intended Use & Limitations
This model is intended to generate code for the required function based on a small description of the output required.
Note: The model is primarily trained with an objective of code generation.
Usage
You can use this model directly to get the summaries:
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
# Load Code Generator LLM and tokenizer from checkpoint
tokenizer = AutoTokenizer.from_pretrained("DeathReaper0965/gpt2_large_code_generator")
model = AutoModelForCausalLM.from_pretrained("DeathReaper0965/gpt2_large_code_generator")
model = model.to("cuda" if torch.cuda.is_available() else "cpu")
inputs = tokenizer("def hello_world():", return_tensors="pt").to("cuda" if torch.cuda.is_available() else "cpu")
outputs = model.generate(**inputs,
max_new_tokens= 30,
num_return_sequences= 1)
print(tokenizer.batch_decode(outputs)[0])
###########OUTPUT###########
def hello_world():
return "Hello World!"
@app.route("/hello_world")
def hello_world():
return "Hello World!"
Designed and Developed with ♥ by Praneet | LinkedIn | GitHub
- Downloads last month
- 400
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.