Instructions to use hasan-toontech/codegen-phaser with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use hasan-toontech/codegen-phaser with Transformers:

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

pipe = pipeline("text-generation", model="hasan-toontech/codegen-phaser")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("hasan-toontech/codegen-phaser")
model = AutoModelForCausalLM.from_pretrained("hasan-toontech/codegen-phaser", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use hasan-toontech/codegen-phaser with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "hasan-toontech/codegen-phaser"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "hasan-toontech/codegen-phaser",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/hasan-toontech/codegen-phaser

SGLang

How to use hasan-toontech/codegen-phaser with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "hasan-toontech/codegen-phaser" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "hasan-toontech/codegen-phaser",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "hasan-toontech/codegen-phaser" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "hasan-toontech/codegen-phaser",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use hasan-toontech/codegen-phaser with Docker Model Runner:
```
docker model run hf.co/hasan-toontech/codegen-phaser
```

codegen-phaser / README.md

hasan-toontech

Update README.md

4edf9ac verified almost 2 years ago

preview code

Raw

History Blame Contribute Delete

2.89 kB

	---
	library_name: transformers
	tags: []
	---

	### Model Description

	CodeGen-Mono 350M Fine-Tuned for Phaser.js
	This model is a fine-tuned version of the CodeGen-Mono 350M, specifically optimized for generating Phaser.js code based on user prompts.
	Phaser.js is a popular 2D game framework used for creating web-based games, and this model aims to assist developers by generating relevant
	game code snippets, enhancing productivity and creativity.


	- Developed by: Toontech Ltd.
	- Funded by: Toontech Ltd.
	- Language(s) (NLP): English
	- Finetuned from model: CodeGen (CodeGen-Mono 350M)

	Model Overview

	Base Model: CodeGen-Mono 350M
	Training Data: Fine-tuned on a custom dataset of Phaser.js code examples, covering various aspects like User Input Handling,
	Audio Management, Text Display, Tilemap Handling, Scene Management, Camera Controls, and Particles and Effects.

	Purpose: To generate Phaser.js code snippets based on natural language prompts, helping developers to quickly prototype and implement game
	features.

	Key Features
	User-Friendly: Accepts plain English prompts and translates them into functional Phaser.js code.
	Versatile: Capable of handling a variety of game development tasks, from simple input handling to complex scene management.
	Optimized for Game Development: Specifically fine-tuned on a wide range of Phaser.js use cases to provide accurate and efficient code
	generation.

	Use Cases
	Game Development Prototyping: Quickly generate code snippets to prototype game mechanics and features.
	Learning Tool: A valuable resource for beginners learning Phaser.js, providing examples and code structure.
	Coding Assistance: Acts as an intelligent code assistant for experienced developers, helping to reduce development time and improve
	workflow.

	How to Use
	You can use this model with the Hugging Face transformers library or integrate it into your custom applications using the provided API.
	Here’s a basic example:

	from transformers import pipeline

	codegen_pipeline = pipeline("text2text-generation", model="your-huggingface-username/CodeGen-Mono-350M-Phaser")

	prompt = "Create a Phaser.js code snippet to move a sprite with arrow keys."
	response = codegen_pipeline(prompt)

	print(response[0]['generated_text'])

	Model Limitations
	Phaser.js Specific: The model is specialized for Phaser.js and may not generalize well to other frameworks or coding languages.
	Code Quality: Generated code might require additional refinement and testing for edge cases.

	Future Work
	Planned improvements include expanding the training dataset, incorporating more advanced game development scenarios, and enhancing model
	accuracy and usability.

	Feedback and Contributions
	Your feedback is invaluable! If you encounter any issues or have suggestions, please feel free to open an issue or contribute directly to
	the repository.