Instructions to use nirajsaran/AdTextGeneration with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nirajsaran/AdTextGeneration with Transformers:

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

pipe = pipeline("text-generation", model="nirajsaran/AdTextGeneration")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("nirajsaran/AdTextGeneration")
model = AutoModelForCausalLM.from_pretrained("nirajsaran/AdTextGeneration", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use nirajsaran/AdTextGeneration with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/nirajsaran/AdTextGeneration

SGLang

How to use nirajsaran/AdTextGeneration 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 "nirajsaran/AdTextGeneration" \
    --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": "nirajsaran/AdTextGeneration",
		"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 "nirajsaran/AdTextGeneration" \
        --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": "nirajsaran/AdTextGeneration",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use nirajsaran/AdTextGeneration with Docker Model Runner:
```
docker model run hf.co/nirajsaran/AdTextGeneration
```

AdTextGeneration / README.md

nirajsaran

Update README.md

5bbb6d8 about 4 years ago

preview code

Raw

History Blame

1.4 kB

	---
	license: mit

	inference:
	parameters:
	temperature: 0.7
	use_cache: false
	max_length: 200
	top_k: 5
	top_p: 0.9

	widget:
	- text: "Sony TV"
	example_title: "Amazon Ad text Electronics"
	- text: "Apple Watch"
	example_title: "Amazon Ad text Wearables"
	- text: "Last minute shopping for Samsung headphones for"
	example_title: "Ads for shopping deals"
	- text: "Labor Day discounts for"
	example_title: "Ads for Holiday deals"

	metrics:
	- bleu
	- sacrebleu

	---

	Generates Ad copy, currently for ads for Amazon shopping (fine tuned for electronics and wearables).

	Usage Examples:
	Enter the bolded text below to get the Amazon ad generated by the model.

	Big savings on the new Roku Streaming Device
	Mothers Day discounts for Apple Watch Wireless Charger USB Charging Cable
	Big savings on the new Sony
	Last minute shopping for Samsung headphones for

	You can try entering brand and product names like Samsung Galaxy to see the ad text generator in action.


	Currently fine tuned on the EleutherAI/gpt-neo-125M model

	Model Performance:
	The model does quite well on the Electronics and Wearables categories on which it has been fine-tuned. There are, however, occasional hallucinations, though the ad copy is mostly coherent.

	In other domains, it doesn't do quite as well...
	Tesla for Christmas today,
	Honda on sale