Instructions to use mrutyunjay-patil/keywordGen-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mrutyunjay-patil/keywordGen-v1 with Transformers:

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

pipe = pipeline("text-generation", model="mrutyunjay-patil/keywordGen-v1")

# Load model directly
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

tokenizer = AutoTokenizer.from_pretrained("mrutyunjay-patil/keywordGen-v1")
model = AutoModelForSeq2SeqLM.from_pretrained("mrutyunjay-patil/keywordGen-v1", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use mrutyunjay-patil/keywordGen-v1 with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/mrutyunjay-patil/keywordGen-v1

SGLang

How to use mrutyunjay-patil/keywordGen-v1 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 "mrutyunjay-patil/keywordGen-v1" \
    --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": "mrutyunjay-patil/keywordGen-v1",
		"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 "mrutyunjay-patil/keywordGen-v1" \
        --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": "mrutyunjay-patil/keywordGen-v1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use mrutyunjay-patil/keywordGen-v1 with Docker Model Runner:
```
docker model run hf.co/mrutyunjay-patil/keywordGen-v1
```

YAML Metadata Warning:The pipeline tag "text2text-generation" is not in the official list: text-classification, token-classification, table-question-answering, question-answering, zero-shot-classification, translation, summarization, feature-extraction, text-generation, fill-mask, sentence-similarity, text-to-speech, text-to-audio, automatic-speech-recognition, audio-to-audio, audio-classification, audio-text-to-text, voice-activity-detection, depth-estimation, image-classification, object-detection, image-segmentation, text-to-image, image-to-text, image-to-image, image-to-video, unconditional-image-generation, video-classification, reinforcement-learning, robotics, tabular-classification, tabular-regression, tabular-to-text, table-to-text, multiple-choice, text-ranking, text-retrieval, time-series-forecasting, text-to-video, image-text-to-text, image-text-to-image, image-text-to-video, visual-question-answering, document-question-answering, zero-shot-image-classification, graph-ml, mask-generation, zero-shot-object-detection, text-to-3d, image-to-3d, image-feature-extraction, video-text-to-text, keypoint-detection, visual-document-retrieval, any-to-any, video-to-video, other

KeywordGen-v1 Model

KeywordGen-v1 is a T5-based model fine-tuned for keyword generation from a piece of text. Given an input text, the model will return relevant keywords.

Model details

This model was trained using the T5 base model, and was fine-tuned on a custom dataset. The training data consists of text and corresponding keywords. The model generates keywords by predicting the relevant words or phrases present in the input text.

Important Usage Note

This model is optimized for processing larger inputs and will generate 3 keywords as output. For the most accurate results, I recommend using inputs of at least 4-5 sentences. Inputs shorter than this may lead to suboptimal keyword generation.

Suggestion for Usage

This model was made to generate keywords from reviews. To perform efficiently combine multiple reviews as one and give it as input to the model.

How to use

You can use this model in your application using the Hugging Face Transformers library. Make sure to prefix your input with "Keyword: " for the model to generate keywords. Here is an example:

from transformers import T5TokenizerFast, T5ForConditionalGeneration

# Load the tokenizer and model
tokenizer = T5TokenizerFast.from_pretrained('mrutyunjay-patil/keywordGen-v1')
model = T5ForConditionalGeneration.from_pretrained('mrutyunjay-patil/keywordGen-v1')

# Define the input text
input_text = "Keyword: I recently purchased the new headphones and they are incredible. The sound quality is superb, providing crystal clear audio in all ranges. The noise-cancelling feature is very effective, blocking out almost all ambient noise. I also love the comfortable design - they fit perfectly over my ears and don't cause any discomfort, even after long periods of use. The battery life is also impressive, lasting up to 20 hours on a single charge. Overall, I'm extremely satisfied with this product."

# Encode the input text
input_ids = tokenizer.encode(input_text, return_tensors='pt')

# Generate the keywords
outputs = model.generate(input_ids)

# Decode the outputs
keywords = tokenizer.decode(outputs[0])

Limitations and bias

As this is the first version, the model might perform poorly on texts that are very different from the texts in the training data. It might also be biased towards the types of text or keywords that are overrepresented in the training data.

Downloads last month: 6