Instructions to use oliverguhr/spelling-correction-multilingual-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use oliverguhr/spelling-correction-multilingual-base with Transformers:

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

pipe = pipeline("text-generation", model="oliverguhr/spelling-correction-multilingual-base")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

tokenizer = AutoTokenizer.from_pretrained("oliverguhr/spelling-correction-multilingual-base")
model = AutoModelForSeq2SeqLM.from_pretrained("oliverguhr/spelling-correction-multilingual-base", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use oliverguhr/spelling-correction-multilingual-base with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/oliverguhr/spelling-correction-multilingual-base

SGLang

How to use oliverguhr/spelling-correction-multilingual-base 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 "oliverguhr/spelling-correction-multilingual-base" \
    --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": "oliverguhr/spelling-correction-multilingual-base",
		"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 "oliverguhr/spelling-correction-multilingual-base" \
        --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": "oliverguhr/spelling-correction-multilingual-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use oliverguhr/spelling-correction-multilingual-base with Docker Model Runner:
```
docker model run hf.co/oliverguhr/spelling-correction-multilingual-base
```

spelling-correction-multilingual-base / README.md

oliverguhr

updated sampels

1c97bf7 about 3 years ago

preview code

Raw History Blame

1.37 kB

metadata

language:
  - en
  - de
license: mit
widget:
  - text: fix:lets do a comparsion
    example_title: EN 1
  - text: fix:Their going to be here so0n
    example_title: EN 2
  - text: fix:das idst ein neuZr test
    example_title: DE 1
  - text: >-
      fix:ein dransformer isd ein mthode mit der ein compuder eine volge von
      zeichn übersetz
    example_title: DE 2
  - text: fix:can we mix the languages können wir die sprachen mischen
    example_title: EN and DE
metrics:
  - cer
pipeline_tag: text2text-generation

This is an experimental model that should fix your typos and punctuation. If you like to run your own experiments or train for a different language, take a look at the code.

Model description

This is a proof of concept spelling correction model for English and German.

Intended uses & limitations

This project is work in progress, be aware that the model can produce artefacts. You can test the model using the pipeline-interface:

from transformers import pipeline

fix_spelling_pipeline = pipeline("text2text-generation",model="oliverguhr/spelling-correction-multilingual-base")
def fix_spelling(text, max_length = 256):
    return fix_spelling_pipeline("fix:"+text,max_length = max_length)

print(fix_spelling_pipeline("can we mix the languages können wir die sprachen mischen",max_length=2048))