Instructions to use cactode/gpt2_urbandict_textgen with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cactode/gpt2_urbandict_textgen with Transformers:

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

pipe = pipeline("text-generation", model="cactode/gpt2_urbandict_textgen")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("cactode/gpt2_urbandict_textgen")
model = AutoModelForCausalLM.from_pretrained("cactode/gpt2_urbandict_textgen")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use cactode/gpt2_urbandict_textgen with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "cactode/gpt2_urbandict_textgen"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "cactode/gpt2_urbandict_textgen",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/cactode/gpt2_urbandict_textgen

SGLang

How to use cactode/gpt2_urbandict_textgen 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 "cactode/gpt2_urbandict_textgen" \
    --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": "cactode/gpt2_urbandict_textgen",
		"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 "cactode/gpt2_urbandict_textgen" \
        --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": "cactode/gpt2_urbandict_textgen",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use cactode/gpt2_urbandict_textgen with Docker Model Runner:
```
docker model run hf.co/cactode/gpt2_urbandict_textgen
```

cactode commited on Oct 19, 2021

Commit

53749b0

1 Parent(s): 65d8931

Update README.md

Browse files

Files changed (1) hide show

README.md +11 -1

README.md CHANGED Viewed

@@ -1,2 +1,12 @@
 # GPT2 Fine Tuned on UrbanDictionary
-Honestly a little horrifying, but still funny.

 # GPT2 Fine Tuned on UrbanDictionary
+Honestly a little horrifying, but still funny.
+## Usage
+Use with GPT2Tokenizer. Pad token should be set to the EOS token.
+Inputs should be of the form "define <your word>: ".
+## Training Data
+All training data was obtained from [Urban Dictionary Words And Definitions on Kaggle](https://www.kaggle.com/therohk/urban-dictionary-words-dataset). Data was additionally filtered, normalized, and spell-checked.
+## Bias
+This model was trained on public internet data and will almost definitely produce offensive results. Some efforts were made to reduce this (i.e definitions with ethnic / gender-based slurs were removed), but the final model should not be trusted to produce non-offensive definitions.