Instructions to use madatnlp/skgpt-base-kormath with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use madatnlp/skgpt-base-kormath with Transformers:

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

pipe = pipeline("text-generation", model="madatnlp/skgpt-base-kormath")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("madatnlp/skgpt-base-kormath")
model = AutoModelForCausalLM.from_pretrained("madatnlp/skgpt-base-kormath")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use madatnlp/skgpt-base-kormath with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/madatnlp/skgpt-base-kormath

SGLang

How to use madatnlp/skgpt-base-kormath 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 "madatnlp/skgpt-base-kormath" \
    --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": "madatnlp/skgpt-base-kormath",
		"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 "madatnlp/skgpt-base-kormath" \
        --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": "madatnlp/skgpt-base-kormath",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use madatnlp/skgpt-base-kormath with Docker Model Runner:
```
docker model run hf.co/madatnlp/skgpt-base-kormath
```

madatnlp/skgpt-base-kormath

This model is a fine-tuned version of madatnlp/sk-kogptv2-kormath-causal on an unknown dataset. It achieves the following results on the evaluation set:

Train Loss: 0.4912
Validation Loss: 1.1109
Epoch: 25

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

optimizer: {'name': 'Adam', 'learning_rate': 5.3799995e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
training_precision: float32

Training results

Train Loss	Validation Loss	Epoch
2.0777	1.5042	0
1.4350	1.1904	1
1.3248	1.2871	2
1.1661	1.4441	3
1.0920	1.0973	4
1.0431	1.3860	5
0.9541	1.2228	6
0.9315	1.0385	7
0.8875	1.2156	8
0.8838	1.0195	9
0.8029	1.1956	10
0.7533	1.1139	11
0.7526	1.4868	12
0.6986	1.1045	13
0.6999	1.1083	14
0.6462	1.0082	15
0.6325	1.0643	16
0.6350	1.0729	17
0.6373	1.0455	18
0.5922	1.2834	19
0.5606	1.1031	20
0.5241	1.3085	21
0.5394	1.2911	22
0.5486	1.2407	23
0.5239	1.2075	24
0.4912	1.1109	25

Framework versions

Transformers 4.19.2
TensorFlow 2.8.0
Datasets 2.2.1
Tokenizers 0.12.1

Downloads last month: 3