Instructions to use kazuma313/code-instruct-facebook-opt-350m with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kazuma313/code-instruct-facebook-opt-350m with Transformers:

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

pipe = pipeline("text-generation", model="kazuma313/code-instruct-facebook-opt-350m")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("kazuma313/code-instruct-facebook-opt-350m")
model = AutoModelForCausalLM.from_pretrained("kazuma313/code-instruct-facebook-opt-350m")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use kazuma313/code-instruct-facebook-opt-350m with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kazuma313/code-instruct-facebook-opt-350m"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kazuma313/code-instruct-facebook-opt-350m",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/kazuma313/code-instruct-facebook-opt-350m

SGLang

How to use kazuma313/code-instruct-facebook-opt-350m 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 "kazuma313/code-instruct-facebook-opt-350m" \
    --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": "kazuma313/code-instruct-facebook-opt-350m",
		"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 "kazuma313/code-instruct-facebook-opt-350m" \
        --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": "kazuma313/code-instruct-facebook-opt-350m",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use kazuma313/code-instruct-facebook-opt-350m with Docker Model Runner:
```
docker model run hf.co/kazuma313/code-instruct-facebook-opt-350m
```

results

This model is a fine-tuned version of facebook/opt-350m on CodeAlpaca-20k with 601 rows.

Model description

More information needed

Intended uses & limitations

this use only 601 data from CodeAlpaca-20k so need to train further.
use tag "### Question: " for asking/instruct and it will generate "### Answer: " for inference.
the dataset is about coding with various languages, so need to train with specify programming language.

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 1e-05
train_batch_size: 2
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.03
num_epochs: 3

Framework versions

Transformers 4.38.2
Pytorch 2.2.1+cu121
Datasets 2.18.0
Tokenizers 0.15.2

Downloads last month: 8

Safetensors

Model size

0.3B params

Tensor type

F32

Model tree for kazuma313/code-instruct-facebook-opt-350m

Base model

facebook/opt-350m

Finetuned

(155)

this model