Instructions to use jinaai/starcoder-1b-textbook with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use jinaai/starcoder-1b-textbook with Transformers:

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

pipe = pipeline("text-generation", model="jinaai/starcoder-1b-textbook")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("jinaai/starcoder-1b-textbook")
model = AutoModelForCausalLM.from_pretrained("jinaai/starcoder-1b-textbook")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use jinaai/starcoder-1b-textbook with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "jinaai/starcoder-1b-textbook"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jinaai/starcoder-1b-textbook",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/jinaai/starcoder-1b-textbook

SGLang

How to use jinaai/starcoder-1b-textbook 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 "jinaai/starcoder-1b-textbook" \
    --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": "jinaai/starcoder-1b-textbook",
		"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 "jinaai/starcoder-1b-textbook" \
        --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": "jinaai/starcoder-1b-textbook",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use jinaai/starcoder-1b-textbook with Docker Model Runner:
```
docker model run hf.co/jinaai/starcoder-1b-textbook
```

samsja commited on Sep 6, 2023

Commit

ba33f58

1 Parent(s): a9c43b5

Update README.md

Browse files

Files changed (1) hide show

README.md +11 -1

README.md CHANGED Viewed

@@ -32,7 +32,14 @@ That is an improvement of almost 12 points over the starcoder 1b baseline, almos
 The results (on the human eval benchmark) are on par with other open-source models like StarCoderBase (30.4) StarCoder(33.6) CodeGen-16B-Mono(29.3) while the model being 15 times smaller.
-It still underperforms compared to other models like CodeLLama (53%) or chat gpt 4 (82) or wizard coder (73.2), but these model are more than 30 times bigger.
 ## Disclaimer
@@ -43,3 +50,6 @@ It still underperforms compared to other models like CodeLLama (53%) or chat gpt
 * This model has not been aligned with human preferences and therefore could potentially generate harmful content
 * This model has been trained on a dataset generated by ChatGPT 3.5, and you should check the legal status of AI-generated content in your jurisdiction before using it. You should make sure that your usage complies with the OpenAI Terms of Use, in so far as legally applicable.

 The results (on the human eval benchmark) are on par with other open-source models like StarCoderBase (30.4) StarCoder(33.6) CodeGen-16B-Mono(29.3) while the model being 15 times smaller.
+It still underperforms compared to other models like CodeLLama (53%) chat gpt 4 (82) or wizard coder (73.2), but these model are more than 30 times bigger.
+## Finetuning details
+We did full parameter fine-tuning and used a Nvidia a40 for 12 hours using a batch size of 128 and a micro-batch size of 8.
+To reproduce the training just follow the training instructions in our [open source codebase](https://github.com/jina-ai/textbook)
 ## Disclaimer
 * This model has not been aligned with human preferences and therefore could potentially generate harmful content
 * This model has been trained on a dataset generated by ChatGPT 3.5, and you should check the legal status of AI-generated content in your jurisdiction before using it. You should make sure that your usage complies with the OpenAI Terms of Use, in so far as legally applicable.
+## Credits
+This model was trained and released by [Jina.ai](https://jina.ai/)