Instructions to use DDIDU/ETRI_CodeLLaMA_7B_CPP with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DDIDU/ETRI_CodeLLaMA_7B_CPP with Transformers:

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

pipe = pipeline("text-generation", model="DDIDU/ETRI_CodeLLaMA_7B_CPP")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("DDIDU/ETRI_CodeLLaMA_7B_CPP")
model = AutoModelForCausalLM.from_pretrained("DDIDU/ETRI_CodeLLaMA_7B_CPP")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use DDIDU/ETRI_CodeLLaMA_7B_CPP with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "DDIDU/ETRI_CodeLLaMA_7B_CPP"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DDIDU/ETRI_CodeLLaMA_7B_CPP",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/DDIDU/ETRI_CodeLLaMA_7B_CPP

SGLang

How to use DDIDU/ETRI_CodeLLaMA_7B_CPP 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 "DDIDU/ETRI_CodeLLaMA_7B_CPP" \
    --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": "DDIDU/ETRI_CodeLLaMA_7B_CPP",
		"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 "DDIDU/ETRI_CodeLLaMA_7B_CPP" \
        --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": "DDIDU/ETRI_CodeLLaMA_7B_CPP",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use DDIDU/ETRI_CodeLLaMA_7B_CPP with Docker Model Runner:
```
docker model run hf.co/DDIDU/ETRI_CodeLLaMA_7B_CPP
```

DDIDU commited on Apr 12, 2024

Commit

526ccf0

verified ·

1 Parent(s): 060754b

Update README.md

Browse files

Files changed (1) hide show

README.md +3 -6

README.md CHANGED Viewed

@@ -20,7 +20,7 @@ model-index:
 We used LoRa to further pre-train Meta's CodeLLaMA-7B-hf model with high-quality C++ code tokens.
-Furthermore, we was fine-tuned on CodeM's C++ instruction data.
 ## Model Details
@@ -35,10 +35,7 @@ We pre-trained CodeLLaMA-7B further using 543 GB of C++ code collected online, a
 ## Requirements
 ```
-peft==0.3.0.dev0
-tokenizers==0.13.3
-transformers==4.33.0
-bitsandbytes==0.41.1
 ```
 ## How to reproduce HumanEval-X results
@@ -86,7 +83,7 @@ pipeline = transformers.pipeline(
 )
 sequences = pipeline(
-    'import socket\n\ndef ping_exponential_backoff(host: str):',
     do_sample=True,
     top_k=10,
     temperature=0.1,

 We used LoRa to further pre-train Meta's CodeLLaMA-7B-hf model with high-quality C++ code tokens.
+Furthermore, we fine-tuned on CodeM's C++ instruction data.
 ## Model Details
 ## Requirements
 ```
+pip install torch transformers accelerate
 ```
 ## How to reproduce HumanEval-X results
 )
 sequences = pipeline(
+    '#include <iostream>\n#include <vector>\n\nusing namespace std;\n\nvoid quickSort(int *data, int start, int end) {',
     do_sample=True,
     top_k=10,
     temperature=0.1,