Instructions to use Wanfq/Explore-LM-Ext-7B-Math with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Wanfq/Explore-LM-Ext-7B-Math with Transformers:

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

pipe = pipeline("text-generation", model="Wanfq/Explore-LM-Ext-7B-Math")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Wanfq/Explore-LM-Ext-7B-Math")
model = AutoModelForCausalLM.from_pretrained("Wanfq/Explore-LM-Ext-7B-Math")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Wanfq/Explore-LM-Ext-7B-Math with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Wanfq/Explore-LM-Ext-7B-Math"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Wanfq/Explore-LM-Ext-7B-Math",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Wanfq/Explore-LM-Ext-7B-Math

SGLang

How to use Wanfq/Explore-LM-Ext-7B-Math 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 "Wanfq/Explore-LM-Ext-7B-Math" \
    --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": "Wanfq/Explore-LM-Ext-7B-Math",
		"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 "Wanfq/Explore-LM-Ext-7B-Math" \
        --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": "Wanfq/Explore-LM-Ext-7B-Math",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Wanfq/Explore-LM-Ext-7B-Math with Docker Model Runner:
```
docker model run hf.co/Wanfq/Explore-LM-Ext-7B-Math
```

Wanfq commited on Oct 16, 2023

Commit

a860184

1 Parent(s): 239e3d8

Update README.md

Browse files

Files changed (1) hide show

README.md +4 -4

README.md CHANGED Viewed

@@ -198,7 +198,7 @@ We fine-tune LLaMA-7B with the following hyperparameters:
 | Hyperparameter  |  Global Batch Size |  Learning rate |  Epochs |  Max length |  Weight decay |
 |:----------------|-------------------:|---------------:|--------:|------------:|--------------:|
-| LLaMA 7B        |                128 |           2e-5 |       3 |        2048 |             0 |
 To reproduce the training procedure, please use the following command:
@@ -259,7 +259,7 @@ python3 ./eval/generate.py \
     --prompt_type alpaca \
     --do_sample
-2. Evaluation
 python3 ./eval/chatgpt_score.py \
     --baseline_file ./eval/answer/<MODEL_1>.jsonl \  # answer of baseline model to compare with
     --answer_file ./eval/answer/<MODEL_2>.jsonl \  # answer of evaluation model
@@ -285,10 +285,10 @@ python3 ./eval/generate.py \
     --max_new_tokens 512 \
     --prompt_type alpaca
-2. Evaluation
 python3 ./eval/auto_eval.py \
     --question_file ./eval/question/MATH_eval_set_sample.jsonl \
-    --answer_file ./eval/answer/<MODEL_ID>.jsonl
 ```
 ## Limitations

 | Hyperparameter  |  Global Batch Size |  Learning rate |  Epochs |  Max length |  Weight decay |
 |:----------------|-------------------:|---------------:|--------:|------------:|--------------:|
+| LLaMA 7B        |                128 |           2e-5 |       3 |          512|             0 |
 To reproduce the training procedure, please use the following command:
     --prompt_type alpaca \
     --do_sample
+# 2. Evaluation
 python3 ./eval/chatgpt_score.py \
     --baseline_file ./eval/answer/<MODEL_1>.jsonl \  # answer of baseline model to compare with
     --answer_file ./eval/answer/<MODEL_2>.jsonl \  # answer of evaluation model
     --max_new_tokens 512 \
     --prompt_type alpaca
+# 2. Evaluation
 python3 ./eval/auto_eval.py \
     --question_file ./eval/question/MATH_eval_set_sample.jsonl \
+    --answer_file ./eval/answer/<MODEL_ID>.jsonl  # answer of evaluation model
 ```
 ## Limitations