Instructions to use xu1998hz/InstructScore with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use xu1998hz/InstructScore with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="xu1998hz/InstructScore")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("xu1998hz/InstructScore") model = AutoModelForCausalLM.from_pretrained("xu1998hz/InstructScore") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use xu1998hz/InstructScore with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "xu1998hz/InstructScore" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "xu1998hz/InstructScore", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/xu1998hz/InstructScore
- SGLang
How to use xu1998hz/InstructScore 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 "xu1998hz/InstructScore" \ --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": "xu1998hz/InstructScore", "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 "xu1998hz/InstructScore" \ --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": "xu1998hz/InstructScore", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use xu1998hz/InstructScore with Docker Model Runner:
docker model run hf.co/xu1998hz/InstructScore
Wenda Xu commited on
Commit ·
3752a16
1
Parent(s): e31eb42
updates readme
Browse files- README.md +26 -0
- requirements.txt +16 -0
README.md
CHANGED
|
@@ -1,3 +1,29 @@
|
|
| 1 |
---
|
| 2 |
license: cc
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: cc
|
| 3 |
---
|
| 4 |
+
|
| 5 |
+
# InstructScore (SEScore3)
|
| 6 |
+
|
| 7 |
+
An amazing explanation metric (diagnostic report) for text generation evaluation
|
| 8 |
+
|
| 9 |
+
First step, you may download all required dependencies through: pip3 install -r requirements.txt
|
| 10 |
+
|
| 11 |
+
<div align="center">
|
| 12 |
+
<img src="figs/InstructScore_teaser.jpg" width=400px>
|
| 13 |
+
</div>
|
| 14 |
+
|
| 15 |
+
To run our metric, you only need five lines
|
| 16 |
+
|
| 17 |
+
````
|
| 18 |
+
```
|
| 19 |
+
from InstructScore import *
|
| 20 |
+
refs = ["Normally the administration office downstairs would call me when there’s a delivery."]
|
| 21 |
+
outs = ["Usually when there is takeaway, the management office downstairs will call."]
|
| 22 |
+
scorer = InstructScore()
|
| 23 |
+
batch_outputs, scores_ls = scorer.score(refs, outs)
|
| 24 |
+
```
|
| 25 |
+
````
|
| 26 |
+
|
| 27 |
+

|
| 28 |
+
|
| 29 |
+
|
requirements.txt
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
sentencepiece
|
| 2 |
+
torch
|
| 3 |
+
pandas
|
| 4 |
+
matplotlib
|
| 5 |
+
sklearn
|
| 6 |
+
wandb
|
| 7 |
+
datasets
|
| 8 |
+
numpy
|
| 9 |
+
wandb
|
| 10 |
+
deepspeed
|
| 11 |
+
flask
|
| 12 |
+
flask_cors
|
| 13 |
+
daal
|
| 14 |
+
fire
|
| 15 |
+
tokenizers
|
| 16 |
+
git+https://github.com/huggingface/transformers@c612628045822f909020f7eb6784c79700813eda
|