2023-06-17 03:07:40 -07:00
|
|
|
# Benchmarking vLLM
|
2023-05-28 03:20:05 -07:00
|
|
|
|
|
|
|
|
## Downloading the ShareGPT dataset
|
|
|
|
|
|
|
|
|
|
You can download the dataset by running:
|
2025-02-08 20:25:15 +08:00
|
|
|
|
2023-05-28 03:20:05 -07:00
|
|
|
```bash
|
|
|
|
|
wget https://huggingface.co/datasets/anon8231489123/ShareGPT_Vicuna_unfiltered/resolve/main/ShareGPT_V3_unfiltered_cleaned_split.json
|
|
|
|
|
```
|
2024-11-05 11:30:02 -08:00
|
|
|
|
|
|
|
|
## Downloading the ShareGPT4V dataset
|
|
|
|
|
|
|
|
|
|
The json file refers to several image datasets (coco, llava, etc.). The benchmark scripts
|
|
|
|
|
will ignore a datapoint if the referred image is missing.
|
2025-02-08 20:25:15 +08:00
|
|
|
|
2024-11-05 11:30:02 -08:00
|
|
|
```bash
|
|
|
|
|
wget https://huggingface.co/datasets/Lin-Chen/ShareGPT4V/resolve/main/sharegpt4v_instruct_gpt4-vision_cap100k.json
|
|
|
|
|
mkdir coco -p
|
|
|
|
|
wget http://images.cocodataset.org/zips/train2017.zip -O coco/train2017.zip
|
|
|
|
|
unzip coco/train2017.zip -d coco/
|
|
|
|
|
```
|
