title: Primate Detection Gradio | |
emoji: 🐢 | |
colorFrom: red | |
colorTo: indigo | |
sdk: gradio | |
sdk_version: 4.23.0 | |
app_file: app.py | |
pinned: false | |
license: apache-2.0 | |
# Primate Detection | |
## Setup | |
1. Clone this repository through your local terminal by entering: | |
``` shell | |
git lfs install | |
git clone https://huggingface.co/spaces/caltech-animal-tracking/Primate-Detection-GPU | |
``` | |
2. Install the requirements using conda and pip. | |
``` shell | |
conda create -n primates | |
conda activate primates | |
pip install -r requirements.txt | |
``` | |
3. Change the environment variables to fit your system. | |
In ```app.py``` change ```BATCH_SIZE``` and ```CUDA_PATH``` to reflect current CUDA version and the size of your GPU. You can find your CUDA version by navigating to ```/usr/local``` on your computer. You may need to test a few different batch sizes to find one that works for your computer. | |
4. (Optional) Install FFmpeg if it's missing on your system by following the directions [here](https://ffmpeg.org/download.html). | |
5. Run ```app.py``` to install GroundingDINO locally and load the interface. | |
``` shell | |
python3 app.py | |
``` | |