Instructions to use replicate/cv_utils with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Kernels
How to use replicate/cv_utils with Kernels:
# !pip install kernels from kernels import get_kernel kernel = get_kernel("replicate/cv_utils") - Notebooks
- Google Colab
- Kaggle
Download build.toml from replicate/cv_utils: direct link, hf CLI and curl.
- Browser
- Download file 421 Bytes
-
https://huggingface.co/replicate/cv_utils/resolve/main/build.toml
- Command line
-
hf download hf://replicate/cv_utils/build.toml
-
curl -L -o build.toml https://huggingface.co/replicate/cv_utils/resolve/main/build.toml
421 Bytes
| [general] | |
| name = "cv_utils" | |
| universal = false | |
| [torch] | |
| src = [ | |
| "torch-ext/torch_binding.cpp", | |
| "torch-ext/torch_binding.h", | |
| ] | |
| [kernel.cv_utils] | |
| depends = ["torch"] | |
| backend = "cuda" | |
| src = [ | |
| "cv_utils/connected_components.cu", | |
| "cv_utils/generic_nms.cu", | |
| ] | |
| cuda-flags = [ | |
| "-DCUDA_HAS_FP16=1", | |
| "-D__CUDA_NO_HALF_OPERATORS__", | |
| "-D__CUDA_NO_HALF_CONVERSIONS__", | |
| "-D__CUDA_NO_HALF2_OPERATORS__", | |
| ] | |