Skip to content

build(deps): add s3 and gcs extras and floor fsspec (#1256) #2372

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
chethanuk wants to merge 1 commit into
base: main
Choose a base branch
from
Open

build(deps): add s3 and gcs extras and floor fsspec (#1256) #2372

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions fern/versions/main/pages/get-started/installation.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -219,6 +219,8 @@ NeMo Curator provides several installation extras to install only the components
| **inference_server** | `uv pip install nemo-curator[inference_server]` | Ray Serve + vLLM for serving LLMs alongside curation pipelines |
| **sdg_cpu** | `uv pip install nemo-curator[sdg_cpu]` | CPU-only synthetic data generation with Data Designer |
| **sdg_cuda12** | `uv pip install nemo-curator[sdg_cuda12]` | GPU-accelerated synthetic data generation with local inference server support |
| **s3** | `uv pip install nemo-curator[s3]` | `s3fs` backend for reading and writing `s3://` paths |
| **gcs** | `uv pip install nemo-curator[gcs]` | `gcsfs` backend for reading and writing `gs://` paths |

For pip installations of `inference_server` or an extra that includes it, add
the public PyTorch and vLLM CUDA 12.9 wheel indexes:
Expand Down
18 changes: 15 additions & 3 deletions pyproject.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -58,7 +58,7 @@ dependencies = [
"comment_parser",
"cosmos-xenna==0.2.0",
"datasets>=4.0.0",
"fsspec",
"fsspec>=2025.3.0",
"hydra-core",
"jieba==0.42.1",
"lmdb>=1.4",
Expand All @@ -78,6 +78,16 @@ cuda12 = [
"gpustat",
"nvidia-ml-py",
]

# Cloud storage backends for fsspec (s3:// and gs:// paths)
s3 = [
"s3fs>=2025.3.0",
]

gcs = [
"gcsfs>=2025.3.0",
]

# Opt-in cv2 features; kept off `[all]` — wheel vendors CVE-flagged ffmpeg.
cv2 = [
"opencv-python-headless",
Expand Down Expand Up @@ -273,7 +283,7 @@ interleaved_cpu = [
"open_clip_torch",
"Pillow",
"pypdfium2",
"s3fs>=2024.12.0",
"s3fs>=2025.3.0",
"timm",
]

Expand All @@ -297,11 +307,13 @@ sdg_cuda12 = [
# All dependencies
all = [
"nemo_curator[audio_cuda12]",
"nemo_curator[gcs]",
"nemo_curator[image_cuda12]",
"nemo_curator[inference_server]",
"nemo_curator[interleaved_cuda12]",
"nemo_curator[lance]",
"nemo_curator[math_cuda12]",
"nemo_curator[s3]",
"nemo_curator[sdg_cuda12]",
"nemo_curator[text_cuda12]",
"nemo_curator[translation_all]",
Expand All @@ -325,7 +337,7 @@ test = [
"pytest-loguru",
"pyyaml",
"rich",
"s3fs>=2024.12.0", # added for testing cloud fs
"s3fs>=2025.3.0", # added for testing cloud fs
"slack_sdk",
]

Expand Down
Loading