pycocotools
Official APIs for the MS-COCO dataset
Decision gist · record as of 2026-08-14
Yes, if you work with COCO-format datasets. The package is actively maintained, widely used (top 5000 on PyPI), has no known vulnerabilities, and the fork fixes real issues in the original. Verify the FreeBSD license terms for your use case before committing to a production dependency.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires Python >= 3.9.
- Compiled extension; installation may require build tools on platforms without pre-built wheels.
- Medium install friction due to compiled components, but pre-built wheels are available for common platforms (macOS, Linux x86_64/aarch64, Windows).
License · maintenance · safety
FreeBSD (unclear) — License treatment is unclear; the raw license is listed as FreeBSD but SPDX mapping is absent. Verify the actual license terms before use in proprietary or restricted-license projects.
last release 2025-12-15 (242 days) · last repo commit 2026-03-23 · 171 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 5,686,892 downloads/mo, #2,053 on PyPI
Alternatives
Verify before relying
pip install pycocotools
from pycocotools.coco import COCO
import numpy as np
coco = COCO('annotations.json')
img_ids = coco.getImgIds()
img = coco.loadImgs(img_ids[0])[0]- Exact scope of bug fixes and improvements over the original cocoapi—which specific use cases benefit most.
- Whether the FreeBSD license is correctly mapped and what restrictions it imposes in commercial contexts.
- Performance characteristics when working with very large annotation files or datasets.
What it is and what it does
pycocotools is a maintained fork of the official MS-COCO dataset API, providing Python bindings to load and manipulate COCO-format annotations for object detection, instance segmentation, and keypoint detection tasks. It wraps the original C++ implementation and depends only on numpy. The package includes utilities to parse JSON annotation files, query images and annotations, compute evaluation metrics, and encode/decode run-length encoded segmentation masks.
The fork addresses long-standing issues in the original cocoapi: it installs cleanly via pip on Windows and modern Python versions, avoids unnecessary matplotlib imports, fixes file handle leaks, and resolves segmentation faults in RLE decoding. It maintains API compatibility with the original to ensure existing code continues to work without modification.
Use it for
- Load COCO-format annotations and retrieve image metadata and ground-truth bounding boxes for training object detection models.
- Evaluate detection and segmentation model predictions against COCO ground truth using standard metrics (AP, AR).
- Decode and manipulate instance segmentation masks stored in run-length encoded format within COCO JSON files.
- Query and filter annotations by category, image ID, or area to prepare dataset splits for training and validation.
- Integrate COCO dataset loading into computer vision pipelines that require standard annotation formats.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you work with COCO-format datasets.
The package is actively maintained, widely used (top 5000 on PyPI), has no known vulnerabilities, and the fork fixes real issues in the original. Verify the FreeBSD license terms for your use case before committing to a production dependency.
Install
pycocotools on PyPI
Before you install
Medium install friction due to compiled components, but pre-built wheels are available for common platforms (macOS, Linux x86_64/aarch64, Windows). Last release was 242 days ago; repository is active with recent commits.
Requires Python >= 3.9. Compiled extension; installation may require build tools on platforms without pre-built wheels.
License in practice
License treatment is unclear; the raw license is listed as FreeBSD but SPDX mapping is absent. Verify the actual license terms before use in proprietary or restricted-license projects.
Quickstart
pip install pycocotools
from pycocotools.coco import COCO
import numpy as np
coco = COCO('annotations.json')
img_ids = coco.getImgIds()
img = coco.loadImgs(img_ids[0])[0]Verify before relying
- Exact scope of bug fixes and improvements over the original cocoapi—which specific use cases benefit most.
- Whether the FreeBSD license is correctly mapped and what restrictions it imposes in commercial contexts.
- Performance characteristics when working with very large annotation files or datasets.
Package facts
| License | FreeBSD unclear |
| Python support | Supports the current Python release >=3.9 |
| Install friction | Medium. Platform-specific wheel |
| Runtime dependencies | 1 packagenumpy |
| Maintenance | Actively maintained 242 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 5,686,892 / month, #2,053 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
Evidence: pycocotools-2.0.11-cp310-cp310-macosx_10_9_universal2.whl; pycocotools-2.0.11-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl; pycocotools-2.0.11-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl; pycocotools-2.0.11-cp310-cp310-musllinux_1_2_aarch64.whl; pycocotools-2.0.11-cp310-cp310-musllinux_1_2_x86_64.whl; pycocotools-2.0.11-cp310-cp310-win_amd64.whl; pycocotools-2.0.11-cp310-cp310-win_arm64.whl; pycocotools-2.0.11-cp311-cp311-macosx_10_9_universal2.whl; pycocotools-2.0.11-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl; pycocotools-2.0.11-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl; pycocotools-2.0.11-cp311-cp311-musllinux_1_2_aarch64.whl; pycocotools-2.0.11-cp311-cp311-musllinux_1_2_x86_64.whl; pycocotools-2.0.11-cp311-cp311-win_amd64.whl; pycocotools-2.0.11-cp311-cp311-win_arm64.whl; pycocotools-2.0.11-cp312-abi3-macosx_10_13_universal2.whl; pycocotools-2.0.11-cp312-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl; pycocotools-2.0.11-cp312-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl; pycocotools-2.0.11-cp312-abi3-musllinux_1_2_aarch64.whl; pycocotools-2.0.11-cp312-abi3-musllinux_1_2_x86_64.whl; pycocotools-2.0.11-cp312-abi3-win_amd64.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “coco dataset api”
- pycocotoolsProvides Python APIs for loading, parsing, and working with the…
- faster-coco-evalFaster-COCO-Eval provides a high-performance replacement for COCO…
- pycocoevalcapEvaluates image captions using multiple automatic metrics (BLEU,…
Give your agent the search over MCP, or paste the wish link into any chat.
More Artificial Intelligence packages
LiteLLM provides a unified Python interface to call 100+ LLM providers (OpenAI, Anthropic, Gemini, Bedrock, Azure, and others) using OpenAI-compatible API format, available as both a Python SDK and a self-hosted AI Gateway proxy server.
Install it if you need to work with multiple LLM providers or want to centralize LLM routing in your organization.
Client library and CLI tool for downloading, uploading, and managing models, datasets, and repositories on the Hugging Face Hub platform.
Install it if you work with Hugging Face Hub models or datasets.
LangChain provides a framework for building agents and LLM-powered applications by composing language models, tools, and memory through a unified API that abstracts over multiple model providers.
hf-xet provides chunk-based deduplication and efficient file transfer for the Hugging Face Hub, enabling faster uploads and downloads of large files with local disk caching.
Tokenizers converts raw text into token sequences for NLP models, with support for training custom vocabularies and using pre-built tokenizers (BPE, WordPiece) optimized for speed via Rust.
Transformers provides a unified framework for loading, fine-tuning, and running state-of-the-art pretrained models across text, vision, audio, video, and multimodal tasks using PyTorch, JAX, or TensorFlow.
Install it if you need to run or train any transformer-based model for NLP, vision, audio, or multimodal tasks.
See also faster-coco-eval · pycocoevalcap · clip-benchmark · pybboxes · supervision · nuscenes-devkit · cvat-sdk · ultralytics · sahi · groundingdino-py