No project description provided
Project description
Introduction
echo1-coco-builder
provides a faster, safer way to build coco formatted data.
See: https://cocodataset.org/#format-data for more information
Installation
# If using pip
pip install echo1-coco-builder
# If using poetry
poetry add echo1-coco-builder
Example use
import pandas as pd
from echo1_coco_builder.echo1_coco_builder import CocoBuilder
# Open a CSV using pandas
df = pd.read_csv("example.csv")
# Initialize the coco generator
coco_builder = CocoBuilder()
# For each row in the csv
for annotation_id, row in df.iterrows():
# image_id must be an integer
image_id = int(re.sub("[^0-9]", "", row["image_name"]))
# image_name must be a string
image_name = row["image_name"]
# image_width and image_height must be an integer
image_width = row["image_width"]
image_height = row["image_height"]
# category_id must be an integer
category_id = row["category_id"]
# category_name must be a string
category_name = row["category_name"]
# bbox format: [x,y,width,height]
bbox = row["bbox"].split(",")
# add a new image
coco_builder.add_image(image_id, image_name, image_width, image_height)
# add a new category
coco_builder.add_category(category_id, category_name)
# add a new annotation
coco_builder.add_annotation(annotation_id, image_id, category_id, bbox, [], 0, 0)
# add info
coco_builder.add_info(2022, "v1.0", "Echo1")
# print the data in the coco format as a python object
print(coco_builder.get())
# print the data in the coco format as json
print(coco_builder.get())
# save the data in the coco format as json
python_file = open("example-data.json", "w")
python_file.write(coco_builder.get())
python_file.close()
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Close
Hashes for echo1_coco_builder-0.1.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | ca92e02dd27452ae313043c1cdff3adc808e55ddfd27ae5de3ea3f460e3279e7 |
|
MD5 | 69dac807e62e37ed82530e75199b91fe |
|
BLAKE2b-256 | 3d988b6ad771bac74977ea4053e6cce285ca39aa2de9b2d66f9273578602e536 |