Skip to main content

Object Classifier

Project description

OBJECT CLASSIFIER

Clone this repository

  • for SSH
git clone git@github.com:1chimaruGin/Object-classifier.git
  • for https
https://github.com/1chimaruGin/Object-classifier.git

Requirements

pip install -U requirements.txt

Dataset

  • the dataset directory should be the following format.

# for example, dog vs cat classification
data/
    -train/
        dog/
            -*.jpg or *.png
        cat/
            -*.jpg or *.png
    -val/
        dog/
            -*.jpg or *.png
        cat/
            -*.jpg or *.png

Usage with Argparse

cd Object-classifier/objifier
  • Update number of classes and names in data.yaml

  • For training model(ResNet)

$ python main.py -m [mode: train] -opt [optimizer: (default='SGD')]  -epochs [epochs: (default=25)] 
  • For training model(EfficientNet)
$ python main.py -m [mode: train] -opt [optimizer]  -epochs [epochs] -backbone [backbone: efficientNet] -lvl [efficientNet level]
  • For prediction
$ python main.py -m [mode: predict] -im [input image] - backbone [backbone: ResNet or efficientNe] -lvl [efficientNet level]

Usage with YAML (via pip)

Create a YAML file as sample below:

  • For training [train.yaml]
nc: 10
# names: ['mantled_howler', 'patas_monkey', 'bald_uakari', 'japanese_macaque', 'pygmy_marmoset', 
#       'white_headed_capuchin', 'silvery_marmoset', 'common_squirrel_monkey', 'black_headed_night_monkey','nilgiri_langur' ]

names: ['airplane', 'automobile', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck']

backbone: 'efficientNet'

efftlvl: 0

optimizer: 'Adam'

mode: 'train'

epoch: 2

load: False

output: 'output'

dataset_path: null

  • For prediction [predict.yaml]
nc: 10
# names: ['mantled_howler', 'patas_monkey', 'bald_uakari', 'japanese_macaque', 'pygmy_marmoset', 
#       'white_headed_capuchin', 'silvery_marmoset', 'common_squirrel_monkey', 'black_headed_night_monkey','nilgiri_langur' ]

names: ['airplane', 'automobile', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck']

backbone: 'efficientNet'

efftlvl: 0

output: 'output'

image: 'baobao.jpg'

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for objifier, version 0.1.1
Filename, size File type Python version Upload date Hashes
Filename, size objifier-0.1.1-py3-none-any.whl (24.4 kB) File type Wheel Python version py3 Upload date Hashes View
Filename, size objifier-0.1.1.tar.gz (22.0 kB) File type Source Python version None Upload date Hashes View

Supported by

Pingdom Pingdom Monitoring Google Google Object Storage and Download Analytics Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page