wk-classify 0.0.0.2

A package of tools for building deep-learning classification programs.

wk-classify

A package of tools for building deep-learning classification programs. Easy to use, light and powerful.

Install

pip3 install wk-classify

Usage

quick experience

from wcf.packages.resnet.training import train, BaseConfig
class Config(BaseConfig):
    TRAIN_DIR = 'path for train set'
    VAL_DIR = 'path for val set'
cfg=Config()
train(cfg)

a real example

from wcf.packages.resnet.training import train, BaseConfig
from torchvision import transforms
class Config(BaseConfig):
    GEN_CLASSES_FILE = True
    USE_tqdm_TRAIN = False # use tqdm to format output
    INPUT_SIZE = (252,196)
    BATCH_SIZE = 16
    NUM_EPOCHS = 50
    BALANCE_CLASSES = True
    VAL_INTERVAL = 0.2 # val time insterval: 0.2 epoch (0.2* num_batches_per_epoch)
    WEIGHTS_SAVE_INTERVAL = 0.2 #  the same as above
    TRAIN_DIR = '<your train path>'
    VAL_DIR = '<your val path>'
    train_transform = transforms.Compose([
        transforms.ColorJitter(brightness=0.1, contrast=0.1, saturation=0.1, hue=0.5),
        transforms.Resize(INPUT_SIZE),
        transforms.ToTensor(),
        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
    ])
    val_transform = transforms.Compose([
        transforms.Resize(INPUT_SIZE),
        transforms.ToTensor(),
        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
    ])
cfg=Config()
train(cfg)

all options

check out the BaseConfig class for all options

how to predict?

check out demo_predict.py