import argparse import logging as log from pprint import pformat import sys import time sys.path.insert(0, '.') from engine._gain_pagesize import GainPageSize from utils.envs import initEnv if __name__ == '__main__': parser = argparse.ArgumentParser(description='patent spider: an excellent spider program') parser.add_argument('patent_class', help='patent class', default=None, choices=['publish','authorization','utility_model','design']) args = parser.parse_args() config = initEnv(patent_class=args.patent_class) log.info('config\n\n%s\n' % pformat(config)) eng = GainPageSize(config) # run eng eng.start_spider()
import time from statistics import mean import numpy as np import torch from torchvision import transforms as tf from pprint import pformat import sys sys.path.insert(0, '.') import brambox.boxes as bbb import vedanet as vn from utils.envs import initEnv if __name__ == '__main__': parser = argparse.ArgumentParser( description='OneDet: an one stage framework based on PyTorch') parser.add_argument('model_name', help='model name', default=None) args = parser.parse_args() train_flag = 0 config = initEnv(train_flag=train_flag, model_name=args.model_name) log.info('Config\n\n%s\n' % pformat(config)) # init env hyper_params = vn.hyperparams.HyperParams(config, train_flag=train_flag) # init and run eng vn.engine.speed(hyper_params)
resize = 1 thresh = 0.3 print(result) for label in result.keys(): for value in result[label]: score = value[0] if score>thresh: bbox = np.int0(value[1]) cv2.rectangle(src,(bbox[0],bbox[1]),(bbox[2],bbox[3]),(255,255,0),4,2) cv2.putText(src,'%s %.2f'%(label,score),(bbox[0], bbox[1]-2), cv2.FONT_HERSHEY_COMPLEX,1, (255,255,0), 3,1) end2_time = time.time() print("net:",end1_time-start_time) print("net+draw:",end2_time-start_time) dst = cv2.resize(src,(src.shape[1]//resize,src.shape[0]//resize),cv2.INTER_CUBIC) cv2.imshow("dst",dst) cv2.waitKey(0) if __name__ == '__main__': # parser = argparse.ArgumentParser(description='OneDet: an one stage framework based on PyTorch') # parser.add_argument('model_name', help='model name', default='Yolov3') # args = parser.parse_args() train_flag = 2 config = initEnv(train_flag=train_flag, model_name='Yolov3') log.info('Config\n\n%s\n' % pformat(config)) # init env hyper_params = vn.hyperparams.HyperParams(config, train_flag=train_flag) # init and run eng mytest(hyper_params)
import sys sys.path.insert(0, '.') import brambox.boxes as bbb import vedanet as vn from utils.envs import initEnv, randomSeeding if __name__ == '__main__': parser = argparse.ArgumentParser(description='OneDet: an one stage framework based on PyTorch') parser.add_argument('model_name', help='model name', default=None) args = parser.parse_args() train_flag = 1 config = initEnv(train_flag=train_flag, model_name=args.model_name) #randomSeeding(0) log.info('Config\n\n%s\n' % pformat(config)) # init env hyper_params = vn.hyperparams.HyperParams(config, train_flag=train_flag) # int eng eng = vn.engine.VOCTrainingEngine(hyper_params) # run eng b1 = eng.batch t1 = time.time() eng() t2 = time.time()
test_loss / len(cifar100_test_loader.dataset), epoch) writer.add_scalar('Test/Accuracy', correct.float() / len(cifar100_test_loader.dataset), epoch) return correct.float() / len(cifar100_test_loader.dataset) if __name__ == '__main__': parser = argparse.ArgumentParser() parser.add_argument('model_name', help='model name', default=None) args = parser.parse_args() config = initEnv(args.model_name, train_flag=1) net = get_network(args, use_gpu=True) cifar100_training_loader = get_training_dataloader( config['CIFAR100_TRAIN_MEAN'], config['CIFAR100_TRAIN_STD'], num_workers=config['nworkers'], batch_size=config['batch_size'], shuffle=config['shuffle']) cifar100_test_loader = get_test_dataloader(config['CIFAR100_TRAIN_MEAN'], config['CIFAR100_TRAIN_STD'], num_workers=config['nworkers'], batch_size=config['batch_size'], shuffle=config['shuffle'])