test code

Author: ajithvcoder

config.py

@@ -22,5 +22,6 @@ def argument_parser():
  parser.add_argument('--device', default=0, type=str, help='gpu device ids for CUDA_VISIBLE_DEVICES')
  parser.add_argument("--redirector", action='store_false')
  parser.add_argument('--use_bn', action='store_false')
+ parser.add_argument("--checkpoint", type=str)
 
  return parser

test.py

@@ -0,0 +1,150 @@
+#%%writefile test.py
+
+import os
+import pprint
+from collections import OrderedDict, defaultdict
+
+import numpy as np
+import torch
+from torch.optim.lr_scheduler import ReduceLROnPlateau
+from torch.utils.data import DataLoader
+
+from batch_engine import valid_trainer, batch_trainer
+from config import argument_parser
+from dataset.AttrDataset import AttrDataset, get_transform
+from loss.CE_loss import CEL_Sigmoid
+from models.base_block import FeatClassifier, BaseClassifier
+from models.vgg import vgg16
+from tools.function import get_model_log_path, get_pedestrian_metrics
+from tools.utils import time_str, save_ckpt, ReDirectSTD, set_seed
+
+set_seed(605)
+
+
+def main(args):
+ visenv_name = args.dataset
+ exp_dir = os.path.join('exp_result', args.dataset)
+ model_dir, log_dir = get_model_log_path(exp_dir, visenv_name)
+ stdout_file = os.path.join(log_dir, f'stdout_{time_str()}.txt')
+ 
+ checkpoint_file = args.checkpoint
+
+ if args.redirector:
+ print('redirector stdout')
+ ReDirectSTD(stdout_file, 'stdout', False)
+
+ pprint.pprint(OrderedDict(args.__dict__))
+
+ print('-' * 60)
+ print(f'use GPU{args.device} for testing')
+ #print(f'train set: {args.dataset} {args.train_split}, test set: {args.valid_split}')
+
+ train_tsfm, valid_tsfm = get_transform(args)
+ #print(train_tsfm)
+
+ train_set = AttrDataset(args=args, split=args.train_split, transform=train_tsfm)
+
+ train_loader = DataLoader(
+ dataset=train_set,
+ batch_size=args.batchsize,
+ shuffle=True,
+ num_workers=4,
+ pin_memory=True,
+ )
+ valid_set = AttrDataset(args=args, split=args.valid_split, transform=valid_tsfm)
+
+ valid_loader = DataLoader(
+ dataset=valid_set,
+ batch_size=args.batchsize,
+ shuffle=False,
+ num_workers=4,
+ pin_memory=True,
+ )
+
+ print(f'{args.train_split} set: {len(train_loader.dataset)}, '
+ f'{args.valid_split} set: {len(valid_loader.dataset)}, '
+ f'attr_num : {train_set.attr_num}')
+
+ labels = train_set.label
+ sample_weight = labels.mean(0)
+
+ backbone = vgg16()
+ classifier = BaseClassifier(nattr=train_set.attr_num)
+ model = FeatClassifier(backbone, classifier)
+ 
+ #model.load_state_dict(torch.load(filename))
+ if torch.cuda.is_available():
+ model = torch.nn.DataParallel(model).cuda()
+ 
+ #print(checkpoint_file['state_dicts'])
+ checkpoint = torch.load(checkpoint_file)
+ model.load_state_dict(checkpoint["state_dicts"])
+
+ criterion = CEL_Sigmoid(sample_weight)
+
+ param_groups = [{'params': model.module.finetune_params(), 'lr': args.lr_ft},
+ {'params': model.module.fresh_params(), 'lr': args.lr_new}]
+ optimizer = torch.optim.SGD(param_groups, momentum=args.momentum, weight_decay=args.weight_decay, nesterov=False)
+ lr_scheduler = ReduceLROnPlateau(optimizer, factor=0.1, patience=4)
+ 
+ 
+
+ tester(epoch=args.train_epoch,
+ model=model,
+ train_loader=train_loader,
+ valid_loader=valid_loader,
+ criterion=criterion,
+ optimizer=optimizer,
+ lr_scheduler=lr_scheduler,
+ )
+
+ 
+
+
+def tester(epoch, model, train_loader, valid_loader, criterion, optimizer, lr_scheduler):
+ maximum = float(-np.inf)
+ best_epoch = 0
+
+ result_list = defaultdict()
+
+ 
+ 
+ valid_loss, valid_gt, valid_probs = valid_trainer(
+ model=model,
+ valid_loader=valid_loader,
+ criterion=criterion,
+ )
+
+ lr_scheduler.step(metrics=valid_loss, epoch=1)
+
+ 
+ valid_result = get_pedestrian_metrics(valid_gt, valid_probs)
+
+ print(f'Evaluation on test set, \n',
+ 'ma: {:.4f}, pos_recall: {:.4f} , neg_recall: {:.4f} \n'.format(
+ valid_result.ma, np.mean(valid_result.label_pos_recall), np.mean(valid_result.label_neg_recall)),
+ 'Acc: {:.4f}, Prec: {:.4f}, Rec: {:.4f}, F1: {:.4f}'.format(
+ valid_result.instance_acc, valid_result.instance_prec, valid_result.instance_recall,
+ valid_result.instance_f1))
+
+ print(f'{time_str()}')
+ print('-' * 60)
+
+ 
+
+
+if __name__ == '__main__':
+ parser = argument_parser()
+
+ args = parser.parse_args()
+ print(args)
+ main(args)
+
+ # os.path.abspath()
+
+"""
+载入的时候要：
+from tools.function import LogVisual
+sys.modules['LogVisual'] = LogVisual
+log = torch.load('./save/2018-10-29_21:17:34trlog')
+"""