def main(args):
args.cuda = args.use_cuda and torch.cuda.is_available()
train_set, validate_set, test_set, train_loader, validate_loader, test_loader = get_data.get_data_atomic(
args)
#model = models.Atomic(args)
model = models.Atomic_edge_only(args)
optimizer = torch.optim.Adam(model.parameters(), lr=args.lr)
#optimizer = torch.optim.SGD(model.parameters(), lr=args.lr)
#{'single': 0, 'mutual': 1, 'avert': 2, 'refer': 3, 'follow': 4, 'share': 5}
criterion = [
torch.nn.CrossEntropyLoss(
weight=torch.Tensor([0.05, 0.05, 0.25, 0.25, 0.25, 0.15])),
torch.nn.MSELoss()
]
# {'NA': 0, 'single': 1, 'mutual': 2, 'avert': 3, 'refer': 4, 'follow': 5, 'share': 6}
scheduler = ReduceLROnPlateau(optimizer,
factor=args.lr_decay,
patience=1,
verbose=True,
mode='max')
#--------------------------------------------------
# ------------------------
# use multi-gpu
if args.cuda and torch.cuda.device_count() > 1:
print("Now Using ", len(args.device_ids), " GPUs!")
model = torch.nn.DataParallel(model,
device_ids=args.device_ids,
output_device=args.device_ids[0]).cuda()
#model=model.cuda()
criterion[0] = criterion[0].cuda()
criterion[1] = criterion[1].cuda()
elif args.cuda:
model = model.cuda()
criterion[0] = criterion[0].cuda()
criterion[1] = criterion[1].cuda()
if args.load_best_checkpoint:
loaded_checkpoint = utils.load_best_checkpoint(args,
model,
optimizer,
path=args.resume)
if loaded_checkpoint:
args, best_epoch_acc, avg_epoch_acc, model, optimizer = loaded_checkpoint
if args.load_last_checkpoint:
loaded_checkpoint = utils.load_last_checkpoint(
args,
model,
optimizer,
path=args.resume,
version=args.model_load_version)
if loaded_checkpoint:
args, best_epoch_acc, avg_epoch_acc, model, optimizer = loaded_checkpoint
# ------------------------------------------------------------------------------
# Start Training!
since = time.time()
train_epoch_acc_all = []
val_epoch_acc_all = []
best_acc = 0
avg_epoch_acc = 0
for epoch in range(args.start_epoch, args.epochs):
train_epoch_loss, train_epoch_acc = train(train_loader, model,
criterion, optimizer, epoch,
args)
train_epoch_acc_all.append(train_epoch_acc)
val_epoch_loss, val_epoch_acc = validate(validate_loader, model,
criterion, epoch, args)
val_epoch_acc_all.append(val_epoch_acc)
print('Epoch {}/{} Training Acc: {:.4f} Validation Acc: {:.4f}'.format(
epoch, args.epochs - 1, train_epoch_acc, val_epoch_acc))
print('*' * 15)
scheduler.step(val_epoch_acc)
is_best = val_epoch_acc > best_acc
if is_best:
best_acc = val_epoch_acc
avg_epoch_acc = np.mean(val_epoch_acc_all)
utils.save_checkpoint(
{
'epoch': epoch + 1,
'state_dict': model.state_dict(),
'best_epoch_acc': best_acc,
'avg_epoch_acc': avg_epoch_acc,
'optimizer': optimizer.state_dict(),
'args': args
},
is_best=is_best,
directory=args.resume,
version='epoch_{}'.format(str(epoch)))
time_elapsed = time.time() - since
print('Training complete in {:.0f}m {:.0f}s'.format(
time_elapsed // 60, time_elapsed % 60))
print('Best Val Acc: {}, Final Avg Val Acc: {}'.format(
best_acc, avg_epoch_acc))
# ----------------------------------------------------------------------------------------------------------
# test
loaded_checkpoint = utils.load_best_checkpoint(args,
model,
optimizer,
path=args.resume)
if loaded_checkpoint:
args, best_epoch_acc, avg_epoch_acc, model, optimizer = loaded_checkpoint
test_loader.dataset.round_cnt = {
'single': 0,
'mutual': 0,
'avert': 0,
'refer': 0,
'follow': 0,
'share': 0
}
test_loss, test_acc, confmat, top2_acc = test(test_loader, model,
criterion, args)
# save test results
if not isdir(args.save_test_res):
os.mkdir(args.save_test_res)
with open(os.path.join(args.save_test_res, 'raw_test_results.pkl'),
'w') as f:
pickle.dump([test_loss, test_acc, confmat, top2_acc], f)
print("Test Acc {}".format(test_acc))
print("Top 2 Test Acc {}".format(top2_acc))
# todo: need to change the mode here!
get_metric_from_confmat(confmat, 'atomic')
def main(args):
args.cuda = args.use_cuda and torch.cuda.is_available()
train_set, validate_set, test_set, train_loader, validate_loader, test_loader = get_data.get_data_attmat(
args)
model = models.AttMat(args)
optimizer = torch.optim.Adam(model.parameters(), lr=args.lr)
criterion = torch.nn.CrossEntropyLoss()
scheduler = ReduceLROnPlateau(optimizer,
factor=args.lr_decay,
patience=1,
verbose=True,
mode='max')
#------------------------
# use multi-gpu
if args.cuda and torch.cuda.device_count() > 1:
print("Now Using ", len(args.device_ids), " GPUs!")
model = torch.nn.DataParallel(model,
device_ids=args.device_ids,
output_device=args.device_ids[0]).cuda()
criterion = criterion.cuda()
elif args.cuda:
model = model.cuda()
criterion = criterion.cuda()
if args.load_best_checkpoint:
loaded_checkpoint = utils.load_best_checkpoint(args,
model,
optimizer,
path=args.resume)
if loaded_checkpoint:
args, best_epoch_error, avg_epoch_error, model, optimizer = loaded_checkpoint
if args.load_last_checkpoint:
loaded_checkpoint = utils.load_last_checkpoint(
args,
model,
optimizer,
path=args.resume,
version=args.model_load_version)
if loaded_checkpoint:
args, best_epoch_error, avg_epoch_error, model, optimizer = loaded_checkpoint
# ------------------------------------------------------------------------------
# Start Training!
since = time.time()
train_epoch_acc_all = []
val_epoch_acc_all = []
best_acc = 0
avg_epoch_acc = 0
for epoch in range(args.start_epoch, args.epochs):
train_epoch_loss, train_epoch_acc = train(train_loader, model,
criterion, optimizer, epoch,
args)
train_epoch_acc_all.append(train_epoch_acc)
val_epoch_loss, val_epoch_acc = validate(validate_loader, model,
criterion, epoch, args)
val_epoch_acc_all.append(val_epoch_acc)
print('Epoch {}/{} Training Acc: {:.4f} Validation Acc: {:.4f}'.format(
epoch, args.epochs - 1, train_epoch_acc, val_epoch_acc))
print('*' * 15)
scheduler.step(val_epoch_acc)
is_best = val_epoch_acc > best_acc
if is_best:
best_acc = val_epoch_acc
avg_epoch_acc = np.mean(val_epoch_acc_all)
utils.save_checkpoint(
{
'epoch': epoch + 1,
'state_dict': model.state_dict(),
'best_epoch_acc': best_acc,
'avg_epoch_acc': avg_epoch_acc,
'optimizer': optimizer.state_dict(),
'args': args
},
is_best=is_best,
directory=args.resume,
version='epoch_{}'.format(str(epoch)))
time_elapsed = time.time() - since
print('Training complete in {:.0f}m {:.0f}s'.format(
time_elapsed // 60, time_elapsed % 60))
print('Best Val Acc: {}, Final Avg Val Acc: {}'.format(
best_acc, avg_epoch_acc))
#----------------------------------------------------------------------------------------------------------
# test
loaded_checkpoint = utils.load_best_checkpoint(args,
model,
optimizer,
path=args.resume)
if loaded_checkpoint:
args, best_epoch_acc, avg_epoch_acc, model, optimizer = loaded_checkpoint
#
test_loss, test_acc, one_acc, zero_acc = test(test_loader, model,
criterion, args)
print("Test Acc {}, One Acc {}, Zero Acc {}".format(
test_acc, one_acc, zero_acc))
# save test results
if not isdir(args.save_test_res):
os.mkdir(args.save_test_res)
with open(os.path.join(args.save_test_res, 'raw_test_results.pkl'),
'w') as f:
pickle.dump([test_loss, test_acc, one_acc, zero_acc], f)
def main(args):
args.cuda = args.use_cuda and torch.cuda.is_available()
train_set, validate_set, test_set, train_loader, validate_loader, test_loader = get_data.get_data_AttMat_msg_lstm(
args)
model_args = {'roi_feature_size': args.roi_feature_size, 'edge_feature_size': args.roi_feature_size,
'node_feature_size': args.roi_feature_size,
'message_size': args.message_size, 'link_hidden_size': args.link_hidden_size,
'link_hidden_layers': args.link_hidden_layers, 'propagate_layers': args.propagate_layers,
'big_attr_classes': args.big_attr_class_num, 'lstm_hidden_size': args.lstm_hidden_size}
model = models.AttMat_msg_lstm(model_args, args)
# TODO: check grads and then to set the learning rate for Adam
optimizer = torch.optim.Adam(model.parameters(), lr=args.lr)
# criterion=torch.nn.CrossEntropyLoss()
if args.cuda:
model = model.cuda()
# criterion=criterion.cuda( )
if args.load_best_checkpoint:
loaded_checkpoint = utils.load_best_checkpoint(args, model, optimizer, path=args.resume)
if loaded_checkpoint:
args, best_epoch_error, avg_epoch_error, model, optimizer = loaded_checkpoint
if args.load_last_checkpoint:
loaded_checkpoint = utils.load_last_checkpoint(args, model, optimizer, path=args.resume,
version=args.model_load_version)
if loaded_checkpoint:
args, best_epoch_error, avg_epoch_error, model, optimizer = loaded_checkpoint
train_error_history = list()
train_loss_history = list()
val_error_history = list()
val_loss_history = list()
best_epoch_error = np.inf
for epoch in range(args.start_epoch, args.epochs):
train_error_rate_cur_epoch, train_loss_cur_epoch = train(train_loader, model, criterion, optimizer, epoch, args)
train_error_history.append(train_error_rate_cur_epoch)
train_loss_history.append(train_loss_cur_epoch)
val_error_rate_cur_epoch, val_loss_cur_epoch = validate(validate_loader, model, criterion, args)
val_error_history.append(val_error_rate_cur_epoch)
val_loss_history.append(val_loss_cur_epoch)
# TODO: why use this schedule for adjusting learning rate, there is no need to decrease lr for Adam every epoch
if epoch > 0 and epoch % 1 == 0:
args.lr *= args.lr_decay
for param_group in optimizer.param_groups:
param_group['lr'] = args.lr
is_best = val_error_rate_cur_epoch < best_epoch_error
best_epoch_error = min(val_error_rate_cur_epoch, best_epoch_error)
avg_epoch_error = np.mean(val_error_history)
utils.save_checkpoint({
'epoch': epoch + 1,
'state_dict': model.state_dict(),
'best_epoch_error': best_epoch_error,
'avg_epoch_error': avg_epoch_error,
'optimizer': optimizer.state_dict(), }, is_best=is_best, directory=args.resume,
version='epoch_{}'.format(str(epoch)))
print('best_epoch_error: {}, avg_epoch_error: {}'.format(best_epoch_error, avg_epoch_error))
# test
# loaded_checkpoint=utils.load_best_checkpoint(args,model,optimizer,path=args.resume)
loaded_checkpoint = utils.load_last_checkpoint(args, model, optimizer, path=args.resume,
version=args.model_load_version)
if loaded_checkpoint:
args, best_epoch_error, avg_epoch_error, model, optimizer = loaded_checkpoint
test(test_loader, model, args)
def main(args):
args.cuda = args.use_cuda and torch.cuda.is_available()
train_set, validate_set, test_set, train_loader, validate_loader, test_loader = get_data.get_data_atomic(
args)
#model = models.Atomic(args)
model = models.Atomic_2branch(args)
optimizer = torch.optim.Adam(model.parameters(), lr=args.lr)
#{'single': 0, 'mutual': 1, 'avert': 2, 'refer': 3, 'follow': 4, 'share': 5}
criterion = [
torch.nn.CrossEntropyLoss(
weight=torch.Tensor([0.05, 0.05, 0.25, 0.25, 0.25, 0.15])),
torch.nn.MSELoss()
]
# {'NA': 0, 'single': 1, 'mutual': 2, 'avert': 3, 'refer': 4, 'follow': 5, 'share': 6}
scheduler = ReduceLROnPlateau(optimizer,
factor=args.lr_decay,
patience=1,
verbose=True,
mode='max')
#--------------------------------------------------
# ------------------------
# use multi-gpu
if args.cuda and torch.cuda.device_count() > 1:
print("Now Using ", len(args.device_ids), " GPUs!")
model = torch.nn.DataParallel(model,
device_ids=args.device_ids,
output_device=args.device_ids[0]).cuda()
#model=model.cuda()
criterion[0] = criterion[0].cuda()
criterion[1] = criterion[1].cuda()
elif args.cuda:
model = model.cuda()
criterion[0] = criterion[0].cuda()
criterion[1] = criterion[1].cuda()
# ----------------------------------------------------------------------------------------------------------
# test
loaded_checkpoint = utils.load_best_checkpoint(args,
model,
optimizer,
path=args.resume)
if loaded_checkpoint:
args, best_epoch_acc, avg_epoch_acc, model, optimizer = loaded_checkpoint
test_loader.dataset.round_cnt = {
'single': 0,
'mutual': 0,
'avert': 0,
'refer': 0,
'follow': 0,
'share': 0
}
test_loss, test_acc, confmat, top2_acc, correct_rec, error_rec = test(
test_loader, model, criterion, args)
#fw = open(os.path.join(args.tmp_root, 'correct_list.txt'), 'w')
fc_single = open(os.path.join(args.tmp_root, 'correct_single.txt'), 'w')
fc_mutual = open(os.path.join(args.tmp_root, 'correct_mutual.txt'), 'w')
fc_avert = open(os.path.join(args.tmp_root, 'correct_avert.txt'), 'w')
fc_refer = open(os.path.join(args.tmp_root, 'correct_refer.txt'), 'w')
fc_follow = open(os.path.join(args.tmp_root, 'correct_follow.txt'), 'w')
fc_share = open(os.path.join(args.tmp_root, 'correct_share.txt'), 'w')
fe_single = open(os.path.join(args.tmp_root, 'error_single.txt'), 'w')
fe_mutual = open(os.path.join(args.tmp_root, 'error_mutual.txt'), 'w')
fe_avert = open(os.path.join(args.tmp_root, 'error_avert.txt'), 'w')
fe_refer = open(os.path.join(args.tmp_root, 'error_refer.txt'), 'w')
fe_follow = open(os.path.join(args.tmp_root, 'error_follow.txt'), 'w')
fe_share = open(os.path.join(args.tmp_root, 'error_share.txt'), 'w')
for item in correct_rec:
if item[1] == '0':
for i in range(5):
fc_single.write(str(item[0][i].cpu().numpy()))
fc_single.write(' ')
fc_single.write(str(item[1].cpu().numpy()))
fc_single.write(' ')
fc_single.write(str(item[2].cpu().numpy()))
fc_single.write('\n')
elif item[1] == '1':
for i in range(5):
fc_mutual.write(str(item[0][i].cpu().numpy()))
fc_mutual.write(' ')
fc_mutual.write(str(item[1].cpu().numpy()))
fc_mutual.write(' ')
fc_mutual.write(str(item[2].cpu().numpy()))
fc_mutual.write('\n')
elif item[1] == '2':
for i in range(5):
fc_avert.write(str(item[0][i].cpu().numpy()))
fc_avert.write(' ')
fc_avert.write(str(item[1].cpu().numpy()))
fc_avert.write(' ')
fc_avert.write(str(item[2].cpu().numpy()))
fc_avert.write('\n')
elif item[1] == '3':
for i in range(5):
fc_refer.write(str(item[0][i].cpu().numpy()))
fc_refer.write(' ')
fc_refer.write(str(item[1].cpu().numpy()))
fc_refer.write(' ')
fc_refer.write(str(item[2].cpu().numpy()))
fc_refer.write('\n')
elif item[1] == '4':
for i in range(5):
fc_follow.write(str(item[0][i].cpu().numpy()))
fc_follow.write(' ')
fc_follow.write(str(item[1].cpu().numpy()))
fc_follow.write(' ')
fc_follow.write(str(item[2].cpu().numpy()))
fc_follow.write('\n')
elif item[1] == '5':
for i in range(5):
fc_share.write(str(item[0][i].cpu().numpy()))
fc_share.write(' ')
fc_share.write(str(item[1].cpu().numpy()))
fc_share.write(' ')
fc_share.write(str(item[2].cpu().numpy()))
fc_share.write('\n')
#fw2 = open(os.path.join(args.tmp_root, 'error_list.txt'), 'w')
# for item in error_rec:
# for i in range(5):
# fw2.write(str(item[0][i].cpu().numpy()))
# fw2.write(' ')
# fw2.write(str(item[1].cpu().numpy()))
# fw2.write(' ')
# fw2.write(str(item[2].cpu().numpy()))
# fw2.write('\n')
for item in error_rec:
if item[1] == '0':
for i in range(5):
fe_single.write(str(item[0][i].cpu().numpy()))
fe_single.write(' ')
fe_single.write(str(item[1].cpu().numpy()))
fe_single.write(' ')
fe_single.write(str(item[2].cpu().numpy()))
fe_single.write('\n')
elif item[1] == '1':
for i in range(5):
fe_mutual.write(str(item[0][i].cpu().numpy()))
fe_mutual.write(' ')
fe_mutual.write(str(item[1].cpu().numpy()))
fe_mutual.write(' ')
fe_mutual.write(str(item[2].cpu().numpy()))
fe_mutual.write('\n')
elif item[1] == '2':
for i in range(5):
fe_avert.write(str(item[0][i].cpu().numpy()))
fe_avert.write(' ')
fe_avert.write(str(item[1].cpu().numpy()))
fe_avert.write(' ')
fe_avert.write(str(item[2].cpu().numpy()))
fe_avert.write('\n')
elif item[1] == '3':
for i in range(5):
fe_refer.write(str(item[0][i].cpu().numpy()))
fe_refer.write(' ')
fe_refer.write(str(item[1].cpu().numpy()))
fe_refer.write(' ')
fe_refer.write(str(item[2].cpu().numpy()))
fe_refer.write('\n')
elif item[1] == '4':
for i in range(5):
fe_follow.write(str(item[0][i].cpu().numpy()))
fe_follow.write(' ')
fe_follow.write(str(item[1].cpu().numpy()))
fe_follow.write(' ')
fe_follow.write(str(item[2].cpu().numpy()))
fe_follow.write('\n')
elif item[1] == '5':
for i in range(5):
fe_share.write(str(item[0][i].cpu().numpy()))
fe_share.write(' ')
fe_share.write(str(item[1].cpu().numpy()))
fe_share.write(' ')
fe_share.write(str(item[2].cpu().numpy()))
fe_share.write('\n')
def main(args):
args.cuda = args.use_cuda and torch.cuda.is_available()
train_set, validate_set, test_set, train_loader, validate_loader, test_loader = get_data.get_data_resnet_msgpassing_balanced_lstm(args)
model_args = {'roi_feature_size':args.roi_feature_size,'edge_feature_size': args.roi_feature_size, 'node_feature_size': args.roi_feature_size,
'message_size': args.message_size, 'link_hidden_size': args.link_hidden_size,
'link_hidden_layers': args.link_hidden_layers, 'propagate_layers': args.propagate_layers,
'big_attr_classes': args.big_attr_class_num, 'lstm_hidden_size':args.lstm_hidden_size}
model = models.HGNN_resnet_msgpassing_balanced_lstm(model_args)
optimizer=torch.optim.Adam(model.parameters(),lr=args.lr)
criterion=torch.nn.CrossEntropyLoss()
if args.cuda:
model=model.cuda()
criterion=criterion.cuda( )
if args.load_best_checkpoint:
loaded_checkpoint=utils.load_best_checkpoint(args,model,optimizer,path=args.resume)
if loaded_checkpoint:
args, best_epoch_error, avg_epoch_error, model, optimizer=loaded_checkpoint
if args.load_last_checkpoint:
loaded_checkpoint=utils.load_last_checkpoint(args,model,optimizer,path=args.resume)
if loaded_checkpoint:
args, best_epoch_error, avg_epoch_error, model, optimizer=loaded_checkpoint
train_error_history=list()
train_loss_history=list()
val_error_history=list()
val_loss_history=list()
best_epoch_error=np.inf
for epoch in range(args.start_epoch, args.epochs):
train_error_rate_cur_epoch, train_loss_cur_epoch=train(train_loader,model,criterion,optimizer,epoch,args)
train_error_history.append(train_error_rate_cur_epoch)
train_loss_history.append(train_loss_cur_epoch)
val_error_rate_cur_epoch, val_loss_cur_epoch=validate(validate_loader,model,criterion,args)
val_error_history.append(val_error_rate_cur_epoch)
val_loss_history.append(val_loss_cur_epoch)
if epoch>0 and epoch%1==0:
args.lr*=args.lr_decay
for param_group in optimizer.param_groups:
param_group['lr']=args.lr
is_best=val_error_rate_cur_epoch<best_epoch_error
best_epoch_error=min(val_error_rate_cur_epoch,best_epoch_error)
avg_epoch_error=np.mean(val_error_history)
utils.save_checkpoint({
'epoch':epoch+1,
'state_dict':model.state_dict(),
'best_epoch_error':best_epoch_error,
'avg_epoch_error':avg_epoch_error,
'optimizer':optimizer.state_dict(),},is_best=is_best,directory=args.resume)
print('best_epoch_error: {}, avg_epoch_error: {}'.format(best_epoch_error, avg_epoch_error))
# test
#loaded_checkpoint=utils.load_best_checkpoint(args,model,optimizer,path=args.resume)
loaded_checkpoint = utils.load_last_checkpoint(args, model, optimizer, path=args.resume)
if loaded_checkpoint:
args, best_epoch_error, avg_epoch_error, model, optimizer=loaded_checkpoint
test(test_loader,model,args)
def main(args):
args.cuda = args.use_cuda and torch.cuda.is_available()
train_set, validate_set, test_set, train_loader, validate_loader, test_loader = get_data.get_data_resnet_fc(
args)
model = models.HGNN_resnet_fc()
# TODO: try to use the step policy for Adam, also consider the step interval
optimizer = torch.optim.Adam(model.parameters(), lr=args.lr)
# TODO: related to line 99 (e.g. max for auc, min for loss; try to check the definition of this method)
scheduler = ReduceLROnPlateau(optimizer,
factor=args.lr_decay,
patience=1,
verbose=True,
mode='min')
# TODO; double check this loss, also the output of the network
criterion = torch.nn.CrossEntropyLoss()
# criterion=torch.nn.MSELoss()
#------------------------
# use multi-gpu
if args.cuda and torch.cuda.device_count() > 1:
print("Now Using ", len(args.device_ids), " GPUs!")
#model=model.to(device_ids[0])
model = torch.nn.DataParallel(model,
device_ids=args.device_ids,
output_device=args.device_ids[0]).cuda()
criterion = criterion.cuda()
elif args.cuda:
model = model.cuda()
criterion = criterion.cuda()
if args.load_best_checkpoint:
loaded_checkpoint = utils.load_best_checkpoint(args,
model,
optimizer,
path=args.resume)
if loaded_checkpoint:
args, best_epoch_error, avg_epoch_error, model, optimizer = loaded_checkpoint
if args.load_last_checkpoint:
loaded_checkpoint = utils.load_last_checkpoint(
args,
model,
optimizer,
path=args.resume,
version=args.model_load_version)
if loaded_checkpoint:
args, best_epoch_error, avg_epoch_error, model, optimizer = loaded_checkpoint
# for param_group in optimizer.param_groups:
# param_group['lr'] = args.lr
#------------------------------------------------------------------------------
# Train
since = time.time()
train_epoch_loss_all = []
val_epoch_loss_all = []
best_loss = np.inf
avg_epoch_loss = np.inf
for epoch in range(args.start_epoch, args.epochs):
train_epoch_loss = train(train_loader, model, criterion, optimizer,
epoch, args)
train_epoch_loss_all.append(train_epoch_loss)
#visdom_viz(vis, train_epoch_loss_all, win=0, ylabel='Training Epoch Loss', title=args.project_name, color='green')
val_epoch_loss = validate(validate_loader, model, criterion, epoch,
args)
val_epoch_loss_all.append(val_epoch_loss)
#visdom_viz(vis, val_epoch_loss_all, win=1, ylabel='Validation Epoch Loss', title=args.project_name,color='blue')
print(
'Epoch {}/{} Training Loss: {:.4f} Validation Loss: {:.4f}'.format(
epoch, args.epochs - 1, train_epoch_loss, val_epoch_loss))
print('*' * 15)
#TODO: reducing lr when there is no gains on validation metric results (e.g. auc, loss)
scheduler.step(val_epoch_loss)
is_best = val_epoch_loss < best_loss
if is_best:
best_loss = val_epoch_loss
avg_epoch_loss = np.mean(val_epoch_loss_all)
utils.save_checkpoint(
{
'epoch': epoch + 1,
'state_dict': model.state_dict(),
'best_epoch_error': best_loss,
'avg_epoch_error': avg_epoch_loss,
'optimizer': optimizer.state_dict()
},
is_best=is_best,
directory=args.resume,
version='epoch_{}'.format(str(epoch)))
time_elapsed = time.time() - since
print('Training complete in {:.0f}m {:.0f}s'.format(
time_elapsed // 60, time_elapsed % 60))
print('Best Val Loss: {}, Final Avg Val Loss: {}'.format(
best_loss, avg_epoch_loss))
#-------------------------------------------------------------------------------------------------------------
# test
loaded_checkpoint = utils.load_best_checkpoint(args,
model,
optimizer,
path=args.resume)
if loaded_checkpoint:
args, best_epoch_error, avg_epoch_error, model, optimizer = loaded_checkpoint
pred_label, gt_label, test_loss = test(test_loader, model, criterion, args)
print("Test Epoch Loss {}".format(test_loss))
# save test results
if not isdir(args.save_test_res):
os.mkdir(args.save_test_res)
with open(os.path.join(args.save_test_res, 'raw_test_results.pkl'),
'w') as f:
pickle.dump([pred_label, gt_label, test_loss], f)
#todo: check get_test_metric
recall, precision, F_one_score, acc, avg_acc, ConfMat = get_test_metric(
pred_label, gt_label, args)
print(
'[====Test results Small Attr====] \n recall: {} \n precision: {} \n F1 score: {} \n acc: {} \n avg acc: {} \n Confusion Matrix: \n {}'
.format(recall, precision, F_one_score, acc, avg_acc, ConfMat))