def test(epoch):
global best_acc
net.eval()
test_loss = 0
correct = 0
total = 0
for batch_idx, (inputs, targets) in enumerate(testloader):
if use_cuda:
inputs, targets = inputs.cuda(), targets.cuda()
inputs, targets = Variable(inputs, volatile=True), Variable(targets)
outputs = net(inputs)
loss = criterion(outputs, targets)
test_loss += loss.data[0]
_, predicted = torch.max(outputs.data, 1)
total += targets.size(0)
correct += predicted.eq(targets.data).cpu().sum()
progress_bar(
batch_idx, len(testloader), 'Loss: %.3f | Acc: %.3f%% (%d/%d)' %
(test_loss /
(batch_idx + 1), 100. * correct / total, correct, total))
test_accuracy_list[epoch] = 100. * correct / total
acc = 100. * correct / total
if acc > best_acc:
dump_acc_record(acc, net, use_cuda, epoch, args)
dump_record(train_accuracy_list, test_accuracy_list,
learning_rate_list, loss_function_list, args)
best_acc = acc
default=0,
type=float,
help='incorrect_penalty')
parser.add_argument('--normalize',
action='store_true',
help='normalize the rewards')
parser.add_argument('--epochs_to_train', default=300, type=int)
args = parser.parse_args()
use_cuda = torch.cuda.is_available()
trainloader, testloader, _ = transform_dataset(batch_size=400)
# Model
storedNet, trainList = _initilization_(args, use_cuda)
(net, best_acc, start_epoch), (train_accuracy_list, test_accuracy_list,
learning_rate_list,
loss_function_list) = storedNet, trainList
criterion = parser_loss_function(args=args)
optimizer = optim.SGD(net.parameters(),
lr=args.lr,
momentum=0.9,
weight_decay=5e-4)
# Training
for epoch in range(start_epoch, start_epoch + args.epochs_to_train):
train(epoch)
test(epoch)
dump_record(train_accuracy_list, test_accuracy_list,
learning_rate_list, loss_function_list, args)
u16 key_sz;
char key[];
};
'''
filename = sys.argv[1]
fd = open(filename, 'rb')
data = fd.read()
p = 0
p += utils.dump_record(
'index header',
[
'magic', 'last_record_logno', 'last_record_offset', 'keys_stored',
'checksum'
],
"<IiQQI",
data,
p,
magic=0x43211234,
)
i = itertools.count(1)
while p < len(data):
p += utils.dump_record(
'item %i' % i.next(),
[
'magic', 'checksum', 'logno', 'value_offset', 'key_sz', 'r',
'value_sz'
],
"<IIiQHHI",
struct ydb_value_record{
u32 checksum;
};
'''
filename = sys.argv[1]
fd = open(filename, 'rb')
data = fd.read()
p = 0
i = itertools.count(1)
while p < len(data):
p += utils.dump_record('record',
['magic', 'checksum', 'flags', 'key_sz', 'value_sz', 'key'],
"<IIHHI", data, p,
magic = 0x7DB5EC5D,
)
p += utils.dump_record('item %i' % i.next(),
['magic', 'checksum', 'logno', 'value_offset', 'key_sz', 'r','value_sz'],
"<IIiQHHI", data, p,
indent=4,
magic = 0x12344321,
)
u64 value_offset;
u32 value_sz;
u16 key_sz;
char key[];
};
'''
filename = sys.argv[1]
fd = open(filename, 'rb')
data = fd.read()
p = 0
p += utils.dump_record('index header',
['magic', 'last_record_logno', 'last_record_offset', 'keys_stored', 'checksum'],
"<IiQQI", data, p,
magic = 0x43211234,
)
i = itertools.count(1)
while p < len(data):
p += utils.dump_record('item %i' % i.next(),
['magic', 'checksum', 'logno', 'value_offset', 'key_sz', 'r','value_sz'],
"<IIiQHHI", data, p,
indent=4,
key_len = 'key_sz',
magic = 0x12344321,
)
