epoch_cl_val_acc = []
max_val_acc = 0
optimizer = optim.Adam(model.parameters(), lr=learning_rate_2)
for epoch in range(n_epochs_2):
# TRAIN
model.train()
correct = 0
train2_loss = 0
train_num = 0
# freeze params except for the classifier
trained_names = ['classifier.0.bias', 'classifier.0.weight']
for name, param in model.named_parameters():
if name in trained_names:
param.requires_grad = True
else:
param.requires_grad = False
for i, (XI, XB, y) in enumerate(train_loader):
if model.header == 'CNN':
x = XI
else:
x = XB
x, y = x.to(device), y.long().to(device)
if x.size()[0] != batch_size:
# print("batch {} size {} < {}, skip".format(i, x.size()[0], batch_size))
break
train_num += x.size(0)
for epoch in range(n_epochs_2):
# TRAIN
model_B.train()
model_I.train()
correct_B = 0
train2_loss_B = 0
correct_I = 0
train2_loss_I = 0
train2_loss_tot = 0
train_num = 0
# freeze params except for the classifier
trained_names = ['classifier.0.bias', 'classifier.0.weight']
for name, param in model_B.named_parameters():
if name in trained_names:
param.requires_grad = True
else:
param.requires_grad = False
for name, param in model_I.named_parameters():
if name in trained_names:
param.requires_grad = True
else:
param.requires_grad = False
for i, (XI, XB, y) in enumerate(train_loader):
XI, XB, y = XI.to(device), XB.to(device), y.long().to(device)
if XI.size()[0] != batch_size:
# print("batch {} size {} < {}, skip".format(i, x.size()[0], batch_size))