def main_train(params):
_set_seed(params)
results_logger = ResultsLogger(params)
if params.dataset == 'cross':
base_file = configs.data_dir['miniImagenet'] + 'all.json'
val_file = configs.data_dir['CUB'] + 'val.json'
elif params.dataset == 'cross_char':
base_file = configs.data_dir['omniglot'] + 'noLatin.json'
val_file = configs.data_dir['emnist'] + 'val.json'
else:
base_file = configs.data_dir[params.dataset] + 'base.json'
val_file = configs.data_dir[params.dataset] + 'val.json'
if 'Conv' in params.model:
if params.dataset in ['omniglot', 'cross_char']:
image_size = 28
else:
image_size = 84
else:
image_size = 224
if params.dataset in ['omniglot', 'cross_char']:
assert params.model == 'Conv4' and not params.train_aug, 'omniglot only support Conv4 without augmentation'
params.model = 'Conv4S'
optimization = 'Adam'
if params.stop_epoch == -1:
if params.method in ['baseline', 'baseline++']:
if params.dataset in ['omniglot', 'cross_char']:
params.stop_epoch = 5
elif params.dataset in ['CUB']:
params.stop_epoch = 200 # This is different as stated in the open-review paper. However, using 400 epoch in baseline actually lead to over-fitting
elif params.dataset in ['miniImagenet', 'cross']:
params.stop_epoch = 400
else:
params.stop_epoch = 400 # default
else: # meta-learning methods
if params.n_shot == 1:
params.stop_epoch = 600
elif params.n_shot == 5:
params.stop_epoch = 400
else:
params.stop_epoch = 600 # default
if params.method in ['baseline', 'baseline++']:
base_datamgr = SimpleDataManager(image_size, batch_size=16)
base_loader = base_datamgr.get_data_loader(base_file,
aug=params.train_aug)
val_datamgr = SimpleDataManager(image_size, batch_size=64)
val_loader = val_datamgr.get_data_loader(val_file, aug=False)
if params.dataset == 'omniglot':
assert params.num_classes >= 4112, 'class number need to be larger than max label id in base class'
if params.dataset == 'cross_char':
assert params.num_classes >= 1597, 'class number need to be larger than max label id in base class'
if params.method == 'baseline':
model = BaselineTrain(model_dict[params.model], params.num_classes)
elif params.method == 'baseline++':
model = BaselineTrain(model_dict[params.model],
params.num_classes,
loss_type='dist')
elif params.method in [
'DKT', 'protonet', 'matchingnet', 'relationnet',
'relationnet_softmax', 'maml', 'maml_approx'
]:
n_query = max(
1, int(16 * params.test_n_way / params.train_n_way)
) # if test_n_way is smaller than train_n_way, reduce n_query to keep batch size small
train_few_shot_params = dict(n_way=params.train_n_way,
n_support=params.n_shot)
base_datamgr = SetDataManager(image_size,
n_query=n_query,
**train_few_shot_params) # n_eposide=100
base_loader = base_datamgr.get_data_loader(base_file,
aug=params.train_aug)
test_few_shot_params = dict(n_way=params.test_n_way,
n_support=params.n_shot)
val_datamgr = SetDataManager(image_size,
n_query=n_query,
**test_few_shot_params)
val_loader = val_datamgr.get_data_loader(val_file, aug=False)
# a batch for SetDataManager: a [n_way, n_support + n_query, dim, w, h] tensor
if (params.method == 'DKT'):
model = DKT(model_dict[params.model], **train_few_shot_params)
model.init_summary()
elif params.method == 'protonet':
model = ProtoNet(model_dict[params.model], **train_few_shot_params)
elif params.method == 'matchingnet':
model = MatchingNet(model_dict[params.model],
**train_few_shot_params)
elif params.method in ['relationnet', 'relationnet_softmax']:
if params.model == 'Conv4':
feature_model = backbone.Conv4NP
elif params.model == 'Conv6':
feature_model = backbone.Conv6NP
elif params.model == 'Conv4S':
feature_model = backbone.Conv4SNP
else:
feature_model = lambda: model_dict[params.model](flatten=False)
loss_type = 'mse' if params.method == 'relationnet' else 'softmax'
model = RelationNet(feature_model,
loss_type=loss_type,
**train_few_shot_params)
elif params.method in ['maml', 'maml_approx']:
backbone.ConvBlock.maml = True
backbone.SimpleBlock.maml = True
backbone.BottleneckBlock.maml = True
backbone.ResNet.maml = True
model = MAML(model_dict[params.model],
approx=(params.method == 'maml_approx'),
**train_few_shot_params)
if params.dataset in [
'omniglot', 'cross_char'
]: # maml use different parameter in omniglot
model.n_task = 32
model.task_update_num = 1
model.train_lr = 0.1
else:
raise ValueError('Unknown method')
model = model.cuda()
params.checkpoint_dir = '%s/checkpoints/%s/%s_%s' % (
configs.save_dir, params.dataset, params.model, params.method)
if params.train_aug:
params.checkpoint_dir += '_aug'
if not params.method in ['baseline', 'baseline++']:
params.checkpoint_dir += '_%dway_%dshot' % (params.train_n_way,
params.n_shot)
if not os.path.isdir(params.checkpoint_dir):
os.makedirs(params.checkpoint_dir)
start_epoch = params.start_epoch
stop_epoch = params.stop_epoch
if params.method == 'maml' or params.method == 'maml_approx':
stop_epoch = params.stop_epoch * model.n_task # maml use multiple tasks in one update
if params.resume:
resume_file = get_resume_file(params.checkpoint_dir)
if resume_file is not None:
tmp = torch.load(resume_file)
start_epoch = tmp['epoch'] + 1
model.load_state_dict(tmp['state'])
elif params.warmup: # We also support warmup from pretrained baseline feature, but we never used in our paper
baseline_checkpoint_dir = '%s/checkpoints/%s/%s_%s' % (
configs.save_dir, params.dataset, params.model, 'baseline')
if params.train_aug:
baseline_checkpoint_dir += '_aug'
warmup_resume_file = get_resume_file(baseline_checkpoint_dir)
tmp = torch.load(warmup_resume_file)
if tmp is not None:
state = tmp['state']
state_keys = list(state.keys())
for i, key in enumerate(state_keys):
if "feature." in key:
newkey = key.replace(
"feature.", ""
) # an architecture model has attribute 'feature', load architecture feature to backbone by casting name from 'feature.trunk.xx' to 'trunk.xx'
state[newkey] = state.pop(key)
else:
state.pop(key)
model.feature.load_state_dict(state)
else:
raise ValueError('No warm_up file')
model = train(base_loader, val_loader, model, optimization, start_epoch,
stop_epoch, params, results_logger)
results_logger.save()
loss_type = 'mse' if params.method == 'relationnet' else 'softmax'
model = RelationNet(feature_model,
loss_type=loss_type,
**train_few_shot_params)
elif params.method in ['maml', 'maml_approx']:
backbone.ConvBlock.maml = True
backbone.SimpleBlock.maml = True
backbone.BottleneckBlock.maml = True
backbone.ResNet.maml = True
model = MAML(model_dict[params.model],
approx=(params.method == 'maml_approx'),
**train_few_shot_params)
if params.dataset in ['omniglot', 'cross_char'
]: #maml use different parameter in omniglot
model.n_task = 32
model.task_update_num = 1
model.train_lr = 0.1
else:
raise ValueError('Unknown method')
model = to_cuda(model)
params.checkpoint_dir = '%s/checkpoints/%s/%s_%s' % (
configs.save_dir, params.dataset, params.model, params.method)
if params.train_aug:
params.checkpoint_dir += '_aug'
if not params.method in ['baseline', 'baseline++']:
params.checkpoint_dir += '_%dway_%dshot' % (params.train_n_way,
params.n_shot)
def get_model(params, mode):
'''
Args:
params: argparse params
mode: (str), 'train', 'test'
'''
print('get_model() start...')
# few_shot_params_d = get_few_shot_params(params, None)
# few_shot_params = few_shot_params_d[mode]
few_shot_params = get_few_shot_params(params, mode)
if 'omniglot' in params.dataset or 'cross_char' in params.dataset:
# if params.dataset in ['omniglot', 'cross_char', 'cross_char_half', 'cross_char_quarter', ...]:
# assert params.model == 'Conv4' and not params.train_aug ,'omniglot only support Conv4 without augmentation'
assert 'Conv4' in params.model and not params.train_aug ,'omniglot/cross_char only support Conv4 without augmentation'
params.model = params.model.replace('Conv4', 'Conv4S') # because Conv4Drop should also be Conv4SDrop
if params.recons_decoder is not None:
if 'ConvS' not in params.recons_decoder:
raise ValueError('omniglot / cross_char should use ConvS/HiddenConvS decoder.')
# if mode == 'train':
# params.num_classes = n_base_class_map[params.dataset]
if params.method in ['baseline', 'baseline++'] and mode=='train':
assert params.num_classes >= n_base_classes[params.dataset]
# if params.dataset == 'omniglot': # 4112/688/1692
# assert params.num_classes >= 4112, 'class number need to be larger than max label id in base class'
# if params.dataset == 'cross_char': # 1597/31/31
# assert params.num_classes >= 1597, 'class number need to be larger than max label id in base class'
# if params.dataset == 'cross_char_half': # 758/31/31
# assert params.num_classes >= 758, 'class number need to be larger than max label id in base class'
# if params.dataset in ['cross_char_quarter', 'cross_char_quarter_10shot']: # 350/31/31
# assert params.num_classes >= 350, 'class number need to be larger than max label id in base class'
# if params.dataset == 'cross_char_base3lang': # 69/31/31
# assert params.num_classes >= 69, 'class number need to be larger than max label id in base class'
# if params.dataset == 'miniImagenet': # 64/16/20
# assert params.num_classes >= 64, 'class number need to be larger than max label id in base class'
# if params.dataset == 'CUB': # 100/50/50
# assert params.num_classes >= 100, 'class number need to be larger than max label id in base class'
# if params.dataset == 'cross': # 64+16+20/50/50
# assert params.num_classes >= 100, 'class number need to be larger than max label id in base class'
# if params.dataset == 'cross_base80cl': # 80/50/50
# assert params.num_classes >= 100, 'class number need to be larger than max label id in base class'
if params.recons_decoder == None:
print('params.recons_decoder == None')
recons_decoder = None
else:
recons_decoder = decoder_dict[params.recons_decoder]
print('recons_decoder:\n',recons_decoder)
backbone_func = get_backbone_func(params)
if 'baseline' in params.method:
loss_types = {
'baseline':'softmax',
'baseline++':'dist',
}
loss_type = loss_types[params.method]
if recons_decoder is None and params.min_gram is None: # default baseline/baseline++
if mode == 'train':
model = BaselineTrain(
model_func = backbone_func, loss_type = loss_type,
num_class = params.num_classes, **few_shot_params)
elif mode == 'test':
model = BaselineFinetune(
model_func = backbone_func, loss_type = loss_type,
**few_shot_params, finetune_dropout_p = params.finetune_dropout_p)
else: # other settings for baseline
if params.min_gram is not None:
min_gram_params = {
'min_gram':params.min_gram,
'lambda_gram':params.lambda_gram,
}
if mode == 'train':
model = BaselineTrainMinGram(
model_func = backbone_func, loss_type = loss_type,
num_class = params.num_classes, **few_shot_params, **min_gram_params)
elif mode == 'test':
model = BaselineFinetune(
model_func = backbone_func, loss_type = loss_type,
**few_shot_params, finetune_dropout_p = params.finetune_dropout_p)
# model = BaselineFinetuneMinGram(backbone_func, loss_type = loss_type, **few_shot_params, **min_gram_params)
elif params.method == 'protonet':
# default ProtoNet
if recons_decoder is None and params.min_gram is None:
model = ProtoNet( backbone_func, **few_shot_params )
else: # other settings
if params.min_gram is not None:
min_gram_params = {
'min_gram':params.min_gram,
'lambda_gram':params.lambda_gram,
}
model = ProtoNetMinGram(backbone_func, **few_shot_params, **min_gram_params)
if params.recons_decoder is not None:
if 'Hidden' in params.recons_decoder:
if params.recons_decoder == 'HiddenConv': # 'HiddenConv', 'HiddenConvS'
model = ProtoNetAE2(backbone_func, **few_shot_params, recons_func=recons_decoder, lambda_d=params.recons_lambda, extract_layer = 2)
elif params.recons_decoder == 'HiddenConvS': # 'HiddenConv', 'HiddenConvS'
model = ProtoNetAE2(backbone_func, **few_shot_params, recons_func=recons_decoder, lambda_d=params.recons_lambda, extract_layer = 2, is_color=False)
elif params.recons_decoder == 'HiddenRes10':
model = ProtoNetAE2(backbone_func, **few_shot_params, recons_func=recons_decoder, lambda_d=params.recons_lambda, extract_layer = 6)
elif params.recons_decoder == 'HiddenRes18':
model = ProtoNetAE2(backbone_func, **few_shot_params, recons_func=recons_decoder, lambda_d=params.recons_lambda, extract_layer = 8)
else:
if 'ConvS' in params.recons_decoder:
model = ProtoNetAE(backbone_func, **few_shot_params, recons_func=recons_decoder, lambda_d=params.recons_lambda, is_color=False)
else:
model = ProtoNetAE(backbone_func, **few_shot_params, recons_func=recons_decoder, lambda_d=params.recons_lambda, is_color=True)
elif params.method == 'matchingnet':
model = MatchingNet( backbone_func, **few_shot_params )
elif params.method in ['relationnet', 'relationnet_softmax']:
# if params.model == 'Conv4':
# feature_model = backbone.Conv4NP
# elif params.model == 'Conv6':
# feature_model = backbone.Conv6NP
# elif params.model == 'Conv4S':
# feature_model = backbone.Conv4SNP
# else:
# feature_model = lambda: model_dict[params.model]( flatten = False )
loss_type = 'mse' if params.method == 'relationnet' else 'softmax'
model = RelationNet( backbone_func, loss_type = loss_type , **few_shot_params )
elif params.method in ['maml' , 'maml_approx']:
backbone.ConvBlock.maml = True
backbone.SimpleBlock.maml = True
backbone.BottleneckBlock.maml = True
backbone.ResNet.maml = True
model = MAML( backbone_func, approx = (params.method == 'maml_approx') , **few_shot_params )
if 'omniglot' in params.dataset or 'cross_char' in params.dataset:
# if params.dataset in ['omniglot', 'cross_char', 'cross_char_half']: #maml use different parameter in omniglot
model.n_task = 32
model.task_update_num = 1
model.train_lr = 0.1
else:
raise ValueError('Unexpected params.method: %s'%(params.method))
print('get_model() finished.')
return model