code
stringlengths
3
6.57k
is_apex_available()
the (Sync)
regularized (weight_decay=0)
regularized_params.append(module.weight)
unregularized_params.append(module.weight)
regularized_params.append(module.bias)
unregularized_params.append(module.bias)
_filter_trainable(param_list: List[Any])
list(filter(lambda x: x.requires_grad, param_list)
unregularized_param_list (pre-existing list of parameters not to regularize)
enumerate(parameters_to_unregularize)
enumerate(regularized_param_list)
indices_to_remove_from_regularized.append(reg_param_ind)
indices_to_remove_from_new_unregularized.append(unreg_param_ind)
indices_to_remove_from_regularized.sort(reverse=True)
indices_to_remove_from_new_unregularized.sort(reverse=True)
param_groups (List[Dict])
model.parameters()
list(model.parameters()
model.named_modules()
isinstance(module, nn.Linear)
isinstance(module, _CONV_TYPES)
hasattr(module, "weight_g")
head_regularized_params.append(module.weight_g)
head_regularized_params.append(module.weight_v)
head_regularized_params.append(module.weight)
head_regularized_params.append(module.bias)
head_unregularized_params.append(module.bias)
isinstance(module, _NORM_TYPES)
isinstance(module, nn.Linear)
isinstance(module, _CONV_TYPES)
hasattr(module, "weight_g")
trunk_regularized_params.append(module.weight_g)
trunk_regularized_params.append(module.weight_v)
trunk_regularized_params.append(module.weight)
trunk_regularized_params.append(module.bias)
trunk_unregularized_params.append(module.bias)
isinstance(module, _NORM_TYPES)
len(list(module.children()
them (set recurse=False)
module.parameters(recurse=False)
regularized_params.append(params)
model.named_parameters()
any(hits)
unregularized_params.append(param)
model.named_parameters()
non_trainable_params.append(param)
_filter_trainable(model.parameters()
_filter_trainable(trunk_regularized_params)
_filter_trainable(trunk_unregularized_params)
_filter_trainable(head_regularized_params)
_filter_trainable(head_unregularized_params)
_filter_trainable(regularized_params)
len(trainable_params)
len(non_trainable_params)
len(trunk_regularized_params)
len(trunk_unregularized_params)
len(head_regularized_params)
len(head_unregularized_params)
len(regularized_params)
len(unregularized_params)
len(regularized_params)
len(unregularized_params)
Copyright (C)
upgrade()
sa.Date()
db.session.query(models.Workflow)
date.today()
all()
if ((not all(tasks_start_days)
all(tasks_end_days)
or (not tasks_start_days and not tasks_end_days)
db.session.add(workflow)
db.session.query(models.Workflow)
date.today()
all()
if ((not all(tasks_start_days)
all(tasks_end_days)
or (not tasks_start_days and not tasks_end_days)
db.session.add(workflow)
max([c.start_date for c in workflow.cycles])
base_date.today()
max(base_date, workflow.next_cycle_start_date)
get_cycle_calculator(workflow, base_date=base_date)
calculator.reified_tasks.values()
calculator.reified_tasks.values()
calculator.adjust_date(nancsd_date)
calculator.adjust_date(nancsd_date)
calculator.adjust_date(nancsd_date)
db.session.add(workflow)
db.session.commit()
downgrade()
op.drop_column('workflows', 'non_adjusted_next_cycle_start_date')
Copyright (c)
diele_func_to_coeff(freq, real, imag)
return (2 * sqrt(2)
sqrt(sqrt(real ** 2 + imag ** 2)
DieleFuncData(MSONable, ToJsonFileMixIn)
ave_absorption_coeff(self)
sum(self.diele_func_real[i][:3])