def init_model_with_schemas(
model_name,
sig_input_dim, tanh_input_dim,
pred_dim
):
workspace.ResetWorkspace()
input_record_schema = schema.Struct(
('sig_input', schema.Scalar((np.float32, (sig_input_dim, )))),
('tanh_input', schema.Scalar((np.float32, (tanh_input_dim, ))))
)
output_record_schema = schema.Struct(
('loss', schema.Scalar((np.float32, (1, )))),
('pred', schema.Scalar((np.float32, (pred_dim, ))))
)
# use trainer_extra_schema as the loss input record
trainer_extra_schema = schema.Struct(
('label', schema.Scalar((np.float32, (pred_dim, )))),
('prediction', schema.Scalar((np.float32, (pred_dim, ))))
)
model = layer_model_helper.LayerModelHelper(
model_name,
input_record_schema,
trainer_extra_schema
)
model.output_schema = output_record_schema
return model
def reset_model(self, input_feature_schema=None, trainer_extra_schema=None):
input_feature_schema = input_feature_schema or schema.Struct(
('float_features', schema.Scalar((np.float32, (32,)))),
)
trainer_extra_schema = trainer_extra_schema or schema.Struct()
self.model = layer_model_helper.LayerModelHelper(
'test_model',
input_feature_schema=input_feature_schema,
trainer_extra_schema=trainer_extra_schema)
def setUp(self):
super(TestLayers, self).setUp()
input_feature_schema = schema.Struct(
('float_features', schema.Scalar((np.float32, (32, )))), )
trainer_extra_schema = schema.Struct()
self.model = layer_model_helper.LayerModelHelper(
'test_model',
input_feature_schema=input_feature_schema,
trainer_extra_schema=trainer_extra_schema)
def init_model_with_schemas(
model_name,
input_dim,
output_dim,
):
workspace.ResetWorkspace()
input_record_schema = schema.Struct(
('origin_input', schema.Scalar((np.float32, (input_dim, )))),
('adjoint_input', schema.Scalar((np.float32, (output_dim, )))))
output_record_schema = schema.Struct(
('loss', schema.Scalar((np.float32, (input_dim, )))),
('origin_pred', schema.Scalar((np.float32, (output_dim, )))),
('adjoint_pred', schema.Scalar((np.float32, (input_dim, )))))
# use trainer_extra_schema as the loss input record
pred_dim = input_dim # for adjoint nn, the pred_dim is the same as original input
trainer_extra_schema = schema.Struct(
('label', schema.Scalar((np.float32, (pred_dim, )))),
('prediction', schema.Scalar((np.float32, (pred_dim, )))))
model = layer_model_helper.LayerModelHelper(model_name,
input_record_schema,
trainer_extra_schema)
model.output_schema = output_record_schema
return model
def init_model_with_schemas(model_name,
sig_input_dim,
tanh_input_dim,
pred_dim,
train_target=TrainTarget.ADJOINT):
'''
output_records have to filled with existing blobs.
'''
workspace.ResetWorkspace()
print('>>> Training Target: ' + train_target)
if train_target == TrainTarget.ADJOINT:
# When training the adjoint network, we also need to forward pass
# through the origin network
input_record_schema = schema.Struct(
('sig_input', schema.Scalar(
(np.float32, (sig_input_dim, )))), # sig
('tanh_input', schema.Scalar(
(np.float32, (tanh_input_dim, )))), # tanh
('adjoint_input', schema.Scalar((np.float32, (pred_dim, )))))
output_record_schema = schema.Struct(
('loss', schema.Scalar((np.float32, (1, )))),
('origin_pred', schema.Scalar((np.float32, (pred_dim, )))),
('sig_adjoint_pred', schema.Scalar(
(np.float32, (sig_input_dim, )))),
('tanh_adjoint_pred',
schema.Scalar((np.float32, (tanh_input_dim, )))),
)
# use trainer_extra_schema as the loss input record
trainer_extra_schema = schema.Struct(
('sig_loss_record',
schema.Struct(
('label', schema.Scalar((np.float32, (sig_input_dim, )))),
('prediction', schema.Scalar(
(np.float32, (sig_input_dim, )))))),
('tanh_loss_record',
schema.Struct(
('label', schema.Scalar((np.float32, (tanh_input_dim, )))),
('prediction', schema.Scalar(
(np.float32, (tanh_input_dim, )))))),
('origin_loss_record',
schema.Struct(
('label', schema.Scalar((np.float32, (pred_dim, )))),
('prediction', schema.Scalar((np.float32, (pred_dim, )))))),
)
if train_target == TrainTarget.ORIGIN:
# When training the origin network, no need of the adjoint network
input_record_schema = schema.Struct(
('sig_input', schema.Scalar(
(np.float32, (sig_input_dim, )))), # sig
('tanh_input', schema.Scalar(
(np.float32, (tanh_input_dim, )))), # tanh
('adjoint_input', schema.Scalar((np.float32, (pred_dim, )))))
output_record_schema = schema.Struct(
('loss', schema.Scalar((np.float32, (1, )))),
('origin_pred', schema.Scalar((np.float32, (pred_dim, )))),
)
# use trainer_extra_schema as the loss input record
trainer_extra_schema = schema.Struct(
('origin_loss_record',
schema.Struct(
('label', schema.Scalar((np.float32, (pred_dim, )))),
('prediction', schema.Scalar((np.float32, (pred_dim, )))))), )
model = layer_model_helper.LayerModelHelper(model_name,
input_record_schema,
trainer_extra_schema)
model.output_schema = output_record_schema
return model
