fix merge

python/oneflow/test/modules/test_optim_lbfgs.py

@@ -156,11 +156,7 @@ def compare_with_numpy_lbfgs(
  save_load_by_pickle,
  contiguous_params,
  tensor_num,
-<<<<<<< HEAD
- use_float64
-=======
  use_float64,
->>>>>>> 1d863e972249cb730e54523b9ae9a794e20fc430
 ):
  random_grad_seq = []
  init_value_seq = []
@@ -174,32 +170,18 @@ def compare_with_numpy_lbfgs(
  flow.set_default_tensor_type(flow.FloatTensor)
  for _ in range(tensor_num):
  init_value_seq.append(np.random.uniform(size=x_shape).astype(npType))
-<<<<<<< HEAD
- for _ in range(train_iters):
- random_grad_seq_per_iter = []
- for _ in range(tensor_num):
- random_grad_seq_per_iter.append(
- np.random.uniform(size=x_shape).astype(npType)
- )
- random_grad_seq.append(random_grad_seq_per_iter)
-=======
  for _ in range(tensor_num):
  random_grad_seq.append(np.random.uniform(size=x_shape).astype(npType))
->>>>>>> 1d863e972249cb730e54523b9ae9a794e20fc430
 
  def train_by_oneflow():
  x = []
  for i in range(tensor_num):
  x.append(
-<<<<<<< HEAD
- Parameter(flow.Tensor(init_value_seq[i], device=flow.device(device)))
-=======
  Parameter(
  flow.tensor(
  init_value_seq[i], device=flow.device(device), dtype=flowType
  )
  )
->>>>>>> 1d863e972249cb730e54523b9ae9a794e20fc430
  )
 
  lbfgs = flow.optim.LBFGS(
@@ -311,10 +293,6 @@ def np_train_one_iter(x, state, init_grad):
 
  prev_flat_grad = np.copy(flat_grad)
  prev_loss = loss
-<<<<<<< HEAD
-
-=======
->>>>>>> 1d863e972249cb730e54523b9ae9a794e20fc430
  if state["n_iter"] == 1:
  t = min(1.0, 1.0 / np.sum(np.abs(flat_grad))) * learning_rate
  else:
@@ -393,35 +371,20 @@ class TestLBFGS(flow.unittest.TestCase):
  def test_lbfgs_numpy(test_case):
  arg_dict = OrderedDict()
  arg_dict["device"] = [random_device().value()]
-<<<<<<< HEAD
- arg_dict["x_shape"] = [10,20]
- arg_dict["learning_rate"] = [0.01]
- arg_dict["train_iters"] = [10]
-=======
  arg_dict["x_shape"] = [10, 20]
  arg_dict["learning_rate"] = [0.01]
  arg_dict["train_iters"] = [20]
->>>>>>> 1d863e972249cb730e54523b9ae9a794e20fc430
  arg_dict["max_iter"] = [20]
  arg_dict["max_eval"] = [25]
  arg_dict["tolerance_grad"] = [1e-7]
  arg_dict["tolerance_change"] = [1e-9]
  arg_dict["history_size"] = [100]
-<<<<<<< HEAD
- arg_dict["line_search_fn"] = [None,"strong_wolfe"]
- arg_dict["reload_state_step"] = [5]
- arg_dict["save_load_by_pickle"] = [random_bool().value()]
- arg_dict["contiguous_params"] = [random_bool().value()]
- arg_dict["tensor_num"] = [3, 4,7]
- arg_dict["use_float64"] = [True]
-=======
  arg_dict["line_search_fn"] = [None, "strong_wolfe"]
  arg_dict["reload_state_step"] = [5]
  arg_dict["save_load_by_pickle"] = [random_bool().value()]
  arg_dict["contiguous_params"] = [random_bool().value()]
  arg_dict["tensor_num"] = [3, 4, 7]
  arg_dict["use_float64"] = [True, False]
->>>>>>> 1d863e972249cb730e54523b9ae9a794e20fc430
  for arg in GenArgList(arg_dict):
  compare_with_numpy_lbfgs(test_case, *arg)