Function Differences with torch.nn.GRUCell

class torch.nn.GRUCell(
    input_size,
    hidden_size,
    bias=True)(input, hidden) -> Tensor

class mindspore.nn.GRUCell(
    input_size: int,
    hidden_size: int,
    has_bias: bool=True)(x, hx) -> Tensor

# PyTorch
import torch
from torch import tensor
import numpy as np

grucell = torch.nn.GRUCell(2, 3, bias=False)
input = torch.tensor(np.array([[3.0, 4.0]]).astype(np.float32))
hidden = torch.tensor(np.array([[1.0, 2.0, 3]]).astype(np.float32))
output = grucell(input, hidden)
print(output)
# tensor([[ 0.9948,  0.0913, -0.1633]], grad_fn=<AddBackward0>)

# MindSpore
import mindspore.nn as nn
from mindspore import Tensor
import numpy as np

grucell = nn.GRUCell(2, 3, has_bias=False)
x = Tensor(np.array([[3.0, 4.0]]).astype(np.float32))
hx = Tensor(np.array([[1.0, 2.0, 3]]).astype(np.float32))
output = grucell(x, hx)
print(output)
# [[-0.94861907  0.6191679   2.1289415 ]]