如何将PyTorch recurrent_block转换为Keras equvalet

import torch
import torch.nn as nn
import torch.nn.functional as F
from keras.layers import LSTM, GRU
from keras.models import Sequential

class PyTorchModel(nn.Module):
    def __init__(self, input_size, hidden_size, num_layers):
        super(PyTorchModel, self).__init__()
        self.rnn = nn.RNN(input_size, hidden_size, num_layers)

    def forward(self, x):
        output, hidden = self.rnn(x)
        return output, hidden

input_size = 10
hidden_size = 20
num_layers = 2

pytorch_model = PyTorchModel(input_size, hidden_size, num_layers)

keras_model = Sequential()
keras_model.add(GRU(hidden_size, input_shape=(None, input_size), return_sequences=True))