从零开始的LSTM网络FPGA/verilog实现0--python不调库搭建LSTM

调库实现

import torch
import torch.nn as nn
from torch.nn import Parameter
from torch.nn import init
from torch import Tensor
import math

### test

def reset_weigths(model):
    """reset weights
    """
    for weight in model.parameters():
        init.constant_(weight, 0.5)

inputs = torch.ones(1, 1, 1)
h0 = torch.ones(1, 1, 1)
c0 = torch.ones(1, 1, 1)
# inputs = torch.ones(1, 1, 10)
# h0 = torch.ones(1, 1, 20)
# c0 = torch.ones(1, 1, 20)
print(h0.shape, h0)
print(c0.shape, c0)
print(inputs.shape, inputs)

# Use official lstm with input_size=10, hidden_size=20
lstm = nn.LSTM(1, 1)
# lstm = nn.LSTM(10, 20)
reset_weigths(lstm)
output2, (hn2, cn2) = lstm(inputs, (h0, c0))
print(hn2.shape, cn2.shape, output2.shape)
print(hn2)
print(cn2)
print(output2)

自己搭建

from typing import Tuple

import torch
import torch.nn as nn
from torch.nn import Parameter
from torch.nn import init
from torch import Tensor
import math


class NaiveLSTM(nn.Module):
    """Naive LSTM like nn.LSTM"""

    def __init__(self, input_size: int, hidden_size: int):
        super(NaiveLSTM, self).__init__()
        self.input_size = input_size
        self.hidden_size = hidden_size

        # input gate
        self.w_ii = Parameter(Tensor(hidden_size, input_size))
        self.w_hi = Parameter(Tensor(hidden_size, hidden_size))
        self.b_ii = Parameter(Tensor(hidden_size, 1))
        self.b_hi = Parameter(Tensor(hidden_size, 1))

        # forget gate
        self.w_if = Parameter(Tensor(hidden_size, input_size))
        self.w_hf = Parameter(Tensor(hidden_size, hidden_size))
        self.b_if = Parameter(Tensor(hidden_size, 1))
        self.b_hf = Parameter(Tensor(hidden_size, 1))

        # output gate
        self.w_io = Parameter(Tensor(hidden_size, input_size))
        self.w_ho = Parameter(Tensor(hidden_size, hidden_size))
        self.b_io = Parameter(Tensor(hidden_size, 1))
        self.b_ho = Parameter(Tensor(hidden_size, 1))

        # cell
        self.w_ig = Parameter(Tensor(hidden_size, input_size))
        self.w_hg = Parameter(Tensor(hidden_size, hidden_size))
        self.b_ig = Parameter(Tensor(hidden_size, 1))
        self.b_hg = Parameter(Tensor(hidden_size, 1))

        self.reset_weigths()

    def reset_weigths(self):
        """reset weights
        """
        stdv = 1.0 / math.sqrt(self.hidden_size)
        for weight in self.parameters():
            init.uniform_(weight, -stdv, stdv)

    def forward(self, inputs: Tensor, state: Tuple[Tensor]) \
            -> Tuple[Tensor, Tuple[Tensor, Tensor]]:
        """Forward
        Args:
            inputs: [1, 1, input_size]
            state: ([1, 1, hidden_size], [1, 1, hidden_size])
        """
        #         seq_size, batch_size, _ = inputs.size()

        if state is None:  #tuple 将列表转换为元组
            h_t = torch.zeros(1, self.hidden_size).t()   #1*hidden_size维度全0
            c_t = torch.zeros(1, self.hidden_size).t()
        else:
            (h, c) = state #(1,1,20)
            h_t = h.squeeze(0).t() #.squeeze(0) 去除0维 (1,20)       .t()转置  (20,1)
            c_t = c.squeeze(0).t()

        hidden_seq = []

        seq_size = 1
        for t in range(seq_size):
            x = inputs[:, t, :].t()   #(10,1)
            # input gate     20,1
            i = torch.sigmoid(self.w_ii @ x + self.b_ii + self.w_hi @ h_t +
                              self.b_hi)
            # forget gate    20,1
            f = torch.sigmoid(self.w_if @ x + self.b_if + self.w_hf @ h_t +
                              self.b_hf)
            # cell    20,1
            # g等效于c~
            g = torch.tanh(self.w_ig @ x + self.b_ig + self.w_hg @ h_t
                           + self.b_hg)
            # output gate    20,1
            o = torch.sigmoid(self.w_io @ x + self.b_io + self.w_ho @ h_t +
                              self.b_ho)

            c_next = f * c_t + i * g
            h_next = o * torch.tanh(c_next)
            c_next_t = c_next.t().unsqueeze(0)
            h_next_t = h_next.t().unsqueeze(0)
            hidden_seq.append(h_next_t)

        hidden_seq = torch.cat(hidden_seq, dim=0)
        return hidden_seq, (h_next_t, c_next_t)


def reset_weigths(model):
    """reset weights
    """
    for weight in model.parameters():
        init.constant_(weight, 0.5)


### test

# inputs = torch.ones(1, 1, 10)
# h0 = torch.ones(1, 1, 20)
# c0 = torch.ones(1, 1, 20)
inputs = torch.ones(1, 1, 1)
h0 = torch.ones(1, 1, 1)
c0 = torch.ones(1, 1, 1)

print(h0.shape, h0)
print(c0.shape, c0)
print(inputs.shape, inputs)

# test naive_lstm with input_size=10, hidden_size=20
# naive_lstm = NaiveLSTM(10, 20)  #input_size;hidden_size
naive_lstm = NaiveLSTM(1, 1)  #input_size;hidden_size
reset_weigths(naive_lstm)

output1, (hn1, cn1) = naive_lstm(inputs, (h0, c0))

print(hn1.shape, cn1.shape, output1.shape)
print(hn1)
print(cn1)
print(output1)

运行结果对比

两种方式输出一样