基本信息

---

下载

   友情提示：（无需注册或充值，赞助后即可获取资源下载链接）

---

     嘿，亲！知识可是无价之宝呢，但咱这精心整理的资料也耗费了不少心血呀。小小地破费一下，绝对物超所值哦！如有下载和支付问题，请联系我们QQ(微信同号)：813200300

本次赞助数额为： 20 元　

支付宝支付

微信支付

   源码介绍

---

LSTM入门学习，正弦波序列预测

        for i, input_t in enumerate(input.chunk(input.size(1), dim=1)):
            h_t, c_t = self.lstm1(input_t, (h_t, c_t))
            h_t2, c_t2 = self.lstm2(h_t, (h_t2, c_t2))
            output = self.linear(h_t2)  # output.shape:[batch,1]
            outputs = [output]  # outputs.shape:[[batch,1],...[batch,1]], list composed of n [batch,1],
        for i in range(future):  # if we should predict the future
            h_t, c_t = self.lstm1(output, (h_t, c_t))
            h_t2, c_t2 = self.lstm2(h_t, (h_t2, c_t2))
            output = self.linear(h_t2)  # output.shape:[batch,1]
            outputs = [output]  # outputs.shape:[[batch,1],...[batch,1]], list composed of n [batch,1],
        outputs = torch.stack(outputs, 1).squeeze(2)  # shape after stack:[batch, n, 1], shape after squeeze: [batch,n]
        return outputs