stateless2stateful.ipynb

stateless2stateful.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "stateless2stateful.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/suzusuzu/14de3946d3e257d7db6e6527051850ea/stateless2stateful.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vQC1VT6LKfGc",
        "colab_type": "code",
        "outputId": "250e2687-8365-4632-c692-6f706c4b9931",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "import numpy as np\n",
        "import random\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, LSTM\n",
        "from keras.callbacks import ModelCheckpoint\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "def hankel_matrix(x,seq_len):\n",
        "    n = x.shape[0]\n",
        "    stride = x.strides[0]\n",
        "    return np.lib.stride_tricks.as_strided(x, shape=(n-seq_len+1, seq_len), strides=(stride,stride)).copy()\n",
        "\n",
        "def triangle_wave(t, m=10000):\n",
        "    arange = np.arange(1, m+1)\n",
        "    return 8/(np.pi**2)*np.sum(np.sin(arange*np.pi/2)*np.sin(arange*t)/(arange**2))\n",
        "\n",
        "seed = 0\n",
        "hidden_unit = 100\n",
        "random.seed(seed)\n",
        "np.random.seed(seed)\n",
        "\n",
        "# dataset\n",
        "N = 500\n",
        "x_n = []\n",
        "ts = np.linspace(0, 500, N)\n",
        "for t in ts:\n",
        "    tmp = triangle_wave(t)\n",
        "    x_n.append(tmp)\n",
        "x_n = np.asarray(x_n)\n",
        "seq_len = 100\n",
        "H = hankel_matrix(x_n, seq_len=seq_len)\n",
        "X = H[:-1,:].reshape(-1, seq_len, 1)\n",
        "Y = H[1:,-1]\n",
        "\n",
        "# plot data\n",
        "plt.plot(x_n[:50])\n",
        "plt.show()\n",
        "\n",
        "cp_cb = ModelCheckpoint(filepath = 'model.hdf5', monitor='val_loss', verbose=1, save_best_only=True, mode='auto')\n",
        "\n",
        "# stateless\n",
        "train_model = Sequential()\n",
        "train_model.add(LSTM(hidden_unit, input_shape=(seq_len, 1)))\n",
        "train_model.add(Dense(1))\n",
        "train_model.compile(loss='mean_squared_error', optimizer='adam')\n",
        "train_model.fit(X, Y, epochs=20, callbacks = [cp_cb], validation_split=0.3, shuffle=True)\n",
        "\n",
        "# stateful\n",
        "predict_model = Sequential()\n",
        "predict_model.add(LSTM(hidden_unit, batch_input_shape=(1, 1, 1), stateful=True))\n",
        "predict_model.add(Dense(1))\n",
        "predict_model.load_weights('model.hdf5')\n",
        "\n",
        "\n",
        "input_len = 30\n",
        "x_n = []\n",
        "xs = []\n",
        "ts = np.linspace(500, 1000, N)\n",
        "for t in ts:\n",
        "    tmp = triangle_wave(t)\n",
        "    x_n.append(tmp)\n",
        "    xs.append(tmp)\n",
        "xs = xs[:input_len]\n",
        "\n",
        "# test input\n",
        "for o in xs:\n",
        "    p = predict_model.predict(np.asarray([o]).reshape(1, 1, 1))[0,0]\n",
        "\n",
        "# predict\n",
        "predict_len = 100\n",
        "xs.append(p)\n",
        "for i in range(predict_len-1):\n",
        "    p = predict_model.predict(np.asarray([p]).reshape(1, 1, 1))[0,0]\n",
        "    xs.append(p)\n",
        "\n",
        "# plot predict\n",
        "plt.plot(xs[:input_len], label='input value')\n",
        "plt.plot(np.arange(predict_len) + input_len, x_n[input_len:][:predict_len], label='true value')\n",
        "plt.plot(np.arange(predict_len) + input_len, xs[input_len:], label='lstm predict value')\n",
        "plt.xlabel('time')\n",
        "plt.legend(loc='upper right')\n",
        "plt.show()"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Using TensorFlow backend.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "<p style=\"color: red;\">\n",
              "The default version of TensorFlow in Colab will soon switch to TensorFlow 2.x.<br>\n",
              "We recommend you <a href=\"https://www.tensorflow.org/guide/migrate\" target=\"_blank\">upgrade</a> now \n",
              "or ensure your notebook will continue to use TensorFlow 1.x via the <code>%tensorflow_version 1.x</code> magic:\n",
              "<a href=\"https://colab.research.google.com/notebooks/tensorflow_version.ipynb\" target=\"_blank\">more info</a>.</p>\n"
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAD4CAYAAADhNOGaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nWpz/H9R2pCFxcHt+AAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:66: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:541: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:4432: The name tf.random_uniform is deprecated. Please use tf.random.uniform instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/optimizers.py:793: The name tf.train.Optimizer is deprecated. Please use tf.compat.v1.train.Optimizer instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/ops/math_grad.py:1424: where (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Use tf.where in 2.0, which has the same broadcast rule as np.where\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:1033: The name tf.assign_add is deprecated. Please use tf.compat.v1.assign_add instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:1020: The name tf.assign is deprecated. Please use tf.compat.v1.assign instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3005: The name tf.Session is deprecated. Please use tf.compat.v1.Session instead.\n",
            "\n",
            "Train on 280 samples, validate on 120 samples\n",
            "Epoch 1/20\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:190: The name tf.get_default_session is deprecated. Please use tf.compat.v1.get_default_session instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:197: The name tf.ConfigProto is deprecated. Please use tf.compat.v1.ConfigProto instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:207: The name tf.global_variables is deprecated. Please use tf.compat.v1.global_variables instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:216: The name tf.is_variable_initialized is deprecated. Please use tf.compat.v1.is_variable_initialized instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:223: The name tf.variables_initializer is deprecated. Please use tf.compat.v1.variables_initializer instead.\n",
            "\n",
            "280/280 [==============================] - 4s 13ms/step - loss: 0.3242 - val_loss: 0.2994\n",
            "\n",
            "Epoch 00001: val_loss improved from inf to 0.29943, saving model to model.hdf5\n",
            "Epoch 2/20\n",
            "280/280 [==============================] - 2s 7ms/step - loss: 0.2800 - val_loss: 0.2541\n",
            "\n",
            "Epoch 00002: val_loss improved from 0.29943 to 0.25409, saving model to model.hdf5\n",
            "Epoch 3/20\n",
            "280/280 [==============================] - 2s 7ms/step - loss: 0.2283 - val_loss: 0.1895\n",
            "\n",
            "Epoch 00003: val_loss improved from 0.25409 to 0.18946, saving model to model.hdf5\n",
            "Epoch 4/20\n",
            "280/280 [==============================] - 2s 7ms/step - loss: 0.1491 - val_loss: 0.0847\n",
            "\n",
            "Epoch 00004: val_loss improved from 0.18946 to 0.08466, saving model to model.hdf5\n",
            "Epoch 5/20\n",
            "280/280 [==============================] - 2s 8ms/step - loss: 0.0338 - val_loss: 0.0235\n",
            "\n",
            "Epoch 00005: val_loss improved from 0.08466 to 0.02349, saving model to model.hdf5\n",
            "Epoch 6/20\n",
            "280/280 [==============================] - 2s 7ms/step - loss: 0.0173 - val_loss: 0.0074\n",
            "\n",
            "Epoch 00006: val_loss improved from 0.02349 to 0.00740, saving model to model.hdf5\n",
            "Epoch 7/20\n",
            "280/280 [==============================] - 2s 8ms/step - loss: 0.0102 - val_loss: 0.0093\n",
            "\n",
            "Epoch 00007: val_loss did not improve from 0.00740\n",
            "Epoch 8/20\n",
            "280/280 [==============================] - 2s 8ms/step - loss: 0.0065 - val_loss: 0.0072\n",
            "\n",
            "Epoch 00008: val_loss improved from 0.00740 to 0.00722, saving model to model.hdf5\n",
            "Epoch 9/20\n",
            "280/280 [==============================] - 2s 7ms/step - loss: 0.0067 - val_loss: 0.0059\n",
            "\n",
            "Epoch 00009: val_loss improved from 0.00722 to 0.00592, saving model to model.hdf5\n",
            "Epoch 10/20\n",
            "280/280 [==============================] - 2s 7ms/step - loss: 0.0061 - val_loss: 0.0061\n",
            "\n",
            "Epoch 00010: val_loss did not improve from 0.00592\n",
            "Epoch 11/20\n",
            "280/280 [==============================] - 2s 7ms/step - loss: 0.0058 - val_loss: 0.0060\n",
            "\n",
            "Epoch 00011: val_loss did not improve from 0.00592\n",
            "Epoch 12/20\n",
            "280/280 [==============================] - 2s 8ms/step - loss: 0.0057 - val_loss: 0.0060\n",
            "\n",
            "Epoch 00012: val_loss did not improve from 0.00592\n",
            "Epoch 13/20\n",
            "280/280 [==============================] - 2s 8ms/step - loss: 0.0058 - val_loss: 0.0059\n",
            "\n",
            "Epoch 00013: val_loss improved from 0.00592 to 0.00586, saving model to model.hdf5\n",
            "Epoch 14/20\n",
            "280/280 [==============================] - 2s 7ms/step - loss: 0.0059 - val_loss: 0.0059\n",
            "\n",
            "Epoch 00014: val_loss did not improve from 0.00586\n",
            "Epoch 15/20\n",
            "280/280 [==============================] - 2s 7ms/step - loss: 0.0057 - val_loss: 0.0059\n",
            "\n",
            "Epoch 00015: val_loss did not improve from 0.00586\n",
            "Epoch 16/20\n",
            "280/280 [==============================] - 2s 7ms/step - loss: 0.0057 - val_loss: 0.0059\n",
            "\n",
            "Epoch 00016: val_loss did not improve from 0.00586\n",
            "Epoch 17/20\n",
            "280/280 [==============================] - 2s 7ms/step - loss: 0.0056 - val_loss: 0.0058\n",
            "\n",
            "Epoch 00017: val_loss improved from 0.00586 to 0.00580, saving model to model.hdf5\n",
            "Epoch 18/20\n",
            "280/280 [==============================] - 2s 7ms/step - loss: 0.0057 - val_loss: 0.0058\n",
            "\n",
            "Epoch 00018: val_loss did not improve from 0.00580\n",
            "Epoch 19/20\n",
            "280/280 [==============================] - 2s 7ms/step - loss: 0.0057 - val_loss: 0.0059\n",
            "\n",
            "Epoch 00019: val_loss did not improve from 0.00580\n",
            "Epoch 20/20\n",
            "280/280 [==============================] - 2s 7ms/step - loss: 0.0057 - val_loss: 0.0058\n",
            "\n",
            "Epoch 00020: val_loss did not improve from 0.00580\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEGCAYAAABo25JHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nPNpE0EYbbbRxyqNNBG200UYbpzz+f27V1zBA21EpAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    }
  ]
}