stock.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "stock.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "mount_file_id": "14VPE6MkHzDOeWXESF2cRTPsyrmMD_jVH",
      "authorship_tag": "ABX9TyMT0fmWcuFZ6bj09qxhlPAX",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/kmtsui/hobby_stock/blob/main/stock.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "utHpUcHIwyNs"
      },
      "source": [
        "Stock trend prediction from historic data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "blV7qhJ0RMct",
        "outputId": "28a55b64-a63a-41e9-9494-0b26175893d7"
      },
      "source": [
        "from __future__ import absolute_import, division, print_function\n",
        "\n",
        "import math\n",
        "\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "from matplotlib import pyplot as plt\n",
        "\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "import tensorflow as tf\n",
        "from keras.layers import Dense, Input, Dropout\n",
        "from keras.models import Model\n",
        "from keras.callbacks import EarlyStopping\n",
        "\n",
        "# Check Versions\n",
        "print(tf.__version__)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "2.4.1\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5RGZruGbcpns",
        "outputId": "b8d74da4-d603-4d98-b88d-06c4957e4a86"
      },
      "source": [
        "# Additional tensorflow_addons is needed for the training\n",
        "!pip install tensorflow-addons"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: tensorflow-addons in /usr/local/lib/python3.7/dist-packages (0.12.1)\n",
            "Requirement already satisfied: typeguard>=2.7 in /usr/local/lib/python3.7/dist-packages (from tensorflow-addons) (2.7.1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jvFOR2c5xdPY"
      },
      "source": [
        "# Link to google drive for permanent storage\n",
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "-7ngEJ_ySHE5",
        "outputId": "220fd50d-971a-466d-8b82-e3348d547134"
      },
      "source": [
        "# Historic data you can get from https://www.macrotrends.net/stocks/research\n",
        "# Remember to remove the first ~14 rows for data input\n",
        "intel_data = pd.read_csv(\"/content/drive/MyDrive/Colab Notebooks/MacroTrends_Data_Download_INTC.csv\",names=[\"date\", \"open\", \"high\", \"low\", \"close\",\n",
        "           \"volume\"])\n",
        "intel_data.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>date</th>\n",
              "      <th>open</th>\n",
              "      <th>high</th>\n",
              "      <th>low</th>\n",
              "      <th>close</th>\n",
              "      <th>volume</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>date</td>\n",
              "      <td>open</td>\n",
              "      <td>high</td>\n",
              "      <td>low</td>\n",
              "      <td>close</td>\n",
              "      <td>volume</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>3/17/1980</td>\n",
              "      <td>0.1996</td>\n",
              "      <td>0.2028</td>\n",
              "      <td>0.1996</td>\n",
              "      <td>0.1996</td>\n",
              "      <td>10924800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3/18/1980</td>\n",
              "      <td>0.1996</td>\n",
              "      <td>0.2012</td>\n",
              "      <td>0.198</td>\n",
              "      <td>0.198</td>\n",
              "      <td>17068800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3/19/1980</td>\n",
              "      <td>0.2028</td>\n",
              "      <td>0.206</td>\n",
              "      <td>0.2028</td>\n",
              "      <td>0.2028</td>\n",
              "      <td>18508800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>3/20/1980</td>\n",
              "      <td>0.2028</td>\n",
              "      <td>0.2052</td>\n",
              "      <td>0.202</td>\n",
              "      <td>0.202</td>\n",
              "      <td>11174400</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "        date    open    high     low   close    volume\n",
              "0       date    open    high     low   close    volume\n",
              "1  3/17/1980  0.1996  0.2028  0.1996  0.1996  10924800\n",
              "2  3/18/1980  0.1996  0.2012   0.198   0.198  17068800\n",
              "3  3/19/1980  0.2028   0.206  0.2028  0.2028  18508800\n",
              "4  3/20/1980  0.2028  0.2052   0.202   0.202  11174400"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_CsFnc5hSi3k",
        "outputId": "16857782-609f-422b-d8fd-e6e5ae80b23c"
      },
      "source": [
        "intel_features = intel_data.copy()\n",
        "intel_features = np.array(intel_features)\n",
        "intel_features = intel_features[1:]\n",
        "intel_features"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([['3/17/1980', '0.1996', '0.2028', '0.1996', '0.1996', '10924800'],\n",
              "       ['3/18/1980', '0.1996', '0.2012', '0.198', '0.198', '17068800'],\n",
              "       ['3/19/1980', '0.2028', '0.206', '0.2028', '0.2028', '18508800'],\n",
              "       ...,\n",
              "       ['4/16/2021', '65.33', '65.52', '64.57', '64.75', '24625458'],\n",
              "       ['4/19/2021', '64.7', '64.74', '63.07', '63.63', '23997675'],\n",
              "       ['4/20/2021', '63.64', '63.76', '62.41', '62.7', '19872123']],\n",
              "      dtype=object)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LpJJIeHQTISn"
      },
      "source": [
        "nPoints = 200\n",
        "nPoints_after = 10\n",
        "def get_train_data(filename):\n",
        "  raw_data = pd.read_csv(filename,names=[\"date\", \"open\", \"high\", \"low\", \"close\", \"volume\"])\n",
        "  feature_array = raw_data.copy()\n",
        "  feature_array = np.array(feature_array)\n",
        "  feature_array = feature_array[1:]\n",
        "  feature_array = feature_array[len(feature_array)-4000:len(feature_array)]\n",
        "  volume = np.asfarray(feature_array[:,5],float)\n",
        "  close_value = feature_array[:,4]\n",
        "  xvalue_array = np.empty(shape=(len(feature_array)-nPoints-nPoints_after,2*nPoints))\n",
        "  yvalue_array = np.empty(len(feature_array)-nPoints-nPoints_after)\n",
        "  weigths_array = np.empty(len(feature_array)-nPoints-nPoints_after)\n",
        "  for i in range(len(feature_array)-nPoints-nPoints_after):\n",
        "    tmp_array = np.empty(nPoints)\n",
        "    volume_array = np.empty(nPoints)\n",
        "    #float(close_value[i:i+nPoints])\n",
        "    for j in range(nPoints):\n",
        "      tmp_array[j]=close_value[i+j]\n",
        "      volume_array[j]=volume[i+j]\n",
        "    tmp_array = tmp_array/np.mean(tmp_array)-1\n",
        "    volume_array = volume_array/np.mean(volume_array)-1\n",
        "    xvalue_array[i]=np.concatenate([tmp_array,volume_array])\n",
        "    yvalue_array[i] = np.heaviside(float(close_value[i+nPoints-1+nPoints_after])-float(close_value[i+nPoints-1]),0)\n",
        "    weigths_array[i] = abs(float(close_value[i+nPoints-1+nPoints_after])-float(close_value[i+nPoints-1]))/float(close_value[i+nPoints-1])\n",
        "  return xvalue_array, yvalue_array, weigths_array, np.asfarray(close_value,float)\n",
        "intel_xvalue,  intel_yvalue, intel_weights, intel_close=get_train_data(\"/content/drive/MyDrive/Colab Notebooks/MacroTrends_Data_Download_INTC.csv\")\n",
        "hsbc_xvalue,  hsbc_yvalue, hsbc_weights, hsbc_close=get_train_data(\"/content/drive/MyDrive/Colab Notebooks/MacroTrends_Data_Download_HSBC.csv\")\n",
        "jnj_xvalue,  jnj_yvalue, jnj_weights, jnj_close=get_train_data(\"/content/drive/MyDrive/Colab Notebooks/MacroTrends_Data_Download_JNJ.csv\")\n",
        "tm_xvalue,  tm_yvalue, tm_weights, tm_close=get_train_data(\"/content/drive/MyDrive/Colab Notebooks/MacroTrends_Data_Download_TM.csv\")\n",
        "dis_xvalue,  dis_yvalue, dis_weights, dis_close=get_train_data(\"/content/drive/MyDrive/Colab Notebooks/MacroTrends_Data_Download_DIS.csv\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RLpbjRpOSEzG",
        "outputId": "3e134f0b-8388-410d-be2c-bccdd34f9a16"
      },
      "source": [
        "len(dis_close)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "4000"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 85
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "E6anGdLeZpo4"
      },
      "source": [
        "# set up the DNN model for training\n",
        "inputs = Input((2*nPoints, )) # nPoints-d vector as input\n",
        "hidden_layer_1 = Dense(800, activation='relu')(inputs)\n",
        "dropout_layer_1 = Dropout(0.5)(hidden_layer_1)\n",
        "hidden_layer_2 = Dense(400, activation='relu')(dropout_layer_1)\n",
        "dropout_layer_2 = Dropout(0.5)(hidden_layer_2)\n",
        "hidden_layer_3 = Dense(200, activation='relu')(dropout_layer_2)\n",
        "dropout_layer_3 = Dropout(0.5)(hidden_layer_3)\n",
        "outputs = Dense(1, activation='sigmoid')(dropout_layer_3)\n",
        "\n",
        "model = Model(inputs=inputs, outputs=outputs)\n",
        "\n",
        "earlystopping = EarlyStopping(patience=50, \n",
        "                              verbose=1,\n",
        "                              restore_best_weights=True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5WCNoslJZ4e4",
        "outputId": "19cd330e-9964-4754-ad9e-e27f600242e4"
      },
      "source": [
        "from tensorflow.keras.layers.experimental import preprocessing\n",
        "from tensorflow_addons.optimizers import AdamW\n",
        "\n",
        "#xvalue = intel_xvalue\n",
        "#yvalue = intel_yvalue \n",
        "#weights = intel_weights\n",
        "\n",
        "xvalue = np.concatenate([intel_xvalue,hsbc_xvalue,jnj_xvalue])\n",
        "yvalue = np.concatenate([intel_yvalue,hsbc_yvalue,jnj_yvalue])\n",
        "weights = np.concatenate([intel_weights,hsbc_weights,jnj_weights])\n",
        "\n",
        "layer_reco = preprocessing.Normalization()\n",
        "layer_reco.adapt(xvalue)\n",
        "xvalue_norm = layer_reco(xvalue).numpy()\n",
        "\n",
        "#X_train_1, X_test_1, Y_train_1, Y_test_1, w_train_1, w_test_1 = train_test_split(xvalue_norm, yvalue, weights)\n",
        "X_train_1, Y_train_1, w_train_1=xvalue_norm, yvalue, weights\n",
        "X_test_1, Y_test_1, w_test_1 = layer_reco(np.concatenate([tm_xvalue,dis_xvalue])).numpy(), np.concatenate([tm_yvalue,dis_yvalue]), np.concatenate([tm_weights,dis_weights])\n",
        "\n",
        "optimizer = AdamW(weight_decay=0.0001)#,learning_rate=0.001)\n",
        "\n",
        "model.compile(loss='binary_crossentropy',\n",
        "              #optimizer='Adam',\n",
        "              optimizer= optimizer,\n",
        "              metrics=['accuracy'])\n",
        "model.fit(X_train_1,\n",
        "          Y_train_1,\n",
        "          sample_weight=w_train_1,\n",
        "          epochs=1000,\n",
        "          batch_size=64,\n",
        "          validation_data=(X_test_1, Y_test_1, w_test_1),\n",
        "          callbacks=[earlystopping],\n",
        "          verbose=1)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/1000\n",
            "178/178 [==============================] - 4s 19ms/step - loss: 0.0282 - accuracy: 0.5153 - val_loss: 0.0229 - val_accuracy: 0.5369\n",
            "Epoch 2/1000\n",
            "178/178 [==============================] - 2s 14ms/step - loss: 0.0252 - accuracy: 0.5174 - val_loss: 0.0228 - val_accuracy: 0.5435\n",
            "Epoch 3/1000\n",
            "178/178 [==============================] - 2s 14ms/step - loss: 0.0227 - accuracy: 0.5433 - val_loss: 0.0230 - val_accuracy: 0.5480\n",
            "Epoch 4/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0218 - accuracy: 0.5656 - val_loss: 0.0229 - val_accuracy: 0.5468\n",
            "Epoch 5/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0215 - accuracy: 0.5753 - val_loss: 0.0229 - val_accuracy: 0.5380\n",
            "Epoch 6/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0202 - accuracy: 0.6001 - val_loss: 0.0238 - val_accuracy: 0.5323\n",
            "Epoch 7/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0198 - accuracy: 0.5949 - val_loss: 0.0239 - val_accuracy: 0.5331\n",
            "Epoch 8/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0190 - accuracy: 0.6133 - val_loss: 0.0238 - val_accuracy: 0.5513\n",
            "Epoch 9/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0176 - accuracy: 0.6257 - val_loss: 0.0236 - val_accuracy: 0.5565\n",
            "Epoch 10/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0178 - accuracy: 0.6327 - val_loss: 0.0244 - val_accuracy: 0.5586\n",
            "Epoch 11/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0171 - accuracy: 0.6410 - val_loss: 0.0253 - val_accuracy: 0.5513\n",
            "Epoch 12/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0165 - accuracy: 0.6514 - val_loss: 0.0255 - val_accuracy: 0.5479\n",
            "Epoch 13/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0155 - accuracy: 0.6758 - val_loss: 0.0265 - val_accuracy: 0.5427\n",
            "Epoch 14/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0150 - accuracy: 0.6785 - val_loss: 0.0262 - val_accuracy: 0.5456\n",
            "Epoch 15/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0149 - accuracy: 0.6821 - val_loss: 0.0270 - val_accuracy: 0.5504\n",
            "Epoch 16/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0144 - accuracy: 0.6907 - val_loss: 0.0287 - val_accuracy: 0.5455\n",
            "Epoch 17/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0133 - accuracy: 0.7068 - val_loss: 0.0280 - val_accuracy: 0.5480\n",
            "Epoch 18/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0135 - accuracy: 0.7044 - val_loss: 0.0297 - val_accuracy: 0.5323\n",
            "Epoch 19/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0127 - accuracy: 0.7123 - val_loss: 0.0314 - val_accuracy: 0.5495\n",
            "Epoch 20/1000\n",
            "178/178 [==============================] - 3s 15ms/step - loss: 0.0121 - accuracy: 0.7250 - val_loss: 0.0309 - val_accuracy: 0.5611\n",
            "Epoch 21/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0120 - accuracy: 0.7245 - val_loss: 0.0307 - val_accuracy: 0.5476\n",
            "Epoch 22/1000\n",
            "178/178 [==============================] - 3s 15ms/step - loss: 0.0116 - accuracy: 0.7442 - val_loss: 0.0329 - val_accuracy: 0.5476\n",
            "Epoch 23/1000\n",
            "178/178 [==============================] - 3s 15ms/step - loss: 0.0104 - accuracy: 0.7520 - val_loss: 0.0327 - val_accuracy: 0.5446\n",
            "Epoch 24/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0112 - accuracy: 0.7533 - val_loss: 0.0318 - val_accuracy: 0.5530\n",
            "Epoch 25/1000\n",
            "178/178 [==============================] - 3s 15ms/step - loss: 0.0103 - accuracy: 0.7608 - val_loss: 0.0320 - val_accuracy: 0.5524\n",
            "Epoch 26/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0102 - accuracy: 0.7575 - val_loss: 0.0367 - val_accuracy: 0.5559\n",
            "Epoch 27/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0091 - accuracy: 0.7771 - val_loss: 0.0349 - val_accuracy: 0.5507\n",
            "Epoch 28/1000\n",
            "178/178 [==============================] - 3s 15ms/step - loss: 0.0094 - accuracy: 0.7715 - val_loss: 0.0362 - val_accuracy: 0.5577\n",
            "Epoch 29/1000\n",
            "178/178 [==============================] - 3s 15ms/step - loss: 0.0092 - accuracy: 0.7731 - val_loss: 0.0340 - val_accuracy: 0.5499\n",
            "Epoch 30/1000\n",
            "178/178 [==============================] - 3s 15ms/step - loss: 0.0090 - accuracy: 0.7832 - val_loss: 0.0375 - val_accuracy: 0.5475\n",
            "Epoch 31/1000\n",
            "178/178 [==============================] - 3s 15ms/step - loss: 0.0086 - accuracy: 0.7828 - val_loss: 0.0348 - val_accuracy: 0.5561\n",
            "Epoch 32/1000\n",
            "178/178 [==============================] - 3s 15ms/step - loss: 0.0084 - accuracy: 0.7997 - val_loss: 0.0336 - val_accuracy: 0.5496\n",
            "Epoch 33/1000\n",
            "178/178 [==============================] - 3s 15ms/step - loss: 0.0094 - accuracy: 0.7827 - val_loss: 0.0369 - val_accuracy: 0.5540\n",
            "Epoch 34/1000\n",
            "178/178 [==============================] - 3s 15ms/step - loss: 0.0087 - accuracy: 0.7970 - val_loss: 0.0355 - val_accuracy: 0.5631\n",
            "Epoch 35/1000\n",
            "178/178 [==============================] - 3s 15ms/step - loss: 0.0081 - accuracy: 0.7987 - val_loss: 0.0406 - val_accuracy: 0.5518\n",
            "Epoch 36/1000\n",
            "178/178 [==============================] - 3s 15ms/step - loss: 0.0076 - accuracy: 0.8131 - val_loss: 0.0402 - val_accuracy: 0.5513\n",
            "Epoch 37/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0076 - accuracy: 0.8033 - val_loss: 0.0396 - val_accuracy: 0.5525\n",
            "Epoch 38/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0075 - accuracy: 0.8187 - val_loss: 0.0389 - val_accuracy: 0.5492\n",
            "Epoch 39/1000\n",
            "178/178 [==============================] - 3s 15ms/step - loss: 0.0072 - accuracy: 0.8183 - val_loss: 0.0405 - val_accuracy: 0.5567\n",
            "Epoch 40/1000\n",
            "178/178 [==============================] - 3s 15ms/step - loss: 0.0068 - accuracy: 0.8232 - val_loss: 0.0387 - val_accuracy: 0.5604\n",
            "Epoch 41/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0072 - accuracy: 0.8141 - val_loss: 0.0401 - val_accuracy: 0.5633\n",
            "Epoch 42/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0065 - accuracy: 0.8326 - val_loss: 0.0394 - val_accuracy: 0.5521\n",
            "Epoch 43/1000\n",
            "178/178 [==============================] - 3s 15ms/step - loss: 0.0078 - accuracy: 0.8178 - val_loss: 0.0400 - val_accuracy: 0.5628\n",
            "Epoch 44/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0066 - accuracy: 0.8277 - val_loss: 0.0402 - val_accuracy: 0.5551\n",
            "Epoch 45/1000\n",
            "178/178 [==============================] - 3s 15ms/step - loss: 0.0066 - accuracy: 0.8278 - val_loss: 0.0401 - val_accuracy: 0.5542\n",
            "Epoch 46/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0063 - accuracy: 0.8325 - val_loss: 0.0410 - val_accuracy: 0.5505\n",
            "Epoch 47/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0062 - accuracy: 0.8360 - val_loss: 0.0429 - val_accuracy: 0.5503\n",
            "Epoch 48/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0061 - accuracy: 0.8357 - val_loss: 0.0394 - val_accuracy: 0.5582\n",
            "Epoch 49/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0067 - accuracy: 0.8291 - val_loss: 0.0428 - val_accuracy: 0.5529\n",
            "Epoch 50/1000\n",
            "178/178 [==============================] - 3s 14ms/step - loss: 0.0061 - accuracy: 0.8327 - val_loss: 0.0407 - val_accuracy: 0.5648\n",
            "Epoch 51/1000\n",
            "178/178 [==============================] - 3s 15ms/step - loss: 0.0061 - accuracy: 0.8449 - val_loss: 0.0381 - val_accuracy: 0.5658\n",
            "Epoch 52/1000\n",
            "178/178 [==============================] - 3s 15ms/step - loss: 0.0074 - accuracy: 0.8208 - val_loss: 0.0446 - val_accuracy: 0.5487\n",
            "Restoring model weights from the end of the best epoch.\n",
            "Epoch 00052: early stopping\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tensorflow.python.keras.callbacks.History at 0x7f05bdd56710>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 92
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4SRdXKzLQfEy"
      },
      "source": [
        "def make_predictions(events, close_value):\n",
        "  events_norm = layer_reco(events).numpy()\n",
        "  f = model.predict(events_norm, batch_size=64)\n",
        "  predictions = np.empty(len(f))\n",
        "  for i in range(len(events)):\n",
        "    predictions[i] = (f[i]-0.5)*2*abs(close_value[i+nPoints-1+nPoints_after]-close_value[i+nPoints-1])+close_value[i+nPoints-1]\n",
        "  return predictions\n",
        "intel_predictions = make_predictions(intel_xvalue,intel_close)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 392
        },
        "id": "SceRmlaGgmly",
        "outputId": "8a2fd1e9-fbd5-4059-8890-49f984221219"
      },
      "source": [
        "def revenue(events, close_value):\n",
        "  buy = 0.5\n",
        "  sell = 0.5\n",
        "  profit = 0\n",
        "  record=[]\n",
        "  win = 0\n",
        "  lose = 0 \n",
        "  events_norm = layer_reco(events).numpy()\n",
        "  f = model.predict(events_norm, batch_size=64)\n",
        "  for i in range(len(events)):\n",
        "    result = 0 \n",
        "    if f[i]>buy:\n",
        "      result = close_value[i+nPoints-1+nPoints_after]/close_value[i+nPoints-1]-1\n",
        "      record.append(result)\n",
        "    if f[i]<sell:\n",
        "      result = 1-close_value[i+nPoints-1+nPoints_after]/close_value[i+nPoints-1]\n",
        "      record.append(result)\n",
        "    if result>0:\n",
        "      win = win+1\n",
        "    elif result<0:\n",
        "      lose=lose+1\n",
        "    profit=profit+result\n",
        "  return win, lose, profit, np.array(record)\n",
        "win, lose, profit, record = revenue(dis_xvalue,dis_close)\n",
        "print('Win = {}, Lose = {}, Profit = {}'.format(win, lose, profit))\n",
        "\n",
        "bins = np.linspace(-1, 1, 41)\n",
        "\n",
        "fig, ax = plt.subplots(figsize=(6, 6))\n",
        "ax.hist(record,bins=bins)\n",
        "\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Win = 2111, Lose = 1673, Profit = 21.89843908856276\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_n2ga9XJcStt"
      },
      "source": [
        "record"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BLCZPkeLmhTd",
        "outputId": "0fcba000-9cd9-4c6d-cccd-7d92cee9421e"
      },
      "source": [
        "d"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[1]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 156
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 730
        },
        "id": "riEc8Wd7YmGT",
        "outputId": "a2735126-67d3-4842-a443-8694eb3cbeeb"
      },
      "source": [
        "# Data for plotting\n",
        "day = np.arange(0, len(intel_close), 1)\n",
        "price = intel_close\n",
        "\n",
        "day_pred = np.arange(nPoints+nPoints_after, nPoints+nPoints_after+len(intel_predictions) , 1)\n",
        "price_pred = intel_predictions\n",
        "\n",
        "fig, ax = plt.subplots(figsize=(12, 12))\n",
        "ax.plot(day, price, '-o', label='real', color='blue')\n",
        "ax.plot(day_pred, price_pred, '-o', label='predicted',color=\"red\")\n",
        "\n",
        "ax.set(xlabel='Day', ylabel='Price',\n",
        "       title='Intel price prediction')\n",
        "ax.grid()\n",
        "\n",
        "plt.xlim([len(intel_close)-365,len(intel_close)])\n",
        "\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x864 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "p9tBqpnvWEDo",
        "outputId": "aceceda8-c622-437c-f85a-2db2203748ab"
      },
      "source": [
        "len(intel_predictions)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "10152"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 110
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "O2m8EeAXWFc2",
        "outputId": "ff1cc870-69c3-4087-bf55-74fbb12f792f"
      },
      "source": [
        "len(intel_close)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "10362"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 111
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tA8lmG5HWLAI",
        "outputId": "6d53f030-7b15-4050-a6da-99101733d0b0"
      },
      "source": [
        "intel_xvalue_norm[0]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([ 0.71552014,  0.5894576 ,  1.5162296 ,  1.5691782 ,  0.10943511,\n",
              "       -0.7938307 , -0.59021366, -0.8156513 , -1.675015  , -0.5322857 ],\n",
              "      dtype=float32)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "J9ypQL_KYpBs"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}