{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 04 - Time Series\n",
    "\n",
    "This module was built by Ignacio Oguiza among others, see [this](https://forums.fast.ai/t/timeseries/55861/1) megathread for the discussion. The essential goal here is we can use arrays of any dimension instead of just 1 (such as tabular) or 3 (such as images). \n",
    "\n",
    "* Note, this notebook was heavily influenced from his tutorial notebook [here](https://github.com/timeseriesAI/timeseriesAI2/blob/master/tutorial_nbs/01_How_to_use_numpy_arrays_in_fastai.ipynb)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pip install fastai"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We'll need a few more libraries and the `timeseriesAI2` github:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pip install pyunpack sktime -q\n",
    "!git clone 'https://github.com/timeseriesAI/timeseriesAI2.git'\n",
    "%cd timeseriesAI2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's grab what we need:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from fastai.tabular.all import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from timeseries.imports import *\n",
    "from timeseries.utils import *\n",
    "from timeseries.data import *\n",
    "from timeseries.core import *\n",
    "from timeseries.models import *"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For our data we'll be utilizing the UCR repository which has 128 univariate and 30 multivariate datasets. In the framework we can quickly grab any dataset we want:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "name = 'StarLightCurves'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can now grab our train and validation by calling `get_UCR_data`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset: StarLightCurves\n",
      "Downloading and decompressing data to data/UCR/StarLightCurves...\n",
      "...data downloaded and decompressed\n",
      "X_train: (1000, 1, 1024)\n",
      "y_train: (1000,)\n",
      "X_valid: (8236, 1, 1024)\n",
      "y_valid: (8236,) \n",
      "\n"
     ]
    }
   ],
   "source": [
    "X_train, y_train, X_valid, y_valid = get_UCR_data(name, verbose=True, on_disk=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since data is already split between train and test in the UCR dataset, we are going to merge them and create some indices to split them in the same sets. To save on memory, he figured out a way to utilized the numpy arrays via your disk. We'll do so as follows:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = np.concatenate((X_train, X_valid))\n",
    "y = np.concatenate((y_train, y_valid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.save('./data/UCR/StarLightCurves/X.npy', X)\n",
    "np.save('./data/UCR/StarLightCurves/y.npy', y)\n",
    "del X, y"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we can load them back in and make our splits:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = np.load('./data/UCR/StarLightCurves/X.npy', mmap_mode='r')\n",
    "y = np.load('./data/UCR/StarLightCurves/y.npy', mmap_mode='r')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "splits = (L(np.arange(len(X_train)), use_list=True),\n",
    "          L(np.arange(len(X_train), len(X)), use_list=True))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((#1000) [0,1,2,3,4,5,6,7,8,9...],\n",
       " (#8236) [1000,1001,1002,1003,1004,1005,1006,1007,1008,1009...])"
      ]
     },
     "execution_count": null,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "splits"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since we used memmap the data is being directly read from memory. Now to make and use your own data they need to be in a three dimentional array with a format of:\n",
    "\n",
    "* Samples\n",
    "* Variables\n",
    "* Length (or timesteps)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To use this we have a special `TSTensor` built to handle such data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "t = TSTensor(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "TSTensor(samples:9236, vars:1, len:1024)"
      ]
     },
     "execution_count": null,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So here we can see we had 60 samples with one variable and an overall time length of 570 for each"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can directly use this in our `DataBlock` API too! With a `TSTensorBlock`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "splitter = IndexSplitter(splits[1])\n",
    "getters = [ItemGetter(0), ItemGetter(1)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "dblock = DataBlock(blocks=(TSTensorBlock, CategoryBlock),\n",
    "                   getters=getters,\n",
    "                   splitter=splitter)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next we make a call to `itemify` which zips up our x's and our y's"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "src = itemify(X, y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And now we can make our `DataLoaders`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "dls = dblock.dataloaders(src, bs=64, val_bs=128)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAacAAAEnCAYAAAANX4xAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUREAo7CSUREAs7/D8R029u1/hgrAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAacAAAEoCAYAAAD8CT6VAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nbK+WN8UrAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAacAAAEoCAYAAAD8CT6VAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "dls.show_batch(max_n=3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And there's our data! Now are we still on disk? *Yes*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(memmap([[0.5373029 , 0.53110296, 0.52850294, ..., 0.52640295, 0.51950294,\n",
       "          0.51140296]], dtype=float32), TensorCategory(2))"
      ]
     },
     "execution_count": null,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dls.dataset[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training our Model\n",
    "\n",
    "THe particular model we're using is the [Inception Time](https://towardsdatascience.com/deep-learning-for-time-series-classification-inceptiontime-245703f422db) model. To do so we need the number of input classes and our number of variables:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3"
      ]
     },
     "execution_count": null,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dls.c"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "inp_vars = dls.dataset[0][0].shape[-2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": null,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inp_vars"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "net = InceptionTime(inp_vars, dls.c)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "learn = Learner(dls, net, loss_func=CrossEntropyLossFlat(), metrics=accuracy, opt_func=ranger)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       ""
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "SuggestedLRs(lr_min=0.025118863582611083, lr_steep=0.002511886414140463)"
      ]
     },
     "execution_count": null,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEKCAYAAAAIO8L1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "learn.lr_find()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And now we can fit!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.539881</td>\n",
       "      <td>1.406547</td>\n",
       "      <td>0.641938</td>\n",
       "      <td>00:07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.433147</td>\n",
       "      <td>0.335748</td>\n",
       "      <td>0.851506</td>\n",
       "      <td>00:07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.373533</td>\n",
       "      <td>2.214681</td>\n",
       "      <td>0.559009</td>\n",
       "      <td>00:07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.325850</td>\n",
       "      <td>0.599999</td>\n",
       "      <td>0.852477</td>\n",
       "      <td>00:07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.295900</td>\n",
       "      <td>0.382158</td>\n",
       "      <td>0.837785</td>\n",
       "      <td>00:07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.270379</td>\n",
       "      <td>0.178628</td>\n",
       "      <td>0.942205</td>\n",
       "      <td>00:07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.252287</td>\n",
       "      <td>1.122112</td>\n",
       "      <td>0.765542</td>\n",
       "      <td>00:07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.230175</td>\n",
       "      <td>0.151228</td>\n",
       "      <td>0.949247</td>\n",
       "      <td>00:07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.200199</td>\n",
       "      <td>0.110600</td>\n",
       "      <td>0.972195</td>\n",
       "      <td>00:07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.172292</td>\n",
       "      <td>0.106513</td>\n",
       "      <td>0.973045</td>\n",
       "      <td>00:07</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "learn.fit_flat_cos(10, 0.025)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "name": "python3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}