{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Chapter 13 - Convolutional Neural Networks \n", "\n", "> This is almost same with the original notebook.I took some notes but nothing added. It must be read from start to end.\n", "\n", "- toc: true \n", "- badges: true\n", "- comments: true\n", "- categories: [fastbook]\n", "- image: images/fastbook_images/chapter-13/no_addition.png" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "![](images/chapter-13/no_addition.png)\n" ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [], "source": [ "\n", "#!pip install -Uqq fastbook\n", "import fastbook\n", "fastbook.setup_book()" ] }, { "cell_type": "code", "execution_count": 26, "metadata": {}, "outputs": [], "source": [ "\n", "from fastai.vision.all import *\n", "from fastbook import *\n", "\n", "matplotlib.rc('image', cmap='Greys')" ] }, { "cell_type": "raw", "metadata": {}, "source": [ "[[chapter_convolutions]]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Convolutional Neural Networks" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "In <> we learned how to create a neural network recognizing images. We were able to achieve a bit over 98% accuracy at distinguishing 3s from 7s—but we also saw that fastai's built-in classes were able to get close to 100%. Let's start trying to close the gap.\n", "\n", "In this chapter, we will begin by digging into what convolutions are and building a CNN from scratch. We will then study a range of techniques to improve training stability and learn all the tweaks the library usually applies for us to get great results." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## The Magic of Convolutions" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "One of the most powerful tools that machine learning practitioners have at their disposal is *feature engineering*. A *feature* is a transformation of the data which is designed to make it easier to model. For instance, the `add_datepart` function that we used for our tabular dataset preprocessing in <> added date features to the Bulldozers dataset. What kinds of features might we be able to create from images?" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "> jargon: Feature engineering: Creating new transformations of the input data in order to make it easier to model." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "In the context of an image, a feature is a visually distinctive attribute. For example, the number 7 is characterized by a horizontal edge near the top of the digit, and a top-right to bottom-left diagonal edge underneath that. On the other hand, the number 3 is characterized by a diagonal edge in one direction at the top left and bottom right of the digit, the opposite diagonal at the bottom left and top right, horizontal edges at the middle, top, and bottom, and so forth. So what if we could extract information about where the edges occur in each image, and then use that information as our features, instead of raw pixels?\n", "\n", "It turns out that finding the edges in an image is a very common task in computer vision, and is surprisingly straightforward. To do it, we use something called a *convolution*. A convolution requires nothing more than multiplication, and addition—two operations that are responsible for the vast majority of work that we will see in every single deep learning model in this book!\n", "\n", "A convolution applies a *kernel* across an image. A kernel is a little matrix, such as the 3×3 matrix in the top right of <>." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"Applying" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The 7×7 grid to the left is the *image* we're going to apply the kernel to. The convolution operation multiplies each element of the kernel by each element of a 3×3 block of the image. The results of these multiplications are then added together. The diagram in <> shows an example of applying a kernel to a single location in the image, the 3×3 block around cell 18.\n", "\n", "Let's do this with code. First, we create a little 3×3 matrix like so:" ] }, { "cell_type": "code", "execution_count": 27, "metadata": {}, "outputs": [], "source": [ "top_edge = tensor([[-1,-1,-1],\n", " [ 0, 0, 0],\n", " [ 1, 1, 1]]).float()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We're going to call this our kernel (because that's what fancy computer vision researchers call these). And we'll need an image, of course:" ] }, { "cell_type": "code", "execution_count": 28, "metadata": {}, "outputs": [], "source": [ "path = untar_data(URLs.MNIST_SAMPLE)" ] }, { "cell_type": "code", "execution_count": 29, "metadata": {}, "outputs": [], "source": [ "Path.BASE_PATH = path" ] }, { "cell_type": "code", "execution_count": 30, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "im3 = Image.open(path/'train'/'3'/'12.png')\n", "show_image(im3);" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now we're going to take the top 3×3-pixel square of our image, and multiply each of those values by each item in our kernel. Then we'll add them up, like so:" ] }, { "cell_type": "code", "execution_count": 31, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "tensor([[-0., -0., -0.],\n", " [0., 0., 0.],\n", " [0., 0., 0.]])" ] }, "execution_count": 31, "metadata": {}, "output_type": "execute_result" } ], "source": [ "im3_t = tensor(im3)\n", "im3_t[0:3,0:3] * top_edge" ] }, { "cell_type": "code", "execution_count": 32, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "tensor(0.)" ] }, "execution_count": 32, "metadata": {}, "output_type": "execute_result" } ], "source": [ "(im3_t[0:3,0:3] * top_edge).sum()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Not very interesting so far—all the pixels in the top-left corner are white. But let's pick a couple of more interesting spots:" ] }, { "cell_type": "code", "execution_count": 56, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
000000000000000000000
100000000000000000000
200000000000000000000
300000000000000000000
400000000000000000000
5000129991142155246182155155155155131520000
6000138254254254254254254254254254254254252210122330
7000220254254254235189189189189150189205254254254750
80003574353525000000132242542541530
90000000000000090254254247530
" ], "text/plain": [ "" ] }, "execution_count": 56, "metadata": {}, "output_type": "execute_result" } ], "source": [ "#hide_output\n", "df = pd.DataFrame(im3_t[:10,:20])\n", "df.style.set_properties(**{'font-size':'6pt'}).background_gradient('Greys')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"Top" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "There's a top edge at cell 5,8. Let's repeat our calculation there:" ] }, { "cell_type": "code", "execution_count": 34, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "tensor(762.)" ] }, "execution_count": 34, "metadata": {}, "output_type": "execute_result" } ], "source": [ "(im3_t[4:7,6:9] * top_edge).sum()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "There's a right edge at cell 8,18. What does that give us?:" ] }, { "cell_type": "code", "execution_count": 35, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "tensor(-29.)" ] }, "execution_count": 35, "metadata": {}, "output_type": "execute_result" } ], "source": [ "(im3_t[7:10,17:20] * top_edge).sum()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "As you can see, this little calculation is returning a high number where the 3×3-pixel square represents a top edge (i.e., where there are low values at the top of the square, and high values immediately underneath). That's because the `-1` values in our kernel have little impact in that case, but the `1` values have a lot.\n", "\n", "Let's look a tiny bit at the math. The filter will take any window of size 3×3 in our images, and if we name the pixel values like this:\n", "\n", "$$\\begin{matrix} a1 & a2 & a3 \\\\ a4 & a5 & a6 \\\\ a7 & a8 & a9 \\end{matrix}$$\n", "\n", "it will return $-a1-a2-a3+a7+a8+a9$. If we are in a part of the image where $a1$, $a2$, and $a3$ add up to the same as $a7$, $a8$, and $a9$, then the terms will cancel each other out and we will get 0. However, if $a7$ is greater than $a1$, $a8$ is greater than $a2$, and $a9$ is greater than $a3$, we will get a bigger number as a result. So this filter detects horizontal edges—more precisely, edges where we go from bright parts of the image at the top to darker parts at the bottom.\n", "\n", "Changing our filter to have the row of `1`s at the top and the `-1`s at the bottom would detect horizontal edges that go from dark to light. Putting the `1`s and `-1`s in columns versus rows would give us filters that detect vertical edges. Each set of weights will produce a different kind of outcome.\n", "\n", "Let's create a function to do this for one location, and check it matches our result from before:" ] }, { "cell_type": "code", "execution_count": 36, "metadata": {}, "outputs": [], "source": [ "def apply_kernel(row, col, kernel):\n", " return (im3_t[row-1:row+2,col-1:col+2] * kernel).sum()" ] }, { "cell_type": "code", "execution_count": 37, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "tensor(762.)" ] }, "execution_count": 37, "metadata": {}, "output_type": "execute_result" } ], "source": [ "apply_kernel(5,7,top_edge)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "But note that we can't apply it to the corner (e.g., location 0,0), since there isn't a complete 3×3 square there." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Mapping a Convolution Kernel" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We can map `apply_kernel()` across the coordinate grid. That is, we'll be taking our 3×3 kernel, and applying it to each 3×3 section of our image. For instance, <> shows the positions a 3×3 kernel can be applied to in the first row of a 5×5 image." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"Applying" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "To get a grid of coordinates we can use a *nested list comprehension*, like so:" ] }, { "cell_type": "code", "execution_count": 38, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[[(1, 1), (1, 2), (1, 3), (1, 4)],\n", " [(2, 1), (2, 2), (2, 3), (2, 4)],\n", " [(3, 1), (3, 2), (3, 3), (3, 4)],\n", " [(4, 1), (4, 2), (4, 3), (4, 4)]]" ] }, "execution_count": 38, "metadata": {}, "output_type": "execute_result" } ], "source": [ "[[(i,j) for j in range(1,5)] for i in range(1,5)]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "> note: Nested List Comprehensions: Nested list comprehensions are used a lot in Python, so if you haven't seen them before, take a few minutes to make sure you understand what's happening here, and experiment with writing your own nested list comprehensions." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Here's the result of applying our kernel over a coordinate grid:" ] }, { "cell_type": "code", "execution_count": 39, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEQAAABECAYAAAA4E5OyAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAAALd0lEQVR4nO1bWVMb5xI9s++LBLKAYFN5SFWe8sPyg/LLUpWnpLJgEywkzf7NPnnw7fZo4NoYJJx7S11FsUnDfGd6O6cbaRgGHO2jyV/7Bv5tdgRkYkdAJnYEZGJHQCamfuqXP/300/9tCfrxxx+lh35+9JCJHQGZ2BGQiR0Bmdgnk+rnbNr2P4cGSJL04NcvbY8GZBgGlGWJrutQVRXatkVZlqjrGl3X8c+LosAwDOi6jt8rSRJ0XYcsy5Bl+d7hJUmCYRjQNA2GYcC2bei6DtM0oSgKVPVZz+2L7FF/aRgG9H2Ppmn40HVdI01TFEWBtm1R1zWSJMFqteLvyWNUVYXv+1BVFaqqQpY/RqqiKJBlGZ7nwbIsOI6DrutgWRYDoSjKi3nNZwEZhoEBuLu7Q57n2Gw22G63iOMYaZqi6zr0fY+6rlGWJQNIgEiShDiO2TseCg9d16FpGizLguu6cF0Xs9kMhmFgNptB0zT4vs9eNAX2RQERQqCqKtzd3SGKIvz6669Yr9fIsgxFUez1hhRFYWCCIEAQBPj222/hOA6GYeAwAgBN0/buOY8ChEIA+OD+tm1DCAHTNDEMA2RZhqIoME1z5ylSOAzDsBNC0+t3XYeiKJDnOaqqYi8TQgAA3r9/D9/3Yds2mqaBbduQJAmqqr48IABQ1/U9QNq2heM4cBwHhmHAsiz4vo+zszPoug7XdaGqKjRNwzAMyPMcXdfdC5mmadB1HTabDX9cX18jjmOsVitUVQUASNMUnuehaRoEQQBFUZ5V1Z4MiCzLmM1mcF0X8/kcfd/j+++/xzAMsCwLlmVB13UYhgHHcXBycgJN02CaJnsOgTrOKWRd12EYBrx//x6r1Qp3d3d48+YNoijCb7/9hmEYoKoqTNOErusMBL3vxQFRFAWvX7+GrutYLBYwTRPz+Ry2bcMwjHvldPyZAP2U0aGEEMiyDFEU4e+//8bNzQ1+/vlnCCGw3W4BfEy8VPH6vn/u+e/ZozxE13XYtg3LsmDbNhzHgW3bXEbpdWSSJPFBx/0I2fjJDsPAeYrCh/IS/Q26nqqqnKs0TWPv26d9FhBJkuC6LhzHwWw241xBIUHuT6WWSjAdbFx+gfsAUWNHSTXLMiRJgr7vMZ/PMQwDzs/PH7w3AnOf9qik2jQNmqaBEAKSJEEIgWEYdhIkHbxtW3Zp+n58AHodASeEQF3XEEIgz3MURYEkSQAAlmVBlmUYhsEeSP1O13Wo63rn+i8CSNd1WK/XiKIISZJwAh2XPHL3cWyT+089gjxnu92iLEvc3t4iiiJ+rWmacBwHy+USP/zwA1zXRRiG/LeoIhVFgfV6/fKADMPAhyuKAlVVQQixE79jQNq25feMq8E4V/R9j/V6jTzPsVqtsNls+MBhGMK2bSiKwombPISAPmQb/6iQadsWkiRxHwHsls5xLFPmnyZOIoRxHEMIgZubG2y3W0RRhDRNEYYhgiCA67q4urrC5eUlXr16BU3ToOs6+r5HlmXMmYQQDybs59qjPIQ+P7XMkcfUdc15Io5jRFGEsizR9z0UReEGj6rYuJKMKxF9fJU+5DlGLXvTNLi5uUFRFLi+vkaSJIjjGFVVYbFY4OLiAsvlEqenpzg7O8PFxQVs20YcxwA+eGPbtlitVijLElmWPZif9mEHB4TinsoqsWTqXE3TRBAE8DwPQRDAcRzous4sm65TVRW22+2OdxwilxwcEGLKlFw9z4Ou68yBFosFA+I4DoQQ+OWXX1BVFZIkQdu2O2VeURQmkI7jcFjtC5yDA0JxTwciNSwMQ5imidlsxo2eYRjIsox1FgqRzWYDWZYRhiEMwwAA7lYBMFD7sIMCQkqY4zgIgoDjfqxrkERIjdcYRCEEkiTB9fU15wtVVbFcLmHbNs7Pz2FZFubzOVzXZVCnjPpL7OCAULdJvcVUNRvrstTFUj9TliXyPEcURajrGkVRQJIkbDYbmKaJoigwm81YcyHe9ZzwOSggqqqyPLBcLqFpGlzX5biXZZmrEH1O0xS3t7fYbrcIwxBZlmGxWCDPc9ze3u5oM5vNBnmeQ1VVFEXBCde2bdi2/bR73icAUxuHzGKxYPceJ8KpHpvnOU5OTpBlGebzOdI0RRAE3MxlWYbff/8dRVEgiiL0fQ9JkjCfz/nvUq56ih08qVZVBUVRWA4syxLAx0RIWoosy9yVGoYBz/OgaRqyLIOiKMjzHIZhQAgBz/NQFAUrarZtc8+TZRnLFYqiQNf1L7rngwNC+ihJiJZlcWcqyzInVMorpLo3TQPP8yCEgOu6TASFEFitViiKAm/fvkWe50wHaCZEih2F65fYQQGh5ChJEpIkQV3XTOfLsuThFIEyLp1938M0TUiShNlsxlyoKAqUZQlFURCGIYdfVVV8PSKCT6EaL9K6U4UgjyDeQlopTeuIx3ieB9M0Yds2P+mqqqBpGoqiYKKnaRp3sBSSwAepsW3bJ7X2Bwdk/PWYEQ/DAEVRdoSepmlY3yCaPxaTSKslNZ/0Vdu2dyRM8jpq3L7EXm5o+h+j5owOnOf5jhRJmq3v+1yRTNNkWZLmxE3TsIxJ3evY/lWN2fipUhzTDZLSRs0UgJ1YL8uSEyx5QdM0AICqqrjXoHAggB4CYCxOPdb2Cgh1mTQIHxMzKq2+7/Nk/6GpPnWteZ5jvV7z++j61NoT6JIk8cxmfB90rS8dte4VkGmDRXwE+JDoVFW9p8JPbXzgh9Q3ADuznzEfGl8DAHvWl9heAKGbz/Mcd3d3SNMU796944NpmobFYsEqmGVZ/5Wd0rWqqkKWZWiaBnmeA8A9LwuCYKdSkXVdh7Ztn8SA9wYIVQshBE/buq7jwTc9NXqqD8U8XadpGlbGiN8A4JxChJHyzLSiPEda3AsgZVkiTVPEccwjgqqqoKoqjz9PT0+ZdGmaxkl3fA2i+5vNBmma4s8//+TybJom033yCtJS6GeUY2ib6auFDN0ANWHERsfchLpIej2FE/Ah9xAgaZpivV5jvV7j7du3LCG4rotXr14xQLqu87CdvITy05g0fhVAaEWCll26rsObN2+4de/7Hn/88QcfvG3bneoBgNX48SiTds8uLy/h+z6WyyVc18Xr168RhiELRVSm8zxnfYU41FcBhFrxKbus65oZKQ2m/vrrLwgheLVh/FrqVJum4R0TwzAQhiFc14Xv+/A8D/P5nHVYyk+k39JDGHvgiwNCnSWVP0qYWZYx79hsNsxtyrK89wQpMdIale/7OD8/Z4+gAZbnebi8vOQ+RpZlng/TvIckhqfYXgChUQINmsjGQCiKwk+P8syYfLmuC13X4fs+rq6uEIYhLi8v4Xkerq6u4Louzs7OYFkWTk9PdwgcLQVSUqYc9hTbW9lt25bZKZXaIAhg2zayLMM333yDLMvw7t07pvCUX2gbybZthGGIxWIB3/dxcXEB13VxcnICwzDgui7nKdJaqqpCFEXIsgxpmt4D+qsAQsSLxGRSvGRZxtnZGbquw3fffYe6rrHdbnnmQhrHeF/N930eS9DQynEcZrnjOTENv9brNW8TkHc8tRfZW9mlZop2SMZVAgBXH9/3d7aFiIBR0zXeQ6V1LZIBqqpC3/cMZhRFvFdCv3vuvHdvgAghuIschgGu6+4su/i+/yRKTvyobVsGgqTE9XrNk8F97YnsLWQkSWKWCwBxHO/sqhNFp97jU2sV9EGhKIRA0zQ8E06ShCsVldd9bQLsNammacqr35vNBoZh8Jqm4zjcchNDHSvu1FnSPj1JB1VV8aYAcRv6J4R/7TrEdKmOpm5j7ZPY57h/GNN4EpPGQyviI2ma7gDx1Lb8MXYQxYw6TlmWEUURt9afCpmHFnPGoQNgL0nzc7Z3xWz89aGe4iFNOjTi/2t2/J+7iR0BmdgRkIkdAZnYEZCJHQGZ2D+ACU/MswDoqgAAAABJRU5ErkJggg==\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "rng = range(1,27)\n", "top_edge3 = tensor([[apply_kernel(i,j,top_edge) for j in rng] for i in rng])\n", "\n", "show_image(top_edge3);" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Looking good! Our top edges are black, and bottom edges are white (since they are the *opposite* of top edges). Now that our image contains negative numbers too, `matplotlib` has automatically changed our colors so that white is the smallest number in the image, black the highest, and zeros appear as gray.\n", "\n", "We can try the same thing for left edges:" ] }, { "cell_type": "code", "execution_count": 40, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "left_edge = tensor([[-1,1,0],\n", " [-1,1,0],\n", " [-1,1,0]]).float()\n", "\n", "left_edge3 = tensor([[apply_kernel(i,j,left_edge) for j in rng] for i in rng])\n", "\n", "show_image(left_edge3);" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "As we mentioned before, a convolution is the operation of applying such a kernel over a grid in this way. In the paper [\"A Guide to Convolution Arithmetic for Deep Learning\"](https://arxiv.org/abs/1603.07285) there are many great diagrams showing how image kernels can be applied. Here's an example from the paper showing (at the bottom) a light blue 4×4 image, with a dark blue 3×3 kernel being applied, creating a 2×2 green output activation map at the top. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"Result" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Look at the shape of the result. If the original image has a height of `h` and a width of `w`, how many 3×3 windows can we find? As you can see from the example, there are `h-2` by `w-2` windows, so the image we get has a result as a height of `h-2` and a width of `w-2`." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We won't implement this convolution function from scratch, but use PyTorch's implementation instead (it is way faster than anything we could do in Python)." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Convolutions in PyTorch" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Convolution is such an important and widely used operation that PyTorch has it built in. It's called `F.conv2d` (recall that `F` is a fastai import from `torch.nn.functional`, as recommended by PyTorch). The PyTorch docs tell us that it includes these parameters:\n", "\n", "- input:: input tensor of shape `(minibatch, in_channels, iH, iW)`\n", "- weight:: filters of shape `(out_channels, in_channels, kH, kW)`\n", "\n", "Here `iH,iW` is the height and width of the image (i.e., `28,28`), and `kH,kW` is the height and width of our kernel (`3,3`). But apparently PyTorch is expecting rank-4 tensors for both these arguments, whereas currently we only have rank-2 tensors (i.e., matrices, or arrays with two axes).\n", "\n", "The reason for these extra axes is that PyTorch has a few tricks up its sleeve. The first trick is that PyTorch can apply a convolution to multiple images at the same time. That means we can call it on every item in a batch at once!\n", "\n", "The second trick is that PyTorch can apply multiple kernels at the same time. So let's create the diagonal-edge kernels too, and then stack all four of our edge kernels into a single tensor:" ] }, { "cell_type": "code", "execution_count": 41, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "torch.Size([4, 3, 3])" ] }, "execution_count": 41, "metadata": {}, "output_type": "execute_result" } ], "source": [ "diag1_edge = tensor([[ 0,-1, 1],\n", " [-1, 1, 0],\n", " [ 1, 0, 0]]).float()\n", "diag2_edge = tensor([[ 1,-1, 0],\n", " [ 0, 1,-1],\n", " [ 0, 0, 1]]).float()\n", "\n", "edge_kernels = torch.stack([left_edge, top_edge, diag1_edge, diag2_edge])\n", "edge_kernels.shape" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "To test this, we'll need a `DataLoader` and a sample mini-batch. Let's use the data block API:" ] }, { "cell_type": "code", "execution_count": 42, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "torch.Size([64, 1, 28, 28])" ] }, "execution_count": 42, "metadata": {}, "output_type": "execute_result" } ], "source": [ "mnist = DataBlock((ImageBlock(cls=PILImageBW), CategoryBlock), \n", " get_items=get_image_files, \n", " splitter=GrandparentSplitter(),\n", " get_y=parent_label)\n", "\n", "dls = mnist.dataloaders(path)\n", "xb,yb = first(dls.valid)\n", "xb.shape" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "By default, fastai puts data on the GPU when using data blocks. Let's move it to the CPU for our examples:" ] }, { "cell_type": "code", "execution_count": 43, "metadata": {}, "outputs": [], "source": [ "xb,yb = to_cpu(xb),to_cpu(yb)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "One batch contains 64 images, each of 1 channel, with 28×28 pixels. `F.conv2d` can handle multichannel (i.e., color) images too. A *channel* is a single basic color in an image—for regular full-color images there are three channels, red, green, and blue. PyTorch represents an image as a rank-3 tensor, with dimensions `[channels, rows, columns]`.\n", "\n", "We'll see how to handle more than one channel later in this chapter. Kernels passed to `F.conv2d` need to be rank-4 tensors: `[channels_in, features_out, rows, columns]`. `edge_kernels` is currently missing one of these. We need to tell PyTorch that the number of input channels in the kernel is one, which we can do by inserting an axis of size one (this is known as a *unit axis*) in the first location, where the PyTorch docs show `in_channels` is expected. To insert a unit axis into a tensor, we use the `unsqueeze` method:" ] }, { "cell_type": "code", "execution_count": 44, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(torch.Size([4, 3, 3]), torch.Size([4, 1, 3, 3]))" ] }, "execution_count": 44, "metadata": {}, "output_type": "execute_result" } ], "source": [ "edge_kernels.shape,edge_kernels.unsqueeze(1).shape" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This is now the correct shape for `edge_kernels`. Let's pass this all to `conv2d`:" ] }, { "cell_type": "code", "execution_count": 45, "metadata": {}, "outputs": [], "source": [ "edge_kernels = edge_kernels.unsqueeze(1)" ] }, { "cell_type": "code", "execution_count": 46, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "torch.Size([64, 4, 26, 26])" ] }, "execution_count": 46, "metadata": {}, "output_type": "execute_result" } ], "source": [ "batch_features = F.conv2d(xb, edge_kernels)\n", "batch_features.shape" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The output shape shows we gave 64 images in the mini-batch, 4 kernels, and 26×26 edge maps (we started with 28×28 images, but lost one pixel from each side as discussed earlier). We can see we get the same results as when we did this manually:" ] }, { "cell_type": "code", "execution_count": 47, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "show_image(batch_features[0,0]);" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The most important trick that PyTorch has up its sleeve is that it can use the GPU to do all this work in parallel—that is, applying multiple kernels, to multiple images, across multiple channels. Doing lots of work in parallel is critical to getting GPUs to work efficiently; if we did each of these operations one at a time, we'd often run hundreds of times slower (and if we used our manual convolution loop from the previous section, we'd be millions of times slower!). Therefore, to become a strong deep learning practitioner, one skill to practice is giving your GPU plenty of work to do at a time." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "It would be nice to not lose those two pixels on each axis. The way we do that is to add *padding*, which is simply additional pixels added around the outside of our image. Most commonly, pixels of zeros are added. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Strides and Padding" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "With appropriate padding, we can ensure that the output activation map is the same size as the original image, which can make things a lot simpler when we construct our architectures. <> shows how adding padding allows us to apply the kernels in the image corners." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"A" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "With a 5×5 input, 4×4 kernel, and 2 pixels of padding, we end up with a 6×6 activation map, as we can see in <>." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"A" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "If we add a kernel of size `ks` by `ks` (with `ks` an odd number), the necessary padding on each side to keep the same shape is `ks//2`. An even number for `ks` would require a different amount of padding on the top/bottom and left/right, but in practice we almost never use an even filter size.\n", "\n", "So far, when we have applied the kernel to the grid, we have moved it one pixel over at a time. But we can jump further; for instance, we could move over two pixels after each kernel application, as in <>. This is known as a *stride-2* convolution. The most common kernel size in practice is 3×3, and the most common padding is 1. As you'll see, stride-2 convolutions are useful for decreasing the size of our outputs, and stride-1 convolutions are useful for adding layers without changing the output size." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"A" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "In an image of size `h` by `w`, using a padding of 1 and a stride of 2 will give us a result of size `(h+1)//2` by `(w+1)//2`. The general formula for each dimension is `(n + 2*pad - ks)//stride + 1`, where `pad` is the padding, `ks`, the size of our kernel, and `stride` is the stride." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's now take a look at how the pixel values of the result of our convolutions are computed." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Understanding the Convolution Equations" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "To explain the math behind convolutions, fast.ai student Matt Kleinsmith came up with the very clever idea of showing [CNNs from different viewpoints](https://medium.com/impactai/cnns-from-different-viewpoints-fab7f52d159c). In fact, it's so clever, and so helpful, we're going to show it here too!\n", "\n", "Here's our 3×3 pixel image, with each pixel labeled with a letter:" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"The" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "And here's our kernel, with each weight labeled with a Greek letter:" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"The" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Since the filter fits in the image four times, we have four results:" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"The" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "<> shows how we applied the kernel to each section of the image to yield each result." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"Applying" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The equation view is in <>." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"The" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Notice that the bias term, *b*, is the same for each section of the image. You can consider the bias as part of the filter, just like the weights (α, β, γ, δ) are part of the filter." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Here's an interesting insight—a convolution can be represented as a special kind of matrix multiplication, as illustrated in <>. The weight matrix is just like the ones from traditional neural networks. However, this weight matrix has two special properties:\n", "\n", "1. The zeros shown in gray are untrainable. This means that they’ll stay zero throughout the optimization process.\n", "1. Some of the weights are equal, and while they are trainable (i.e., changeable), they must remain equal. These are called *shared weights*.\n", "\n", "The zeros correspond to the pixels that the filter can't touch. Each row of the weight matrix corresponds to one application of the filter." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"Convolution" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now that we understand what a convolution is, let's use them to build a neural net." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Our First Convolutional Neural Network" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "There is no reason to believe that some particular edge filters are the most useful kernels for image recognition. Furthermore, we've seen that in later layers convolutional kernels become complex transformations of features from lower levels, but we don't have a good idea of how to manually construct these.\n", "\n", "Instead, it would be best to learn the values of the kernels. We already know how to do this—SGD! In effect, the model will learn the features that are useful for classification.\n", "\n", "When we use convolutions instead of (or in addition to) regular linear layers we create a *convolutional neural network* (CNN)." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Creating the CNN" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's go back to the basic neural network we had in <>. It was defined like this:" ] }, { "cell_type": "code", "execution_count": 48, "metadata": {}, "outputs": [], "source": [ "simple_net = nn.Sequential(\n", " nn.Linear(28*28,30),\n", " nn.ReLU(),\n", " nn.Linear(30,1)\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We can view a model's definition:" ] }, { "cell_type": "code", "execution_count": 49, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Sequential(\n", " (0): Linear(in_features=784, out_features=30, bias=True)\n", " (1): ReLU()\n", " (2): Linear(in_features=30, out_features=1, bias=True)\n", ")" ] }, "execution_count": 49, "metadata": {}, "output_type": "execute_result" } ], "source": [ "simple_net" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We now want to create a similar architecture to this linear model, but using convolutional layers instead of linear. `nn.Conv2d` is the module equivalent of `F.conv2d`. It's more convenient than `F.conv2d` when creating an architecture, because it creates the weight matrix for us automatically when we instantiate it.\n", "\n", "Here's a possible architecture:" ] }, { "cell_type": "code", "execution_count": 50, "metadata": {}, "outputs": [], "source": [ "broken_cnn = sequential(\n", " nn.Conv2d(1,30, kernel_size=3, padding=1),\n", " nn.ReLU(),\n", " nn.Conv2d(30,1, kernel_size=3, padding=1)\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "One thing to note here is that we didn't need to specify 28×28 as the input size. That's because a linear layer needs a weight in the weight matrix for every pixel, so it needs to know how many pixels there are, but a convolution is applied over each pixel automatically. The weights only depend on the number of input and output channels and the kernel size, as we saw in the previous section.\n", "\n", "Think about what the output shape is going to be, then let's try it and see:" ] }, { "cell_type": "code", "execution_count": 51, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "torch.Size([64, 1, 28, 28])" ] }, "execution_count": 51, "metadata": {}, "output_type": "execute_result" } ], "source": [ "broken_cnn(xb).shape" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This is not something we can use to do classification, since we need a single output activation per image, not a 28×28 map of activations. One way to deal with this is to use enough stride-2 convolutions such that the final layer is size 1. That is, after one stride-2 convolution the size will be 14×14, after two it will be 7×7, then 4×4, 2×2, and finally size 1.\n", "\n", "Let's try that now. First, we'll define a function with the basic parameters we'll use in each convolution:" ] }, { "cell_type": "code", "execution_count": 52, "metadata": {}, "outputs": [], "source": [ "def conv(ni, nf, ks=3, act=True):\n", " res = nn.Conv2d(ni, nf, stride=2, kernel_size=ks, padding=ks//2)\n", " if act: res = nn.Sequential(res, nn.ReLU())\n", " return res" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "> important: Refactoring: Refactoring parts of your neural networks like this makes it much less likely you'll get errors due to inconsistencies in your architectures, and makes it more obvious to the reader which parts of your layers are actually changing." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "When we use a stride-2 convolution, we often increase the number of features at the same time. This is because we're decreasing the number of activations in the activation map by a factor of 4; we don't want to decrease the capacity of a layer by too much at a time." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "> jargon: channels and features: These two terms are largely used interchangeably, and refer to the size of the second axis of a weight matrix, which is, the number of activations per grid cell after a convolution. _Features_ is never used to refer to the input data, but _channels_ can refer to either the input data (generally channels are colors) or activations inside the network." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Here is how we can build a simple CNN:" ] }, { "cell_type": "code", "execution_count": 53, "metadata": {}, "outputs": [], "source": [ "simple_cnn = sequential(\n", " conv(1 ,4), #14x14\n", " conv(4 ,8), #7x7\n", " conv(8 ,16), #4x4\n", " conv(16,32), #2x2\n", " conv(32,2, act=False), #1x1\n", " Flatten(),\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "> j: I like to add comments like the ones here after each convolution to show how large the activation map will be after each layer. These comments assume that the input size is 28*28" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now the network outputs two activations, which map to the two possible levels in our labels:" ] }, { "cell_type": "code", "execution_count": 54, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "torch.Size([64, 2])" ] }, "execution_count": 54, "metadata": {}, "output_type": "execute_result" } ], "source": [ "simple_cnn(xb).shape" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We can now create our `Learner`:" ] }, { "cell_type": "code", "execution_count": 55, "metadata": {}, "outputs": [], "source": [ "learn = Learner(dls, simple_cnn, loss_func=F.cross_entropy, metrics=accuracy)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "To see exactly what's going on in the model, we can use `summary`:" ] }, { "cell_type": "code", "execution_count": 57, "metadata": {}, "outputs": [ { "data": { "text/html": [], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/plain": [ "Sequential (Input shape: 64)\n", "============================================================================\n", "Layer (type) Output Shape Param # Trainable \n", "============================================================================\n", " 64 x 4 x 14 x 14 \n", "Conv2d 40 True \n", "ReLU \n", "____________________________________________________________________________\n", " 64 x 8 x 7 x 7 \n", "Conv2d 296 True \n", "ReLU \n", "____________________________________________________________________________\n", " 64 x 16 x 4 x 4 \n", "Conv2d 1168 True \n", "ReLU \n", "____________________________________________________________________________\n", " 64 x 32 x 2 x 2 \n", "Conv2d 4640 True \n", "ReLU \n", "____________________________________________________________________________\n", " 64 x 2 x 1 x 1 \n", "Conv2d 578 True \n", "____________________________________________________________________________\n", " [] \n", "Flatten \n", "____________________________________________________________________________\n", "\n", "Total params: 6,722\n", "Total trainable params: 6,722\n", "Total non-trainable params: 0\n", "\n", "Optimizer used: \n", "Loss function: \n", "\n", "Callbacks:\n", " - TrainEvalCallback\n", " - Recorder\n", " - ProgressCallback" ] }, "execution_count": 57, "metadata": {}, "output_type": "execute_result" } ], "source": [ "learn.summary()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Note that the output of the final `Conv2d` layer is `64x2x1x1`. We need to remove those extra `1x1` axes; that's what `Flatten` does. It's basically the same as PyTorch's `squeeze` method, but as a module.\n", "\n", "Let's see if this trains! Since this is a deeper network than we've built from scratch before, we'll use a lower learning rate and more epochs:" ] }, { "cell_type": "code", "execution_count": 58, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracytime
00.0596770.0400940.98577000:05
10.0186340.0250030.99067700:04
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "learn.fit_one_cycle(2, 0.01)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Success! It's getting closer to the `resnet18` result we had, although it's not quite there yet, and it's taking more epochs, and we're needing to use a lower learning rate. We still have a few more tricks to learn, but we're getting closer and closer to being able to create a modern CNN from scratch." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Understanding Convolution Arithmetic" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We can see from the summary that we have an input of size `64x1x28x28`. The axes are `batch,channel,height,width`. This is often represented as `NCHW` (where `N` refers to batch size). Tensorflow, on the other hand, uses `NHWC` axis order. The first layer is:" ] }, { "cell_type": "code", "execution_count": 59, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Sequential(\n", " (0): Conv2d(1, 4, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1))\n", " (1): ReLU()\n", ")" ] }, "execution_count": 59, "metadata": {}, "output_type": "execute_result" } ], "source": [ "m = learn.model[0]\n", "m" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "So we have 1 input channel, 4 output channels, and a 3×3 kernel. Let's check the weights of the first convolution:" ] }, { "cell_type": "code", "execution_count": 60, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "torch.Size([4, 1, 3, 3])" ] }, "execution_count": 60, "metadata": {}, "output_type": "execute_result" } ], "source": [ "m[0].weight.shape" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The summary shows we have 40 parameters, and `4*1*3*3` is 36. What are the other four parameters? Let's see what the bias contains:" ] }, { "cell_type": "code", "execution_count": 61, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "torch.Size([4])" ] }, "execution_count": 61, "metadata": {}, "output_type": "execute_result" } ], "source": [ "m[0].bias.shape" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We can now use this information to clarify our statement in the previous section: \"When we use a stride-2 convolution, we often increase the number of features because we're decreasing the number of activations in the activation map by a factor of 4; we don't want to decrease the capacity of a layer by too much at a time.\"\n", "\n", "There is one bias for each channel. (Sometimes channels are called *features* or *filters* when they are not input channels.) The output shape is `64x4x14x14`, and this will therefore become the input shape to the next layer. The next layer, according to `summary`, has 296 parameters. Let's ignore the batch axis to keep things simple. So for each of `14*14=196` locations we are multiplying `296-8=288` weights (ignoring the bias for simplicity), so that's `196*288=56_448` multiplications at this layer. The next layer will have `7*7*(1168-16)=56_448` multiplications.\n", "\n", "What happened here is that our stride-2 convolution halved the *grid size* from `14x14` to `7x7`, and we doubled the *number of filters* from 8 to 16, resulting in no overall change in the amount of computation. If we left the number of channels the same in each stride-2 layer, the amount of computation being done in the net would get less and less as it gets deeper. But we know that the deeper layers have to compute semantically rich features (such as eyes or fur), so we wouldn't expect that doing *less* computation would make sense." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Another way to think of this is based on receptive fields." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Receptive Fields" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The *receptive field* is the area of an image that is involved in the calculation of a layer. On the [book's website](https://book.fast.ai/), you'll find an Excel spreadsheet called *conv-example.xlsx* that shows the calculation of two stride-2 convolutional layers using an MNIST digit. Each layer has a single kernel. <> shows what we see if we click on one of the cells in the *conv2* section, which shows the output of the second convolutional layer, and click *trace precedents*." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"Immediate" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Here, the cell with the green border is the cell we clicked on, and the blue highlighted cells are its *precedents*—that is, the cells used to calculate its value. These cells are the corresponding 3×3 area of cells from the input layer (on the left), and the cells from the filter (on the right). Let's now click *trace precedents* again, to see what cells are used to calculate these inputs. <> shows what happens." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"Secondary" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "In this example, we have just two convolutional layers, each of stride 2, so this is now tracing right back to the input image. We can see that a 7×7 area of cells in the input layer is used to calculate the single green cell in the Conv2 layer. This 7×7 area is the *receptive field* in the input of the green activation in Conv2. We can also see that a second filter kernel is needed now, since we have two layers.\n", "\n", "As you see from this example, the deeper we are in the network (specifically, the more stride-2 convs we have before a layer), the larger the receptive field for an activation in that layer. A large receptive field means that a large amount of the input image is used to calculate each activation in that layer is. We now know that in the deeper layers of the network we have semantically rich features, corresponding to larger receptive fields. Therefore, we'd expect that we'd need more weights for each of our features to handle this increasing complexity. This is another way of saying the same thing we mentioned in the previous section: when we introduce a stride-2 conv in our network, we should also increase the number of channels." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "When writing this particular chapter, we had a lot of questions we needed answers for, to be able to explain CNNs to you as best we could. Believe it or not, we found most of the answers on Twitter. We're going to take a quick break to talk to you about that now, before we move on to color images." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### A Note About Twitter" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We are not, to say the least, big users of social networks in general. But our goal in writing this book is to help you become the best deep learning practitioner you can, and we would be remiss not to mention how important Twitter has been in our own deep learning journeys.\n", "\n", "You see, there's another part of Twitter, far away from Donald Trump and the Kardashians, which is the part of Twitter where deep learning researchers and practitioners talk shop every day. As we were writing this section, Jeremy wanted to double-check that what we were saying about stride-2 convolutions was accurate, so he asked on Twitter:" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"twitter" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "A few minutes later, this answer popped up:" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"twitter" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Christian Szegedy is the first author of [Inception](https://arxiv.org/pdf/1409.4842.pdf), the 2014 ImageNet winner and source of many key insights used in modern neural networks. Two hours later, this appeared:" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"twitter" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Do you recognize that name? You saw it in <>, when we were talking about the Turing Award winners who established the foundations of deep learning today!\n", "\n", "Jeremy also asked on Twitter for help checking our description of label smoothing in <> was accurate, and got a response again from directly from Christian Szegedy (label smoothing was originally introduced in the Inception paper):" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"twitter" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Many of the top people in deep learning today are Twitter regulars, and are very open about interacting with the wider community. One good way to get started is to look at a list of Jeremy's [recent Twitter likes](https://twitter.com/jeremyphoward/likes), or [Sylvain's](https://twitter.com/GuggerSylvain/likes). That way, you can see a list of Twitter users that we think have interesting and useful things to say.\n", "\n", "Twitter is the main way we both stay up to date with interesting papers, software releases, and other deep learning news. For making connections with the deep learning community, we recommend getting involved both in the [fast.ai forums](https://forums.fast.ai) and on Twitter." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "That said, let's get back to the meat of this chapter. Up until now, we have only shown you examples of pictures in black and white, with one value per pixel. In practice, most colored images have three values per pixel to define their color. We'll look at working with color images next." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Color Images" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "A colour picture is a rank-3 tensor:" ] }, { "cell_type": "code", "execution_count": 62, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "torch.Size([3, 1000, 846])" ] }, "execution_count": 62, "metadata": {}, "output_type": "execute_result" } ], "source": [ "im = image2tensor(Image.open(image_bear()))\n", "im.shape" ] }, { "cell_type": "code", "execution_count": 63, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "show_image(im);" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The first axis contains the channels, red, green, and blue:" ] }, { "cell_type": "code", "execution_count": 64, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "_,axs = subplots(1,3)\n", "for bear,ax,color in zip(im,axs,('Reds','Greens','Blues')):\n", " show_image(255-bear, ax=ax, cmap=color)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We saw what the convolution operation was for one filter on one channel of the image (our examples were done on a square). A convolutional layer will take an image with a certain number of channels (three for the first layer for regular RGB color images) and output an image with a different number of channels. Like our hidden size that represented the numbers of neurons in a linear layer, we can decide to have as many filters as we want, and each of them will be able to specialize, some to detect horizontal edges, others to detect vertical edges and so forth, to give something like we studied in <>.\n", "\n", "In one sliding window, we have a certain number of channels and we need as many filters (we don't use the same kernel for all the channels). So our kernel doesn't have a size of 3 by 3, but `ch_in` (for channels in) is 3 by 3. On each channel, we multiply the elements of our window by the elements of the coresponding filter, then sum the results (as we saw before) and sum over all the filters. In the example given in <>, the result of our conv layer on that window is red + green + blue." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"Convolution" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "So, in order to apply a convolution to a color picture we require a kernel tensor with a size that matches the first axis. At each location, the corresponding parts of the kernel and the image patch are multiplied together.\n", "\n", "These are then all added together, to produce a single number, for each grid location, for each output feature, as shown in <>." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"Adding" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Then we have `ch_out` filters like this, so in the end, the result of our convolutional layer will be a batch of images with `ch_out` channels and a height and width given by the formula outlined earlier. This give us `ch_out` tensors of size `ch_in x ks x ks` that we represent in one big tensor of four dimensions. In PyTorch, the order of the dimensions for those weights is `ch_out x ch_in x ks x ks`.\n", "\n", "Additionally, we may want to have a bias for each filter. In the preceding example, the final result for our convolutional layer would be $y_{R} + y_{G} + y_{B} + b$ in that case. Like in a linear layer, there are as many bias as we have kernels, so the biases is a vector of size `ch_out`.\n", "\n", "There are no special mechanisms required when setting up a CNN for training with color images. Just make sure your first layer has three inputs.\n", "\n", "There are lots of ways of processing color images. For instance, you can change them to black and white, change from RGB to HSV (hue, saturation, and value) color space, and so forth. In general, it turns out experimentally that changing the encoding of colors won't make any difference to your model results, as long as you don't lose information in the transformation. So, transforming to black and white is a bad idea, since it removes the color information entirely (and this can be critical; for instance, a pet breed may have a distinctive color); but converting to HSV generally won't make any difference.\n", "\n", "Now you know what those pictures in <> of \"what a neural net learns\" from the [Zeiler and Fergus paper](https://arxiv.org/abs/1311.2901) mean! This is their picture of some of the layer 1 weights which we showed:" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"Layer" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This is taking the three slices of the convolutional kernel, for each output feature, and displaying them as images. We can see that even though the creators of the neural net never explicitly created kernels to find edges, for instance, the neural net automatically discovered these features using SGD.\n", "\n", "Now let's see how we can train these CNNs, and show you all the techniques fastai uses under the hood for efficient training." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Improving Training Stability" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Since we are so good at recognizing 3s from 7s, let's move on to something harder—recognizing all 10 digits. That means we'll need to use `MNIST` instead of `MNIST_SAMPLE`:" ] }, { "cell_type": "code", "execution_count": 65, "metadata": {}, "outputs": [ { "data": { "text/html": [], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "path = untar_data(URLs.MNIST)" ] }, { "cell_type": "code", "execution_count": 66, "metadata": {}, "outputs": [], "source": [ "#hide\n", "Path.BASE_PATH = path" ] }, { "cell_type": "code", "execution_count": 67, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(#2) [Path('training'),Path('testing')]" ] }, "execution_count": 67, "metadata": {}, "output_type": "execute_result" } ], "source": [ "path.ls()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The data is in two folders named *training* and *testing*, so we have to tell `GrandparentSplitter` about that (it defaults to `train` and `valid`). We did do that in the `get_dls` function, which we create to make it easy to change our batch size later:" ] }, { "cell_type": "code", "execution_count": 68, "metadata": {}, "outputs": [], "source": [ "def get_dls(bs=64):\n", " return DataBlock(\n", " blocks=(ImageBlock(cls=PILImageBW), CategoryBlock), \n", " get_items=get_image_files, \n", " splitter=GrandparentSplitter('training','testing'),\n", " get_y=parent_label,\n", " batch_tfms=Normalize()\n", " ).dataloaders(path, bs=bs)\n", "\n", "dls = get_dls()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Remember, it's always a good idea to look at your data before you use it:" ] }, { "cell_type": "code", "execution_count": 69, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "dls.show_batch(max_n=9, figsize=(4,4))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now that we have our data ready, we can train a simple model on it." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### A Simple Baseline" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Earlier in this chapter, we built a model based on a `conv` function like this:" ] }, { "cell_type": "code", "execution_count": 70, "metadata": {}, "outputs": [], "source": [ "def conv(ni, nf, ks=3, act=True):\n", " res = nn.Conv2d(ni, nf, stride=2, kernel_size=ks, padding=ks//2)\n", " if act: res = nn.Sequential(res, nn.ReLU())\n", " return res" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's start with a basic CNN as a baseline. We'll use the same one as earlier, but with one tweak: we'll use more activations. Since we have more numbers to differentiate, it's likely we will need to learn more filters.\n", "\n", "As we discussed, we generally want to double the number of filters each time we have a stride-2 layer. One way to increase the number of filters throughout our network is to double the number of activations in the first layer–then every layer after that will end up twice as big as in the previous version as well.\n", "\n", "> Important: But there is a subtle problem with this. Consider the kernel that is being applied to each pixel. By default, we use a 3×3-pixel kernel. That means that there are a total of 3×3 = 9 pixels that the kernel is being applied to at each location. Previously, our first layer had four output filters. That meant that there were four values being computed from nine pixels at each location. Think about what happens if we double this output to eight filters. Then when we apply our kernel we will be using nine pixels to calculate eight numbers. That means it isn't really learning much at all: the output size is almost the same as the input size. Neural networks will only create useful features if they're forced to do so—that is, if the number of outputs from an operation is significantly smaller than the number of inputs.\n", "\n", "To fix this, we can use a larger kernel in the first layer. If we use a kernel of 5×5 pixels then there are 25 pixels being used at each kernel application. Creating eight filters from this will mean the neural net will have to find some useful features:" ] }, { "cell_type": "code", "execution_count": 71, "metadata": {}, "outputs": [], "source": [ "def simple_cnn():\n", " return sequential(\n", " conv(1 ,8, ks=5), #14x14\n", " conv(8 ,16), #7x7\n", " conv(16,32), #4x4\n", " conv(32,64), #2x2\n", " conv(64,10, act=False), #1x1\n", " Flatten(),\n", " )" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "As you'll see in a moment, we can look inside our models while they're training in order to try to find ways to make them train better. To do this we use the `ActivationStats` callback, which records the mean, standard deviation, and histogram of activations of every trainable layer (as we've seen, callbacks are used to add behavior to the training loop; we'll explore how they work in <>):" ] }, { "cell_type": "code", "execution_count": 72, "metadata": {}, "outputs": [], "source": [ "from fastai.callback.hook import *" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We want to train quickly, so that means training at a high learning rate. Let's see how we go at 0.06:" ] }, { "cell_type": "code", "execution_count": 73, "metadata": {}, "outputs": [], "source": [ "def fit(epochs=1):\n", " learn = Learner(dls, simple_cnn(), loss_func=F.cross_entropy,\n", " metrics=accuracy, cbs=ActivationStats(with_hist=True))\n", " learn.fit(epochs, 0.06)\n", " return learn" ] }, { "cell_type": "code", "execution_count": 74, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/home/niyazi/anaconda3/envs/fastbook/lib/python3.8/site-packages/fastai/callback/core.py:51: UserWarning: You are shadowing an attribute (modules) that exists in the learner. Use `self.learn.modules` to avoid this\n", " warn(f\"You are shadowing an attribute ({name}) that exists in the learner. Use `self.learn.{name}` to avoid this\")\n" ] }, { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracytime
02.3047752.3070390.09820000:21
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "learn = fit()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This didn't train at all well! Let's find out why.\n", "\n", "One handy feature of the callbacks passed to `Learner` is that they are made available automatically, with the same name as the callback class, except in `snake_case`. So, our `ActivationStats` callback can be accessed through `activation_stats`. I'm sure you remember `learn.recorder`... can you guess how that is implemented? That's right, it's a callback called `Recorder`!\n", "\n", "`ActivationStats` includes some handy utilities for plotting the activations during training. `plot_layer_stats(idx)` plots the mean and standard deviation of the activations of layer number *`idx`*, along with the percentage of activations near zero. Here's the first layer's plot:" ] }, { "cell_type": "code", "execution_count": 75, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "learn.activation_stats.plot_layer_stats(0)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "> Important: Generally our model should have a consistent, or at least smooth, mean and standard deviation of layer activations during training. Activations near zero are particularly problematic, because it means we have computation in the model that's doing nothing at all (since multiplying by zero gives zero). When you have some zeros in one layer, they will therefore generally carry over to the next layer... which will then create more zeros. Here's the penultimate layer of our network:" ] }, { "cell_type": "code", "execution_count": 76, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "learn.activation_stats.plot_layer_stats(-2)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "As expected, the problems get worse towards the end of the network, as the instability and zero activations compound over layers. Let's look at what we can do to make training more stable." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Increase Batch Size" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "> Important: One way to make training more stable is to increase the batch size. Larger batches have gradients that are more accurate, since they're calculated from more data. On the downside, though, a larger batch size means fewer batches per epoch, which means less opportunities for your model to update weights. Let's see if a batch size of 512 helps:" ] }, { "cell_type": "code", "execution_count": 77, "metadata": {}, "outputs": [], "source": [ "dls = get_dls(512)" ] }, { "cell_type": "code", "execution_count": 78, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracytime
02.3160742.3019380.11350000:11
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "learn = fit()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's see what the penultimate layer looks like:" ] }, { "cell_type": "code", "execution_count": 80, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "learn.activation_stats.plot_layer_stats(-2)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Again, we've got most of our activations near zero. Let's see what else we can do to improve training stability." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 1cycle Training" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Our initial weights are not well suited to the task we're trying to solve. Therefore, it is dangerous to begin training with a high learning rate: we may very well make the training diverge instantly, as we've seen. We probably don't want to end training with a high learning rate either, so that we don't skip over a minimum. But we want to train at a high learning rate for the rest of the training period, because we'll be able to train more quickly that way. Therefore, we should change the learning rate during training, from low, to high, and then back to low again.\n", "\n", "Leslie Smith (yes, the same guy that invented the learning rate finder!) developed this idea in his article [\"Super-Convergence: Very Fast Training of Neural Networks Using Large Learning Rates\"](https://arxiv.org/abs/1708.07120). He designed a schedule for learning rate separated into two phases: one where the learning rate grows from the minimum value to the maximum value (*warmup*), and one where it decreases back to the minimum value (*annealing*). Smith called this combination of approaches *1cycle training*.\n", "\n", ">Important: 1cycle training allows us to use a much higher maximum learning rate than other types of training, which gives two benefits:\n", "\n", "- By training with higher learning rates, we train faster—a phenomenon Smith named *super-convergence*.\n", "- By training with higher learning rates, we overfit less because we skip over the sharp local minima to end up in a smoother (and therefore more generalizable) part of the loss.\n", "\n", "> Note:The second point is an interesting and subtle one; it is based on the observation that a model that generalizes well is one whose loss would not change very much if you changed the input by a small amount. If a model trains at a large learning rate for quite a while, and can find a good loss when doing so, it must have found an area that also generalizes well, because it is jumping around a lot from batch to batch (that is basically the definition of a high learning rate). The problem is that, as we have discussed, just jumping to a high learning rate is more likely to result in diverging losses, rather than seeing your losses improve. So we don't jump straight to a high learning rate. Instead, we start at a low learning rate, where our losses do not diverge, and we allow the optimizer to gradually find smoother and smoother areas of our parameters by gradually going to higher and higher learning rates.\n", "Then, once we have found a nice smooth area for our parameters, we want to find the very best part of that area, which means we have to bring our learning rates down again. This is why 1cycle training has a gradual learning rate warmup, and a gradual learning rate cooldown. Many researchers have found that in practice this approach leads to more accurate models and trains more quickly. That is why it is the approach that is used by default for `fine_tune` in fastai.\n", "\n", "In <> we'll learn all about *momentum* in SGD. Briefly, momentum is a technique where the optimizer takes a step not only in the direction of the gradients, but also that continues in the direction of previous steps. Leslie Smith introduced the idea of *cyclical momentums* in [\"A Disciplined Approach to Neural Network Hyper-Parameters: Part 1\"](https://arxiv.org/pdf/1803.09820.pdf). It suggests that the momentum varies in the opposite direction of the learning rate: when we are at high learning rates, we use less momentum, and we use more again in the annealing phase.\n", "\n", "We can use 1cycle training in fastai by calling `fit_one_cycle`:" ] }, { "cell_type": "code", "execution_count": 81, "metadata": {}, "outputs": [], "source": [ "def fit(epochs=1, lr=0.06):\n", " learn = Learner(dls, simple_cnn(), loss_func=F.cross_entropy,\n", " metrics=accuracy, cbs=ActivationStats(with_hist=True))\n", " learn.fit_one_cycle(epochs, lr)\n", " return learn" ] }, { "cell_type": "code", "execution_count": 82, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracytime
00.2070100.0776050.97560000:11
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "learn = fit()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We're finally making some progress! It's giving us a reasonable accuracy now.\n", "\n", "We can view the learning rate and momentum throughout training by calling `plot_sched` on `learn.recorder`. `learn.recorder` (as the name suggests) records everything that happens during training, including losses, metrics, and hyperparameters such as learning rate and momentum:" ] }, { "cell_type": "code", "execution_count": 83, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "learn.recorder.plot_sched()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Smith's original 1cycle paper used a linear warmup and linear annealing. As you can see, we adapted the approach in fastai by combining it with another popular approach: cosine annealing. `fit_one_cycle` provides the following parameters you can adjust:\n", "\n", "- `lr_max`:: The highest learning rate that will be used (this can also be a list of learning rates for each layer group, or a Python `slice` object containing the first and last layer group learning rates)\n", "- `div`:: How much to divide `lr_max` by to get the starting learning rate\n", "- `div_final`:: How much to divide `lr_max` by to get the ending learning rate\n", "- `pct_start`:: What percentage of the batches to use for the warmup\n", "- `moms`:: A tuple `(mom1,mom2,mom3)` where *`mom1`* is the initial momentum, *`mom2`* is the minimum momentum, and *`mom3`* is the final momentum\n", "\n", "Let's take a look at our layer stats again:" ] }, { "cell_type": "code", "execution_count": 84, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "learn.activation_stats.plot_layer_stats(-2)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The percentage of near-zero weights is getting much better, although it's still quite high.\n", "\n", "We can see even more about what's going on in our training using `color_dim`, passing it a layer index:" ] }, { "cell_type": "code", "execution_count": 86, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "learn.activation_stats.color_dim(-2)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "`color_dim` was developed by fast.ai in conjunction with a student, Stefano Giomo. Stefano, who refers to the idea as the *colorful dimension*, provides an [in-depth explanation](https://forums.fast.ai/t/the-colorful-dimension/42908) of the history and details behind the method. The basic idea is to create a histogram of the activations of a layer, which we would hope would follow a smooth pattern such as the normal distribution (colorful_dist)." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"Histogram" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "To create `color_dim`, we take the histogram shown on the left here, and convert it into just the colored representation shown at the bottom. Then we flip it on its side, as shown on the right. We found that the distribution is clearer if we take the log of the histogram values. Then, Stefano describes:\n", "\n", "> : The final plot for each layer is made by stacking the histogram of the activations from each batch along the horizontal axis. So each vertical slice in the visualisation represents the histogram of activations for a single batch. The color intensity corresponds to the height of the histogram, in other words the number of activations in each histogram bin.\n", "\n", "<> shows how this all fits together." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"Summary" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This illustrates why log(f) is more colorful than *f* when *f* follows a normal distribution because taking a log changes the Gaussian in a quadratic, which isn't as narrow." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "So with that in mind, let's take another look at the result for the penultimate layer:" ] }, { "cell_type": "code", "execution_count": 87, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "learn.activation_stats.color_dim(-2)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This shows a classic picture of \"bad training.\" We start with nearly all activations at zero—that's what we see at the far left, with all the dark blue. The bright yellow at the bottom represents the near-zero activations. Then, over the first few batches we see the number of nonzero activations exponentially increasing. But it goes too far, and collapses! We see the dark blue return, and the bottom becomes bright yellow again. It almost looks like training restarts from scratch. Then we see the activations increase again, and collapse again. After repeating this a few times, eventually we see a spread of activations throughout the range.\n", "\n", "It's much better if training can be smooth from the start. The cycles of exponential increase and then collapse tend to result in a lot of near-zero activations, resulting in slow training and poor final results. One way to solve this problem is to use batch normalization." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Batch Normalization" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "To fix the slow training and poor final results we ended up with in the previous section, we need to fix the initial large percentage of near-zero activations, and then try to maintain a good distribution of activations throughout training.\n", "\n", "Sergey Ioffe and Christian Szegedy presented a solution to this problem in the 2015 paper [\"Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift\"](https://arxiv.org/abs/1502.03167). In the abstract, they describe just the problem that we've seen:\n", "\n", "> Important: Training Deep Neural Networks is complicated by the fact that the distribution of each layer's inputs changes during training, as the parameters of the previous layers change. This slows down the training by requiring lower learning rates and careful parameter initialization... We refer to this phenomenon as internal covariate shift, and address the problem by normalizing layer inputs.\n", "\n", "Their solution, they say is:\n", "\n", "> Important: Making normalization a part of the model architecture and performing the normalization for each training mini-batch. Batch Normalization allows us to use much higher learning rates and be less careful about initialization.\n", "\n", "The paper caused great excitement as soon as it was released, because it included the chart in <>, which clearly demonstrated that batch normalization could train a model that was even more accurate than the current state of the art (the *Inception* architecture) and around 5x faster." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\"Impact" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Batch normalization (often just called *batchnorm*) works by taking an average of the mean and standard deviations of the activations of a layer and using those to normalize the activations. However, this can cause problems because the network might want some activations to be really high in order to make accurate predictions. So they also added two learnable parameters (meaning they will be updated in the SGD step), usually called `gamma` and `beta`. After normalizing the activations to get some new activation vector `y`, a batchnorm layer returns `gamma*y + beta`.\n", "\n", "That's why our activations can have any mean or variance, independent from the mean and standard deviation of the results of the previous layer. Those statistics are learned separately, making training easier on our model. The behavior is different during training and validation: during training, we use the mean and standard deviation of the batch to normalize the data, while during validation we instead use a running mean of the statistics calculated during training.\n", "\n", "Let's add a batchnorm layer to `conv`:" ] }, { "cell_type": "code", "execution_count": 88, "metadata": {}, "outputs": [], "source": [ "def conv(ni, nf, ks=3, act=True):\n", " layers = [nn.Conv2d(ni, nf, stride=2, kernel_size=ks, padding=ks//2)]\n", " if act: layers.append(nn.ReLU())\n", " layers.append(nn.BatchNorm2d(nf))\n", " return nn.Sequential(*layers)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "and fit our model:" ] }, { "cell_type": "code", "execution_count": 89, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracytime
00.1399750.0621250.98340000:13
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "learn = fit()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "That's a great result! Let's take a look at `color_dim`:" ] }, { "cell_type": "code", "execution_count": 90, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "learn.activation_stats.color_dim(-4)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This is just what we hope to see: a smooth development of activations, with no \"crashes.\" Batchnorm has really delivered on its promise here! In fact, batchnorm has been so successful that we see it (or something very similar) in nearly all modern neural networks.\n", "\n", "An interesting observation about models containing batch normalization layers is that they tend to generalize better than models that don't contain them. Although we haven't as yet seen a rigorous analysis of what's going on here, most researchers believe that the reason for this is that batch normalization adds some extra randomness to the training process. Each mini-batch will have a somewhat different mean and standard deviation than other mini-batches. Therefore, the activations will be normalized by different values each time. In order for the model to make accurate predictions, it will have to learn to become robust to these variations. In general, adding additional randomization to the training process often helps.\n", "\n", "Since things are going so well, let's train for a few more epochs and see how it goes. In fact, let's *increase* the learning rate, since the abstract of the batchnorm paper claimed we should be able to \"train at much higher learning rates\":" ] }, { "cell_type": "code", "execution_count": 91, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracytime
00.1905940.1219820.96360000:13
10.0823350.0785690.97640000:14
20.0537710.0608120.98050000:13
30.0304230.0297660.98920000:13
40.0163720.0247530.99180000:13
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "learn = fit(5, lr=0.1)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "At this point, I think it's fair to say we know how to recognize digits! It's time to move on to something harder..." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Conclusions" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We've seen that convolutions are just a type of matrix multiplication, with two constraints on the weight matrix: some elements are always zero, and some elements are tied (forced to always have the same value). In <> we saw the eight requirements from the 1986 book *Parallel Distributed Processing*; one of them was \"A pattern of connectivity among units.\" That's exactly what these constraints do: they enforce a certain pattern of connectivity.\n", "\n", "These constraints allow us to use far fewer parameters in our model, without sacrificing the ability to represent complex visual features. That means we can train deeper models faster, with less overfitting. Although the universal approximation theorem shows that it should be *possible* to represent anything in a fully connected network in one hidden layer, we've seen now that in *practice* we can train much better models by being thoughtful about network architecture.\n", "\n", "Convolutions are by far the most common pattern of connectivity we see in neural nets (along with regular linear layers, which we refer to as *fully connected*), but it's likely that many more will be discovered.\n", "\n", "We've also seen how to interpret the activations of layers in the network to see whether training is going well or not, and how batchnorm helps regularize the training and makes it smoother. In the next chapter, we will use both of those layers to build the most popular architecture in computer vision: a residual network." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Questionnaire" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "1. What is a \"feature\"?\n", "1. Write out the convolutional kernel matrix for a top edge detector.\n", "1. Write out the mathematical operation applied by a 3×3 kernel to a single pixel in an image.\n", "1. What is the value of a convolutional kernel apply to a 3×3 matrix of zeros?\n", "1. What is \"padding\"?\n", "1. What is \"stride\"?\n", "1. Create a nested list comprehension to complete any task that you choose.\n", "1. What are the shapes of the `input` and `weight` parameters to PyTorch's 2D convolution?\n", "1. What is a \"channel\"?\n", "1. What is the relationship between a convolution and a matrix multiplication?\n", "1. What is a \"convolutional neural network\"?\n", "1. What is the benefit of refactoring parts of your neural network definition?\n", "1. What is `Flatten`? Where does it need to be included in the MNIST CNN? Why?\n", "1. What does \"NCHW\" mean?\n", "1. Why does the third layer of the MNIST CNN have `7*7*(1168-16)` multiplications?\n", "1. What is a \"receptive field\"?\n", "1. What is the size of the receptive field of an activation after two stride 2 convolutions? Why?\n", "1. Run *conv-example.xlsx* yourself and experiment with *trace precedents*.\n", "1. Have a look at Jeremy or Sylvain's list of recent Twitter \"like\"s, and see if you find any interesting resources or ideas there.\n", "1. How is a color image represented as a tensor?\n", "1. How does a convolution work with a color input?\n", "1. What method can we use to see that data in `DataLoaders`?\n", "1. Why do we double the number of filters after each stride-2 conv?\n", "1. Why do we use a larger kernel in the first conv with MNIST (with `simple_cnn`)?\n", "1. What information does `ActivationStats` save for each layer?\n", "1. How can we access a learner's callback after training?\n", "1. What are the three statistics plotted by `plot_layer_stats`? What does the x-axis represent?\n", "1. Why are activations near zero problematic?\n", "1. What are the upsides and downsides of training with a larger batch size?\n", "1. Why should we avoid using a high learning rate at the start of training?\n", "1. What is 1cycle training?\n", "1. What are the benefits of training with a high learning rate?\n", "1. Why do we want to use a low learning rate at the end of training?\n", "1. What is \"cyclical momentum\"?\n", "1. What callback tracks hyperparameter values during training (along with other information)?\n", "1. What does one column of pixels in the `color_dim` plot represent?\n", "1. What does \"bad training\" look like in `color_dim`? Why?\n", "1. What trainable parameters does a batch normalization layer contain?\n", "1. What statistics are used to normalize in batch normalization during training? How about during validation?\n", "1. Why do models with batch normalization layers generalize better?" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Further Research" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "1. What features other than edge detectors have been used in computer vision (especially before deep learning became popular)?\n", "1. There are other normalization layers available in PyTorch. Try them out and see what works best. Learn about why other normalization layers have been developed, and how they differ from batch normalization.\n", "1. Try moving the activation function after the batch normalization layer in `conv`. Does it make a difference? See what you can find out about what order is recommended, and why." ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "jupytext": { "split_at_heading": true }, "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.8.10" } }, "nbformat": 4, "nbformat_minor": 4 }