schuetoliver
/
handson-ml
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Sync chapter 13 notebook with code samples in that chapter

main
Aurélien Geron 2017-06-05 18:56:44 +02:00
parent e6cd00f9a4
commit 8a57af4efe
1 changed files with 230 additions and 72 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

302
13_convolutional_neural_networks.ipynb
View File

@ -278,25 +278,103 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"import numpy as np\n",
"from sklearn.datasets import load_sample_images\n",
"\n",
"# Load sample images\n",
"china = load_sample_image(\"china.jpg\")\n",
"flower = load_sample_image(\"flower.jpg\")\n",
"dataset = np.array([china, flower], dtype=np.float32)\n", "dataset = np.array([china, flower], dtype=np.float32)\n",
"batch_size, height, width, channels = dataset.shape\n", "batch_size, height, width, channels = dataset.shape\n",
"\n", "\n",
"# Create 2 filters\n",
"filters = np.zeros(shape=(7, 7, channels, 2), dtype=np.float32)\n", "filters = np.zeros(shape=(7, 7, channels, 2), dtype=np.float32)\n",
"filters[:, 3, :, 0] = 1 # vertical line\n", "filters[:, 3, :, 0] = 1 # vertical line\n",
"filters[3, :, :, 1] = 1 # horizontal line\n", "filters[3, :, :, 1] = 1 # horizontal line\n",
"\n", "\n",
"# Create a graph with input X plus a convolutional layer applying the 2 filters\n",
"X = tf.placeholder(tf.float32, shape=(None, height, width, channels))\n", "X = tf.placeholder(tf.float32, shape=(None, height, width, channels))\n",
"convolution = tf.nn.conv2d(X, filters, strides=[1,2,2,1], padding=\"SAME\")\n", "convolution = tf.nn.conv2d(X, filters, strides=[1,2,2,1], padding=\"SAME\")\n",
"\n", "\n",
"with tf.Session() as sess:\n", "with tf.Session() as sess:\n",
" output = sess.run(convolution, feed_dict={X: dataset})\n", " output = sess.run(convolution, feed_dict={X: dataset})\n",
"\n", "\n",
"plt.imshow(output[0, :, :, 1], cmap=\"gray\") # plot 1st image's 2nd feature map\n",
"plt.show()"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"for image_index in (0, 1):\n", "for image_index in (0, 1):\n",
" for feature_map_index in (0, 1):\n", " for feature_map_index in (0, 1):\n",
" plot_image(output[image_index, :, :, feature_map_index])\n", " plot_image(output[image_index, :, :, feature_map_index])\n",
" plt.show()" " plt.show()"
] ]
}, },
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Using `tf.layers.conv2d()`:"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"tf.reset_default_graph()"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"X = tf.placeholder(shape=(None, height, width, channels), dtype=tf.float32)\n",
"conv = tf.layers.conv2d(X, filters=2, kernel_size=7, strides=[2,2],\n",
" padding=\"SAME\")"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"init = tf.global_variables_initializer()\n",
"\n",
"with tf.Session() as sess:\n",
" init.run()\n",
" output = sess.run(conv, feed_dict={X: dataset})"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"plt.imshow(output[0, :, :, 1], cmap=\"gray\") # plot 1st image's 2nd feature map\n",
"plt.show()"
]
},
{ {
"cell_type": "markdown", "cell_type": "markdown",
"metadata": { "metadata": {
@ -309,7 +387,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 12, "execution_count": 17,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -333,7 +411,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 13, "execution_count": 18,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -362,11 +440,9 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 14, "execution_count": 19,
"metadata": { "metadata": {
"collapsed": false, "collapsed": true
"deletable": true,
"editable": true
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
@ -374,14 +450,37 @@
"\n", "\n",
"filters = np.zeros(shape=(7, 7, channels, 2), dtype=np.float32)\n", "filters = np.zeros(shape=(7, 7, channels, 2), dtype=np.float32)\n",
"filters[:, 3, :, 0] = 1 # vertical line\n", "filters[:, 3, :, 0] = 1 # vertical line\n",
"filters[3, :, :, 1] = 1 # horizontal line\n", "filters[3, :, :, 1] = 1 # horizontal line"
"\n", ]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {
"collapsed": false,
"deletable": true,
"editable": true
},
"outputs": [],
"source": [
"X = tf.placeholder(tf.float32, shape=(None, height, width, channels))\n", "X = tf.placeholder(tf.float32, shape=(None, height, width, channels))\n",
"max_pool = tf.nn.max_pool(X, ksize=[1, 2, 2, 1], strides=[1,2,2,1], padding=\"VALID\")\n", "max_pool = tf.nn.max_pool(X, ksize=[1,2,2,1], strides=[1,2,2,1],padding=\"VALID\")\n",
"\n", "\n",
"with tf.Session() as sess:\n", "with tf.Session() as sess:\n",
" output = sess.run(max_pool, feed_dict={X: dataset})\n", " output = sess.run(max_pool, feed_dict={X: dataset})\n",
"\n", "\n",
"plt.imshow(output[0].astype(np.uint8)) # plot the output for the 1st image\n",
"plt.show()"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"plot_color_image(dataset[0])\n", "plot_color_image(dataset[0])\n",
"save_fig(\"china_original\")\n", "save_fig(\"china_original\")\n",
"plt.show()\n", "plt.show()\n",
@ -429,7 +528,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 15, "execution_count": 22,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -495,7 +594,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 16, "execution_count": 23,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -509,7 +608,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 17, "execution_count": 24,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -587,7 +686,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 18, "execution_count": 25,
"metadata": { "metadata": {
"collapsed": true, "collapsed": true,
"deletable": true, "deletable": true,
@ -671,7 +770,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 19, "execution_count": 26,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -695,7 +794,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 20, "execution_count": 27,
"metadata": { "metadata": {
"collapsed": true, "collapsed": true,
"deletable": true, "deletable": true,
@ -728,7 +827,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 21, "execution_count": 28,
"metadata": { "metadata": {
"collapsed": true, "collapsed": true,
"deletable": true, "deletable": true,
@ -758,7 +857,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 22, "execution_count": 29,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -819,7 +918,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 23, "execution_count": 30,
"metadata": { "metadata": {
"collapsed": true, "collapsed": true,
"deletable": true, "deletable": true,
@ -834,7 +933,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 24, "execution_count": 31,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -862,7 +961,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 25, "execution_count": 32,
"metadata": { "metadata": {
"collapsed": true, "collapsed": true,
"deletable": true, "deletable": true,
@ -898,7 +997,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 26, "execution_count": 33,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -911,7 +1010,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 27, "execution_count": 34,
"metadata": { "metadata": {
"collapsed": true, "collapsed": true,
"deletable": true, "deletable": true,
@ -931,7 +1030,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 28, "execution_count": 35,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -944,7 +1043,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 29, "execution_count": 36,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -968,7 +1067,18 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 30, "execution_count": 37,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"tf.reset_default_graph()"
]
},
{
"cell_type": "code",
"execution_count": 38,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -979,11 +1089,10 @@
"from tensorflow.contrib.slim.nets import inception\n", "from tensorflow.contrib.slim.nets import inception\n",
"import tensorflow.contrib.slim as slim\n", "import tensorflow.contrib.slim as slim\n",
"\n", "\n",
"tf.reset_default_graph()\n", "X = tf.placeholder(tf.float32, shape=[None, 299, 299, 3], name=\"X\")\n",
"\n",
"X = tf.placeholder(tf.float32, shape=[None, height, width, channels], name=\"X\")\n",
"with slim.arg_scope(inception.inception_v3_arg_scope()):\n", "with slim.arg_scope(inception.inception_v3_arg_scope()):\n",
" logits, end_points = inception.inception_v3(X, num_classes=1001, is_training=False)\n", " logits, end_points = inception.inception_v3(\n",
" X, num_classes=1001, is_training=False)\n",
"predictions = end_points[\"Predictions\"]\n", "predictions = end_points[\"Predictions\"]\n",
"saver = tf.train.Saver()" "saver = tf.train.Saver()"
] ]
@ -1001,9 +1110,9 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 31, "execution_count": 39,
"metadata": { "metadata": {
"collapsed": true, "collapsed": false,
"deletable": true, "deletable": true,
"editable": true "editable": true
}, },
@ -1027,7 +1136,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 32, "execution_count": 40,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1044,7 +1153,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 33, "execution_count": 41,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1058,7 +1167,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 34, "execution_count": 42,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1071,7 +1180,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 35, "execution_count": 43,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1121,7 +1230,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 36, "execution_count": 44,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1150,7 +1259,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 37, "execution_count": 45,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1173,7 +1282,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 39, "execution_count": 46,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1199,7 +1308,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 40, "execution_count": 47,
"metadata": { "metadata": {
"collapsed": true, "collapsed": true,
"deletable": true, "deletable": true,
@ -1230,7 +1339,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 41, "execution_count": 48,
"metadata": { "metadata": {
"collapsed": true, "collapsed": true,
"deletable": true, "deletable": true,
@ -1254,7 +1363,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 42, "execution_count": 49,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1318,7 +1427,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 43, "execution_count": 50,
"metadata": { "metadata": {
"collapsed": true, "collapsed": true,
"deletable": true, "deletable": true,
@ -1369,7 +1478,10 @@
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
"metadata": {}, "metadata": {
"deletable": true,
"editable": true
},
"source": [ "source": [
"Note: at test time, the preprocessing step should be as light as possible, just the bare minimum necessary to be able to feed the image to the neural network. You may want to tweak the above function to add a `training` parameter: if `False`, preprocessing should be limited to the bare minimum (i.e., no flipping the image, and just the minimum cropping required, preserving the center of the image)." "Note: at test time, the preprocessing step should be as light as possible, just the bare minimum necessary to be able to feed the image to the neural network. You may want to tweak the above function to add a `training` parameter: if `False`, preprocessing should be limited to the bare minimum (i.e., no flipping the image, and just the minimum cropping required, preserving the center of the image)."
] ]
@ -1386,7 +1498,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 44, "execution_count": 51,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1413,7 +1525,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 45, "execution_count": 52,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1442,7 +1554,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 46, "execution_count": 53,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1485,7 +1597,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 47, "execution_count": 54,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1546,7 +1658,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 48, "execution_count": 55,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1602,7 +1714,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 49, "execution_count": 56,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1635,7 +1747,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 50, "execution_count": 57,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1658,7 +1770,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 51, "execution_count": 58,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1681,7 +1793,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 52, "execution_count": 59,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1704,7 +1816,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 53, "execution_count": 60,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1727,7 +1839,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 54, "execution_count": 61,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1750,7 +1862,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 55, "execution_count": 62,
"metadata": { "metadata": {
"collapsed": true, "collapsed": true,
"deletable": true, "deletable": true,
@ -1773,7 +1885,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 56, "execution_count": 63,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1808,7 +1920,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 57, "execution_count": 64,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1836,7 +1948,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 58, "execution_count": 65,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1880,7 +1992,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 59, "execution_count": 66,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1904,7 +2016,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 60, "execution_count": 67,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1930,7 +2042,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 61, "execution_count": 68,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1959,7 +2071,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 62, "execution_count": 69,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -1982,7 +2094,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 63, "execution_count": 70,
"metadata": { "metadata": {
"collapsed": true, "collapsed": true,
"deletable": true, "deletable": true,
@ -2003,7 +2115,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 64, "execution_count": 71,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -2016,7 +2128,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 65, "execution_count": 72,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -2029,7 +2141,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 66, "execution_count": 73,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -2042,7 +2154,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 67, "execution_count": 74,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -2055,7 +2167,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 68, "execution_count": 75,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -2078,7 +2190,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 69, "execution_count": 76,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -2091,7 +2203,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null, "execution_count": 77,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -2124,7 +2236,53 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null, "execution_count": 78,
"metadata": {
"collapsed": false,
"deletable": true,
"editable": true
},
"outputs": [],
"source": [
"X_test, y_test = prepare_batch(flower_paths_and_classes_test, batch_size=len(flower_paths_and_classes_test))"
]
},
{
"cell_type": "code",
"execution_count": 79,
"metadata": {
"collapsed": false,
"deletable": true,
"editable": true
},
"outputs": [],
"source": [
"X_test.shape"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"We could prepare the training set in much the same way, but it would only generate one variant for each image. Instead, it's preferable to generate the training batches on the fly during training, so that we can really benefit from data augmentation, with many variants of each image."
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"And now, we are ready to train the network (or more precisely, the output layer we just added, since all the other layers are frozen). Be aware that this may take a (very) long time."
]
},
{
"cell_type": "code",
"execution_count": 80,
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"deletable": true, "deletable": true,
@ -2188,7 +2346,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null, "execution_count": 81,
"metadata": { "metadata": {
"collapsed": true, "collapsed": true,
"deletable": true, "deletable": true,