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

Use np.random.set_seed(42) and tf.set_random_seed(42) to make notebook's output constant

main
Aurélien Geron 2017-06-07 08:59:58 +02:00
parent a331fe6404
commit 045150bd95
1 changed files with 28 additions and 26 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

54
11_deep_learning.ipynb
View File

@ -55,11 +55,13 @@
"\n", "\n",
"# Common imports\n", "# Common imports\n",
"import numpy as np\n", "import numpy as np\n",
"import numpy.random as rnd\n",
"import os\n", "import os\n",
"\n", "\n",
"# to make this notebook's output stable across runs\n", "# to make this notebook's output stable across runs\n",
"rnd.seed(42)\n", "def reset_graph(seed=42):\n",
" tf.reset_default_graph()\n",
" tf.set_random_seed(seed)\n",
" np.random.seed(seed)\n",
"\n", "\n",
"# To plot pretty figures\n", "# To plot pretty figures\n",
"%matplotlib inline\n", "%matplotlib inline\n",
@ -181,7 +183,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"n_inputs = 28 * 28 # MNIST\n", "n_inputs = 28 * 28 # MNIST\n",
"n_hidden1 = 300\n", "n_hidden1 = 300\n",
@ -281,7 +283,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"X = tf.placeholder(tf.float32, shape=(None, n_inputs), name=\"X\")" "X = tf.placeholder(tf.float32, shape=(None, n_inputs), name=\"X\")"
] ]
@ -322,7 +324,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"n_inputs = 28 * 28 # MNIST\n", "n_inputs = 28 * 28 # MNIST\n",
"n_hidden1 = 300\n", "n_hidden1 = 300\n",
@ -540,7 +542,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"X = tf.placeholder(tf.float32, shape=(None, n_inputs), name=\"X\")" "X = tf.placeholder(tf.float32, shape=(None, n_inputs), name=\"X\")"
] ]
@ -596,7 +598,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"import tensorflow as tf\n", "import tensorflow as tf\n",
"\n", "\n",
@ -632,7 +634,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"X = tf.placeholder(tf.float32, shape=(None, n_inputs), name=\"X\")\n", "X = tf.placeholder(tf.float32, shape=(None, n_inputs), name=\"X\")\n",
"training = tf.placeholder_with_default(False, shape=(), name='training')" "training = tf.placeholder_with_default(False, shape=(), name='training')"
@ -693,7 +695,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"batch_norm_momentum = 0.9\n", "batch_norm_momentum = 0.9\n",
"\n", "\n",
@ -886,7 +888,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"n_inputs = 28 * 28 # MNIST\n", "n_inputs = 28 * 28 # MNIST\n",
"n_hidden1 = 300\n", "n_hidden1 = 300\n",
@ -1070,7 +1072,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()" "reset_graph()"
] ]
}, },
{ {
@ -1336,7 +1338,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"n_inputs = 28 * 28 # MNIST\n", "n_inputs = 28 * 28 # MNIST\n",
"n_hidden1 = 300\n", "n_hidden1 = 300\n",
@ -1431,7 +1433,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"n_hidden4 = 20 # new layer\n", "n_hidden4 = 20 # new layer\n",
"n_outputs = 10 # new layer\n", "n_outputs = 10 # new layer\n",
@ -1517,7 +1519,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"n_inputs = 28 * 28 # MNIST\n", "n_inputs = 28 * 28 # MNIST\n",
"n_hidden1 = 300 # reused\n", "n_hidden1 = 300 # reused\n",
@ -1622,7 +1624,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"n_inputs = 2\n", "n_inputs = 2\n",
"n_hidden1 = 3" "n_hidden1 = 3"
@ -1690,7 +1692,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"n_inputs = 2\n", "n_inputs = 2\n",
"n_hidden1 = 3\n", "n_hidden1 = 3\n",
@ -1802,7 +1804,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"n_inputs = 28 * 28 # MNIST\n", "n_inputs = 28 * 28 # MNIST\n",
"n_hidden1 = 300 # reused\n", "n_hidden1 = 300 # reused\n",
@ -1904,7 +1906,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"n_inputs = 28 * 28 # MNIST\n", "n_inputs = 28 * 28 # MNIST\n",
"n_hidden1 = 300 # reused\n", "n_hidden1 = 300 # reused\n",
@ -2026,7 +2028,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"n_inputs = 28 * 28 # MNIST\n", "n_inputs = 28 * 28 # MNIST\n",
"n_hidden1 = 300 # reused\n", "n_hidden1 = 300 # reused\n",
@ -2104,7 +2106,7 @@
" h2_cache_test = sess.run(hidden2, feed_dict={X: mnist.test.images}) # not shown in the book\n", " h2_cache_test = sess.run(hidden2, feed_dict={X: mnist.test.images}) # not shown in the book\n",
"\n", "\n",
" for epoch in range(n_epochs):\n", " for epoch in range(n_epochs):\n",
" shuffled_idx = rnd.permutation(mnist.train.num_examples)\n", " shuffled_idx = np.random.permutation(mnist.train.num_examples)\n",
" hidden2_batches = np.array_split(h2_cache[shuffled_idx], n_batches)\n", " hidden2_batches = np.array_split(h2_cache[shuffled_idx], n_batches)\n",
" y_batches = np.array_split(mnist.train.labels[shuffled_idx], n_batches)\n", " y_batches = np.array_split(mnist.train.labels[shuffled_idx], n_batches)\n",
" for hidden2_batch, y_batch in zip(hidden2_batches, y_batches):\n", " for hidden2_batch, y_batch in zip(hidden2_batches, y_batches):\n",
@ -2265,7 +2267,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"n_inputs = 28 * 28 # MNIST\n", "n_inputs = 28 * 28 # MNIST\n",
"n_hidden1 = 300\n", "n_hidden1 = 300\n",
@ -2390,7 +2392,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"n_inputs = 28 * 28 # MNIST\n", "n_inputs = 28 * 28 # MNIST\n",
"n_hidden1 = 300\n", "n_hidden1 = 300\n",
@ -2518,7 +2520,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"n_inputs = 28 * 28 # MNIST\n", "n_inputs = 28 * 28 # MNIST\n",
"n_hidden1 = 300\n", "n_hidden1 = 300\n",
@ -2694,7 +2696,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"X = tf.placeholder(tf.float32, shape=(None, n_inputs), name=\"X\")\n", "X = tf.placeholder(tf.float32, shape=(None, n_inputs), name=\"X\")\n",
"y = tf.placeholder(tf.int64, shape=(None), name=\"y\")" "y = tf.placeholder(tf.int64, shape=(None), name=\"y\")"
@ -2807,7 +2809,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"n_inputs = 28 * 28\n", "n_inputs = 28 * 28\n",
"n_hidden1 = 300\n", "n_hidden1 = 300\n",
@ -3012,7 +3014,7 @@
}, },
"outputs": [], "outputs": [],
"source": [ "source": [
"tf.reset_default_graph()\n", "reset_graph()\n",
"\n", "\n",
"n_inputs = 28 * 28\n", "n_inputs = 28 * 28\n",
"n_hidden1 = 300\n", "n_hidden1 = 300\n",