Refactor for TensorFlow 1.0 (#69)

* Refactor for TensorFlow 1.0

Tested with TensorFlow r1.0 alpha

Signed-off-by: Norman Heckscher <norman.heckscher@gmail.com>

* Update comments.

Signed-off-by: Norman Heckscher <norman.heckscher@gmail.com>

* close writer

Signed-off-by: Norman Heckscher <norman.heckscher@gmail.com>

* refactor softmax arguments

Signed-off-by: Norman Heckscher <norman.heckscher@gmail.com>

Author: normanheckscher

.gitignore

@@ -57,4 +57,7 @@ target/
 MNIST_data/
 
 # OSX stuff
-.DS_Store
+.DS_Store
+/in_vis.png
+/word2vec.png
+/pred_vis.png

.travis.yml

@@ -11,9 +11,9 @@ install:
   - pip install matplotlib
   # install TensorFlow from https://storage.googleapis.com/tensorflow/
   - if [[ "$TRAVIS_PYTHON_VERSION" == "2.7" ]]; then
-      pip install https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.9.0-cp27-none-linux_x86_64.whl;
+      pip install https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.0.0a0-cp27-none-linux_x86_64.whl;
     elif [[ "$TRAVIS_PYTHON_VERSION" == "3.4" ]]; then
-      pip install https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.9.0-cp34-cp34m-linux_x86_64.whl;
+      pip install https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.0.0a0-cp34-cp34m-linux_x86_64.whl;
     fi
 script:
   - sed -i -- 's/range(100)/range(1)/g' ??_*.py # change range to 1 for quick testing

01_linear_regression.ipynb

@@ -4,7 +4,7 @@
    "cell_type": "code",
    "execution_count": 1,
    "metadata": {
-    "collapsed": true
+    "collapsed": false
    },
    "outputs": [],
    "source": [
@@ -48,7 +48,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 5,
    "metadata": {
     "collapsed": false
    },
@@ -57,22 +57,31 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "2.00683\n"
+      "2.00863\n"
      ]
     }
    ],
    "source": [
     "# Launch the graph in a session\n",
     "with tf.Session() as sess:\n",
     "    # you need to initialize variables (in this case just variable W)\n",
-    "    tf.initialize_all_variables().run()\n",
+    "    tf.global_variables_initializer().run()\n",
     "\n",
     "    for i in range(100):\n",
     "        for (x, y) in zip(trX, trY):\n",
     "            sess.run(train_op, feed_dict={X: x, Y: y})\n",
     "\n",
     "    print(sess.run(w))  # It should be something around 2"
    ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": []
   }
  ],
  "metadata": {
@@ -91,9 +100,9 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython2",
-   "version": "2.7.6"
+   "version": "2.7.13"
   }
  },
  "nbformat": 4,
- "nbformat_minor": 1
+ "nbformat_minor": 0
 }

01_linear_regression.py

@@ -24,7 +24,7 @@ def model(X, w):
 # Launch the graph in a session
 with tf.Session() as sess:
     # you need to initialize variables (in this case just variable W)
-    tf.initialize_all_variables().run()
+    tf.global_variables_initializer().run()
 
     for i in range(100):
         for (x, y) in zip(trX, trY):

02_logistic_regression.ipynb

@@ -15,22 +15,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
    "metadata": {
     "collapsed": false
    },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Extracting MNIST_data/train-images-idx3-ubyte.gz\n",
-      "Extracting MNIST_data/train-labels-idx1-ubyte.gz\n",
-      "Extracting MNIST_data/t10k-images-idx3-ubyte.gz\n",
-      "Extracting MNIST_data/t10k-labels-idx1-ubyte.gz\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "def init_weights(shape):\n",
     "    return tf.Variable(tf.random_normal(shape, stddev=0.01))\n",
@@ -46,7 +35,7 @@
    "cell_type": "code",
    "execution_count": 3,
    "metadata": {
-    "collapsed": true
+    "collapsed": false
    },
    "outputs": [],
    "source": [
@@ -57,156 +46,58 @@
     "\n",
     "py_x = model(X, w)\n",
     "\n",
-    "cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(py_x, Y)) # compute mean cross entropy (softmax is applied internally)\n",
+    "cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=py_x, labels=Y)) # compute mean cross entropy (softmax is applied internally)\n",
     "train_op = tf.train.GradientDescentOptimizer(0.05).minimize(cost) # construct optimizer\n",
     "predict_op = tf.argmax(py_x, 1) # at predict time, evaluate the argmax of the logistic regression"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": null,
    "metadata": {
     "collapsed": false
    },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "0 0.8842\n",
-      "1 0.8969\n",
-      "2 0.9031\n",
-      "3 0.9076\n",
-      "4 0.9102\n",
-      "5 0.9118\n",
-      "6 0.912\n",
-      "7 0.9132\n",
-      "8 0.9145\n",
-      "9 0.9157\n",
-      "10 0.9164\n",
-      "11 0.9165\n",
-      "12 0.917\n",
-      "13 0.9175\n",
-      "14 0.9174\n",
-      "15 0.9178\n",
-      "16 0.9184\n",
-      "17 0.9192\n",
-      "18 0.9193\n",
-      "19 0.9198\n",
-      "20 0.9197\n",
-      "21 0.92\n",
-      "22 0.9203\n",
-      "23 0.9205\n",
-      "24 0.9207\n",
-      "25 0.9211\n",
-      "26 0.9208\n",
-      "27 0.921\n",
-      "28 0.921\n",
-      "29 0.9212\n",
-      "30 0.9212\n",
-      "31 0.9217\n",
-      "32 0.9215\n",
-      "33 0.9213\n",
-      "34 0.9215\n",
-      "35 0.9214\n",
-      "36 0.9213\n",
-      "37 0.9218\n",
-      "38 0.9217\n",
-      "39 0.9215\n",
-      "40 0.9216\n",
-      "41 0.9219\n",
-      "42 0.9222\n",
-      "43 0.9225\n",
-      "44 0.9226\n",
-      "45 0.9225\n",
-      "46 0.9225\n",
-      "47 0.9225\n",
-      "48 0.9225\n",
-      "49 0.9226\n",
-      "50 0.9225\n",
-      "51 0.9226\n",
-      "52 0.9225\n",
-      "53 0.9224\n",
-      "54 0.9223\n",
-      "55 0.9225\n",
-      "56 0.9225\n",
-      "57 0.9225\n",
-      "58 0.9228\n",
-      "59 0.9229\n",
-      "60 0.9229\n",
-      "61 0.923\n",
-      "62 0.9231\n",
-      "63 0.9233\n",
-      "64 0.9234\n",
-      "65 0.9236\n",
-      "66 0.9236\n",
-      "67 0.9236\n",
-      "68 0.9237\n",
-      "69 0.9235\n",
-      "70 0.9235\n",
-      "71 0.9235\n",
-      "72 0.9235\n",
-      "73 0.9235\n",
-      "74 0.9236\n",
-      "75 0.9236\n",
-      "76 0.9235\n",
-      "77 0.9235\n",
-      "78 0.9236\n",
-      "79 0.9238\n",
-      "80 0.9239\n",
-      "81 0.9238\n",
-      "82 0.9238\n",
-      "83 0.9238\n",
-      "84 0.9236\n",
-      "85 0.9236\n",
-      "86 0.9236\n",
-      "87 0.9238\n",
-      "88 0.9238\n",
-      "89 0.9238\n",
-      "90 0.9237\n",
-      "91 0.9236\n",
-      "92 0.9235\n",
-      "93 0.9235\n",
-      "94 0.9237\n",
-      "95 0.9237\n",
-      "96 0.9237\n",
-      "97 0.9237\n",
-      "98 0.9237\n",
-      "99 0.9237\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "# Launch the graph in a session\n",
     "with tf.Session() as sess:\n",
     "    # you need to initialize all variables\n",
-    "    tf.initialize_all_variables().run()\n",
+    "    tf.global_variables_initializer().run()\n",
     "\n",
     "    for i in range(100):\n",
     "        for start, end in zip(range(0, len(trX), 128), range(128, len(trX)+1, 128)):\n",
     "            sess.run(train_op, feed_dict={X: trX[start:end], Y: trY[start:end]})\n",
     "        print(i, np.mean(np.argmax(teY, axis=1) ==\n",
     "                         sess.run(predict_op, feed_dict={X: teX})))"
    ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": []
   }
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 2",
    "language": "python",
-   "name": "python3"
+   "name": "python2"
   },
   "language_info": {
    "codemirror_mode": {
     "name": "ipython",
-    "version": 3
+    "version": 2
    },
    "file_extension": ".py",
    "mimetype": "text/x-python",
    "name": "python",
    "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3",
-   "version": "3.5.2"
+   "pygments_lexer": "ipython2",
+   "version": "2.7.13"
   }
  },
  "nbformat": 4,

02_logistic_regression.py

@@ -23,14 +23,14 @@ def model(X, w):
 
 py_x = model(X, w)
 
-cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(py_x, Y)) # compute mean cross entropy (softmax is applied internally)
+cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=py_x, labels=Y)) # compute mean cross entropy (softmax is applied internally)
 train_op = tf.train.GradientDescentOptimizer(0.05).minimize(cost) # construct optimizer
 predict_op = tf.argmax(py_x, 1) # at predict time, evaluate the argmax of the logistic regression
 
 # Launch the graph in a session
 with tf.Session() as sess:
     # you need to initialize all variables
-    tf.initialize_all_variables().run()
+    tf.global_variables_initializer().run()
 
     for i in range(100):
         for start, end in zip(range(0, len(trX), 128), range(128, len(trX)+1, 128)):