update solution

Author: dsilver829

LeNet-Lab-Solution.ipynb

@@ -13,7 +13,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### Setup\n",
+    "## Setup\n",
     "The `EPOCH` and `BATCH_SIZE` values affect the training speed and model accuracy.\n",
     "\n",
     "You do not need to modify this section."
@@ -43,7 +43,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Solution: Implement LeNet-5 Moidel\n",
+    "## SOLUTION: Implement LeNet-5\n",
     "Implement the [LeNet-5](http://yann.lecun.com/exdb/lenet/) neural network architecture.\n",
     "\n",
     "This is the only cell you need to edit.\n",
@@ -91,49 +91,46 @@
     "    # Add 2 rows/columns on each side for height and width dimensions.\n",
     "    x = tf.pad(x, [[0, 0], [2, 2], [2, 2], [0, 0]], mode=\"CONSTANT\")\n",
     "    \n",
-    "    \n",
-    "\n",
-    "    \n",
-    "\n",
-    "    \n",
-    "    # TODO: Convolution Layer 1. Input = 32x32x3. Output = 28x28x6.\n",
+    "    # SOLUTION: Convolution Layer 1. Input = 32x32x3. Output = 28x28x6.\n",
     "    conv1_W = tf.Variable(tf.truncated_normal(shape=(5, 5, 1, 6)))\n",
     "    conv1_b = tf.Variable(tf.zeros(6))\n",
-    "    conv1 = tf.nn.conv2d(x, conv1_W, strides=[1, 1, 1, 1], padding='VALID') + conv1_b\n",
+    "    conv1   = tf.nn.conv2d(x, conv1_W, strides=[1, 1, 1, 1], padding='VALID') + conv1_b\n",
     "\n",
-    "    # TODO: Activation 1.\n",
+    "    # SOLUTION: Activation 1.\n",
     "    conv1 = tf.nn.relu(conv1)\n",
     "\n",
-    "    # TODO: Pooling Layer 1. Input = 28x28x6. Output = 14x14x6.\n",
+    "    # SOLUTION: Pooling Layer 1. Input = 28x28x6. Output = 14x14x6.\n",
     "    conv1 = tf.nn.max_pool(conv1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='VALID')\n",
     "\n",
-    "    # TODO: Convolution Layer 2. Output = 10x10x16.\n",
+    "    # SOLUTION: Convolution Layer 2. Output = 10x10x16.\n",
     "    conv2_W = tf.Variable(tf.truncated_normal(shape=(5, 5, 6, 16)))\n",
     "    conv2_b = tf.Variable(tf.zeros(16))\n",
-    "    conv2 = tf.nn.conv2d(conv1, conv2_W, strides=[1, 1, 1, 1], padding='VALID') + conv2_b\n",
+    "    conv2   = tf.nn.conv2d(conv1, conv2_W, strides=[1, 1, 1, 1], padding='VALID') + conv2_b\n",
     "    \n",
-    "    # TODO: Activation 2.\n",
+    "    # SOLUTION: Activation 2.\n",
     "    conv2 = tf.nn.relu(conv2)\n",
     "\n",
-    "    # TODO: Pooling Layer 2. Input = 10x10x16. Output = 5x5x16.\n",
+    "    # SOLUTION: Pooling Layer 2. Input = 10x10x16. Output = 5x5x16.\n",
     "    conv2 = tf.nn.max_pool(conv2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='VALID')\n",
     "\n",
-    "    # TODO: Flatten Layer.\n",
+    "    # SOLUTION: Flatten Layer.\n",
     "    fc1 = flatten(conv2)\n",
     "    \n",
-    "    # TODO: Fully Connected Layer 1. Input = 5x5x16. Output = 120.\n",
+    "    # SOLUTION: Fully Connected Layer 1. Input = 5x5x16. Output = 120.\n",
     "    fc1_shape = (fc1.get_shape().as_list()[-1], 120)\n",
-    "    fc1_W = tf.Variable(tf.truncated_normal(shape=(fc1_shape)))\n",
-    "    fc1_b = tf.Variable(tf.zeros(120))\n",
-    "    fc1 = tf.matmul(fc1, fc1_W) + fc1_b\n",
+    "    fc1_W     = tf.Variable(tf.truncated_normal(shape=(fc1_shape)))\n",
+    "    fc1_b     = tf.Variable(tf.zeros(120))\n",
+    "    fc1       = tf.matmul(fc1, fc1_W) + fc1_b\n",
     "    \n",
-    "    # TODO: Activation 3.\n",
+    "    # SOLUTION: Activation 3.\n",
     "    fc1 = tf.nn.relu(fc1)\n",
     "\n",
-    "    # TODO: Fully Connected Layer 2. Input = 120. Output = 10.\n",
-    "    fc2_W = tf.Variable(tf.truncated_normal(shape=(120, 10)))\n",
-    "    fc2_b = tf.Variable(tf.zeros(10))\n",
-    "    return tf.matmul(fc1, fc2_W) + fc2_b"
+    "    # SOLUTION: Fully Connected Layer 2. Input = 120. Output = 10.\n",
+    "    fc2_W  = tf.Variable(tf.truncated_normal(shape=(120, 10)))\n",
+    "    fc2_b  = tf.Variable(tf.zeros(10))\n",
+    "    logits = tf.matmul(fc1, fc2_W) + fc2_b\n",
+    "    \n",
+    "    return logits"
    ]
   },
   {