Neural Net. Vincent Van starry night on class

Author: shubham1810

class_20/.ipynb_checkpoints/NeuralArt-checkpoint.ipynb

@@ -28,20 +28,20 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 48,
    "metadata": {
     "collapsed": true
    },
    "outputs": [],
    "source": [
-    "base_image_path = '../class_19/2007_000187.jpg'\n",
-    "style_reference_image_path = 'donelli.jpg'\n",
+    "base_image_path = 'image.jpg'\n",
+    "style_reference_image_path = 'style.jpg'\n",
     "result_prefix = 'im'\n",
     "iterations = 10\n",
     "\n",
     "total_variation_weight = 1.0\n",
-    "style_weight = 1.0\n",
-    "content_weight = 0.025\n",
+    "style_weight = 0.7\n",
+    "content_weight = 0.5\n",
     "\n",
     "width, height = load_img(base_image_path).size\n",
     "img_nrows = 400\n",
@@ -50,7 +50,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 49,
    "metadata": {
     "collapsed": true
    },
@@ -81,7 +81,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 50,
    "metadata": {
     "collapsed": true
    },
@@ -93,7 +93,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 51,
    "metadata": {
     "collapsed": true
    },
@@ -108,7 +108,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 52,
    "metadata": {
     "collapsed": false
    },
@@ -120,7 +120,7 @@
       "_________________________________________________________________\n",
       "Layer (type)                 Output Shape              Param #   \n",
       "=================================================================\n",
-      "input_1 (InputLayer)         (3, 400, 533, 3)          0         \n",
+      "input_4 (InputLayer)         (3, 400, 533, 3)          0         \n",
       "_________________________________________________________________\n",
       "block1_conv1 (Conv2D)        (3, 400, 533, 64)         1792      \n",
       "_________________________________________________________________\n",
@@ -173,7 +173,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 53,
    "metadata": {
     "collapsed": false
    },
@@ -182,25 +182,25 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "block4_pool Tensor(\"block4_pool/MaxPool:0\", shape=(3, 25, 33, 512), dtype=float32)\n",
-      "block1_pool Tensor(\"block1_pool/MaxPool:0\", shape=(3, 200, 266, 64), dtype=float32)\n",
-      "block4_conv1 Tensor(\"block4_conv1/Relu:0\", shape=(3, 50, 66, 512), dtype=float32)\n",
-      "block2_conv1 Tensor(\"block2_conv1/Relu:0\", shape=(3, 200, 266, 128), dtype=float32)\n",
-      "block2_conv2 Tensor(\"block2_conv2/Relu:0\", shape=(3, 200, 266, 128), dtype=float32)\n",
-      "block4_conv2 Tensor(\"block4_conv2/Relu:0\", shape=(3, 50, 66, 512), dtype=float32)\n",
-      "block4_conv3 Tensor(\"block4_conv3/Relu:0\", shape=(3, 50, 66, 512), dtype=float32)\n",
-      "block2_pool Tensor(\"block2_pool/MaxPool:0\", shape=(3, 100, 133, 128), dtype=float32)\n",
-      "block5_conv3 Tensor(\"block5_conv3/Relu:0\", shape=(3, 25, 33, 512), dtype=float32)\n",
-      "block5_conv2 Tensor(\"block5_conv2/Relu:0\", shape=(3, 25, 33, 512), dtype=float32)\n",
-      "block5_conv1 Tensor(\"block5_conv1/Relu:0\", shape=(3, 25, 33, 512), dtype=float32)\n",
-      "block3_pool Tensor(\"block3_pool/MaxPool:0\", shape=(3, 50, 66, 256), dtype=float32)\n",
-      "block1_conv2 Tensor(\"block1_conv2/Relu:0\", shape=(3, 400, 533, 64), dtype=float32)\n",
-      "block1_conv1 Tensor(\"block1_conv1/Relu:0\", shape=(3, 400, 533, 64), dtype=float32)\n",
-      "input_1 Tensor(\"concat:0\", shape=(3, 400, 533, 3), dtype=float32)\n",
-      "block3_conv1 Tensor(\"block3_conv1/Relu:0\", shape=(3, 100, 133, 256), dtype=float32)\n",
-      "block3_conv3 Tensor(\"block3_conv3/Relu:0\", shape=(3, 100, 133, 256), dtype=float32)\n",
-      "block3_conv2 Tensor(\"block3_conv2/Relu:0\", shape=(3, 100, 133, 256), dtype=float32)\n",
-      "block5_pool Tensor(\"block5_pool/MaxPool:0\", shape=(3, 12, 16, 512), dtype=float32)\n"
+      "block4_pool Tensor(\"block4_pool_4/MaxPool:0\", shape=(3, 25, 33, 512), dtype=float32)\n",
+      "block1_pool Tensor(\"block1_pool_4/MaxPool:0\", shape=(3, 200, 266, 64), dtype=float32)\n",
+      "block4_conv1 Tensor(\"block4_conv1_4/Relu:0\", shape=(3, 50, 66, 512), dtype=float32)\n",
+      "block2_conv1 Tensor(\"block2_conv1_4/Relu:0\", shape=(3, 200, 266, 128), dtype=float32)\n",
+      "block2_conv2 Tensor(\"block2_conv2_4/Relu:0\", shape=(3, 200, 266, 128), dtype=float32)\n",
+      "block4_conv2 Tensor(\"block4_conv2_4/Relu:0\", shape=(3, 50, 66, 512), dtype=float32)\n",
+      "block4_conv3 Tensor(\"block4_conv3_4/Relu:0\", shape=(3, 50, 66, 512), dtype=float32)\n",
+      "block5_conv2 Tensor(\"block5_conv2_4/Relu:0\", shape=(3, 25, 33, 512), dtype=float32)\n",
+      "block2_pool Tensor(\"block2_pool_4/MaxPool:0\", shape=(3, 100, 133, 128), dtype=float32)\n",
+      "block5_conv3 Tensor(\"block5_conv3_4/Relu:0\", shape=(3, 25, 33, 512), dtype=float32)\n",
+      "block1_conv1 Tensor(\"block1_conv1_4/Relu:0\", shape=(3, 400, 533, 64), dtype=float32)\n",
+      "block5_conv1 Tensor(\"block5_conv1_4/Relu:0\", shape=(3, 25, 33, 512), dtype=float32)\n",
+      "block3_pool Tensor(\"block3_pool_4/MaxPool:0\", shape=(3, 50, 66, 256), dtype=float32)\n",
+      "block1_conv2 Tensor(\"block1_conv2_4/Relu:0\", shape=(3, 400, 533, 64), dtype=float32)\n",
+      "input_4 Tensor(\"concat_3:0\", shape=(3, 400, 533, 3), dtype=float32)\n",
+      "block3_conv1 Tensor(\"block3_conv1_4/Relu:0\", shape=(3, 100, 133, 256), dtype=float32)\n",
+      "block3_conv3 Tensor(\"block3_conv3_4/Relu:0\", shape=(3, 100, 133, 256), dtype=float32)\n",
+      "block3_conv2 Tensor(\"block3_conv2_4/Relu:0\", shape=(3, 100, 133, 256), dtype=float32)\n",
+      "block5_pool Tensor(\"block5_pool_4/MaxPool:0\", shape=(3, 12, 16, 512), dtype=float32)\n"
      ]
     }
    ],
@@ -212,7 +212,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 54,
    "metadata": {
     "collapsed": true
    },
@@ -247,7 +247,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 55,
    "metadata": {
     "collapsed": true
    },
@@ -263,7 +263,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 56,
    "metadata": {
     "collapsed": true
    },
@@ -276,7 +276,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 57,
    "metadata": {
     "collapsed": true
    },
@@ -294,7 +294,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 58,
    "metadata": {
     "collapsed": true
    },
@@ -313,7 +313,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 59,
    "metadata": {
     "collapsed": true
    },
@@ -335,7 +335,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 60,
    "metadata": {
     "collapsed": true
    },
@@ -366,13 +366,61 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 61,
    "metadata": {
-    "collapsed": true
+    "collapsed": false
    },
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "('Start of iteration', 0)\n",
+      "('Current loss value:', 6.0200903e+10)\n",
+      "('Image saved as', 'results/im_at_iteration_0.png')\n",
+      "Iteration 0 completed in 23s\n",
+      "('Start of iteration', 1)\n",
+      "('Current loss value:', 2.8211651e+10)\n",
+      "('Image saved as', 'results/im_at_iteration_1.png')\n",
+      "Iteration 1 completed in 23s\n",
+      "('Start of iteration', 2)\n",
+      "('Current loss value:', 2.2322983e+10)\n",
+      "('Image saved as', 'results/im_at_iteration_2.png')\n",
+      "Iteration 2 completed in 23s\n",
+      "('Start of iteration', 3)\n",
+      "('Current loss value:', 2.0141793e+10)\n",
+      "('Image saved as', 'results/im_at_iteration_3.png')\n",
+      "Iteration 3 completed in 23s\n",
+      "('Start of iteration', 4)\n",
+      "('Current loss value:', 1.9059913e+10)\n",
+      "('Image saved as', 'results/im_at_iteration_4.png')\n",
+      "Iteration 4 completed in 23s\n",
+      "('Start of iteration', 5)\n",
+      "('Current loss value:', 1.8409126e+10)\n",
+      "('Image saved as', 'results/im_at_iteration_5.png')\n",
+      "Iteration 5 completed in 23s\n",
+      "('Start of iteration', 6)\n",
+      "('Current loss value:', 1.7965795e+10)\n",
+      "('Image saved as', 'results/im_at_iteration_6.png')\n",
+      "Iteration 6 completed in 23s\n",
+      "('Start of iteration', 7)\n",
+      "('Current loss value:', 1.7639055e+10)\n",
+      "('Image saved as', 'results/im_at_iteration_7.png')\n",
+      "Iteration 7 completed in 23s\n",
+      "('Start of iteration', 8)\n",
+      "('Current loss value:', 1.738588e+10)\n",
+      "('Image saved as', 'results/im_at_iteration_8.png')\n",
+      "Iteration 8 completed in 26s\n",
+      "('Start of iteration', 9)\n",
+      "('Current loss value:', 1.7179054e+10)\n",
+      "('Image saved as', 'results/im_at_iteration_9.png')\n",
+      "Iteration 9 completed in 24s\n"
+     ]
+    }
+   ],
    "source": [
-    "x = np.random.uniform(0, 255, (1, img_nrows, img_ncols, 3)) - 128.\n",
+    "# x = np.random.uniform(0, 255, (1, img_nrows, img_ncols, 3)) - 128.\n",
+    "x = np.zeros((1, img_nrows, img_ncols, 3))\n",
     "\n",
     "for i in range(iterations):\n",
     "    print('Start of iteration', i)\n",
@@ -388,6 +436,15 @@
     "    print('Image saved as', fname)\n",
     "    print('Iteration %d completed in %ds' % (i, end_time - start_time))"
    ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": []
   }
  ],
  "metadata": {