re-time imagenet example on k20, instead of my laptop

1 files changed, 6 insertions, 6 deletions

diff --git a/examples/imagenet_pretrained.ipynb b/examples/imagenet_pretrained.ipynb
index 022d48f9..339eb7d4 100644
--- a/examples/imagenet_pretrained.ipynb
+++ b/examples/imagenet_pretrained.ipynb
@@ -109,7 +109,7 @@
        "output_type": "pyout",
        "prompt_number": 4,
        "text": [
-        "[<matplotlib.lines.Line2D at 0x13fb5a890>]"
+        "[<matplotlib.lines.Line2D at 0x8400dd0>]"
        ]
       },
       {
@@ -117,7 +117,7 @@
        "output_type": "display_data",
        "png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEACAYAAAC+gnFaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
        "text": [
-        "<matplotlib.figure.Figure at 0x11e4d2c90>"
+        "<matplotlib.figure.Figure at 0x3cb8350>"
        ]
       }
      ],
@@ -147,7 +147,7 @@
        "output_type": "stream",
        "stream": "stdout",
        "text": [
-        "1 loops, best of 3: 512 ms per loop\n"
+        "1 loops, best of 3: 296 ms per loop\n"
        ]
       }
      ],
@@ -203,7 +203,7 @@
        "output_type": "pyout",
        "prompt_number": 7,
        "text": [
-        "[<matplotlib.lines.Line2D at 0x13fd02110>]"
+        "[<matplotlib.lines.Line2D at 0x8771110>]"
        ]
       },
       {
@@ -211,7 +211,7 @@
        "output_type": "display_data",
        "png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEACAYAAAC+gnFaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
        "text": [
-        "<matplotlib.figure.Figure at 0x13fb66cd0>"
+        "<matplotlib.figure.Figure at 0x30be2d0>"
        ]
       }
      ],
@@ -237,7 +237,7 @@
        "output_type": "stream",
        "stream": "stdout",
        "text": [
-        "1 loops, best of 3: 334 ms per loop\n"
+        "10 loops, best of 3: 123 ms per loop\n"
        ]
       }
      ],