update learn-gan

Author: demon386

learn-gan.ipynb

@@ -64,7 +64,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 3,
    "metadata": {
     "autoscroll": false,
     "collapsed": false,
@@ -81,7 +81,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 4,
    "metadata": {
     "autoscroll": false,
     "collapsed": false,
@@ -94,7 +94,7 @@
     {
      "data": {
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x1119974e0>"
+       "<matplotlib.figure.Figure at 0x10ace55c0>"
       ]
      },
      "metadata": {},
@@ -158,7 +158,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 137,
+   "execution_count": 5,
    "metadata": {
     "autoscroll": false,
     "collapsed": false,
@@ -181,7 +181,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 138,
+   "execution_count": 6,
    "metadata": {
     "autoscroll": false,
     "collapsed": false,
@@ -193,9 +193,8 @@
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD8CAYAAAB5Pm/hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x115467198>"
+       "<matplotlib.figure.Figure at 0x10ad3fd68>"
       ]
      },
      "metadata": {},
@@ -209,84 +208,6 @@
     "plt.legend();"
    ]
   },
-  {
-   "cell_type": "code",
-   "execution_count": 133,
-   "metadata": {
-    "autoscroll": false,
-    "collapsed": false,
-    "ein.tags": "worksheet-0",
-    "slideshow": {
-     "slide-type": "-"
-    }
-   },
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "0.50449999999999995"
-      ]
-     },
-     "execution_count": 133,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "def naive_predict(X):\n",
-    "    return np.abs(X - 0.5) < 0.2\n",
-    "\n",
-    "all_data = np.concatenate((real_data, forge_data))\n",
-    "all_labels = [1] * 1000 + [0] * 1000\n",
-    "metrics.accuracy_score(all_labels, naive_predict(all_data))"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 134,
-   "metadata": {
-    "autoscroll": false,
-    "collapsed": false,
-    "ein.tags": "worksheet-0",
-    "slideshow": {
-     "slide-type": "-"
-    }
-   },
-   "outputs": [],
-   "source": [
-    "from sklearn.ensemble import RandomForestClassifier\n",
-    "model = RandomForestClassifier()\n",
-    "model.fit(all_data[:, np.newaxis], all_labels)\n",
-    "pred_res = model.predict(all_data[:, np.newaxis])"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 135,
-   "metadata": {
-    "autoscroll": false,
-    "collapsed": false,
-    "ein.tags": "worksheet-0",
-    "slideshow": {
-     "slide-type": "-"
-    }
-   },
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "0.94750000000000001"
-      ]
-     },
-     "execution_count": 135,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "metrics.accuracy_score(all_labels, pred_res)"
-   ]
-  },
   {
    "cell_type": "code",
    "execution_count": null,