Merge pull request #45 from aai-institute/feature/typos

Minor corrections
aai-institute · Dec 5, 2023 · f7dbad1 · f7dbad1
2 parents 8971784 + 2b99c29
commit f7dbad1
Show file tree

Hide file tree

Showing 2 changed files with 10 additions and 10 deletions.
diff --git a/notebooks/nb_01_0_intro_anomaly_detection.ipynb b/notebooks/nb_01_0_intro_anomaly_detection.ipynb
@@ -103,6 +103,9 @@
   {
    "cell_type": "code",
    "execution_count": null,
+   "metadata": {
+    "collapsed": false
+   },
    "outputs": [],
    "source": [
     "import numpy as np\n",
@@ -120,10 +123,7 @@
     "\n",
     "%matplotlib inline\n",
     "matplotlib.rcParams['figure.figsize'] = (5, 5)\n"
-   ],
-   "metadata": {
-    "collapsed": false
-   }
+   ]
   },
   {
    "cell_type": "markdown",
@@ -312,7 +312,7 @@
     "    - $p$ relative frequency of $F_1$\n",
     "- Data set $D \\stackrel{\\text{IID}}{\\sim} F=(1-p)F_0 + pF_1$\n",
     "\n",
-    "**Task:** Estimate if given $x$ is anomalous\n",
+    "**Task:** Estimate if a given $x \\in D$ is anomalous\n",
     "\n",
     "**Assumptions:**\n",
     "- *Few*: $p \\ll 1/2$\n",

diff --git a/notebooks/nb_06_extreme_value_theory_for_anomaly_detection.ipynb b/notebooks/nb_06_extreme_value_theory_for_anomaly_detection.ipynb
@@ -15,6 +15,9 @@
   {
    "cell_type": "code",
    "execution_count": null,
+   "metadata": {
+    "collapsed": false
+   },
    "outputs": [],
    "source": [
     "import tensorflow as tf\n",
@@ -32,10 +35,7 @@
     "from IPython.core.display import HTML\n",
     "\n",
     "tfd = tfp.distributions"
-   ],
-   "metadata": {
-    "collapsed": false
-   }
+   ]
   },
   {
    "cell_type": "markdown",
@@ -1373,7 +1373,7 @@
    "source": [
     "## A look back at the theory\n",
     "\n",
-    "So, what have we really done and why does it make sense to use the GEV for such problems? What kind of guarantees does the Fisher-Tipett-Genedenko theorem give us about the quality of the fit?"
+    "So, what have we really done and why does it make sense to use the GEV for such problems? What kind of guarantees does the Fisher-Tipett-Gnedenko theorem give us about the quality of the fit?"
    ]
   },
   {