Fixed 'Character level one-hot encoding'. Indexes and characters were the other way around. Need to cut off sample at max_length.

Author: Hiroya Chiba

Diff for: 6.1-one-hot-encoding-of-words-or-characters.ipynb

@@ -109,12 +109,12 @@
     "\n",
     "samples = ['The cat sat on the mat.', 'The dog ate my homework.']\n",
     "characters = string.printable  # All printable ASCII characters.\n",
-    "token_index = dict(zip(range(1, len(characters) + 1), characters))\n",
+    "token_index = dict(zip(characters, range(1, len(characters) + 1)))\n",
     "\n",
     "max_length = 50\n",
-    "results = np.zeros((len(samples), max_length, max(token_index.keys()) + 1))\n",
+    "results = np.zeros((len(samples), max_length, max(token_index.values()) + 1))\n",
     "for i, sample in enumerate(samples):\n",
-    "    for j, character in enumerate(sample):\n",
+    "    for j, character in enumerate(sample[:max_length]):\n",
     "        index = token_index.get(character)\n",
     "        results[i, j, index] = 1."
    ]