Update tokenizers to 0.7.0-rc5 (#3705)

notebooks/01-training-tokenizers.ipynb

@@ -178,7 +178,7 @@
     "from tokenizers.pre_tokenizers import ByteLevel\n",
     "\n",
     "# First we create an empty Byte-Pair Encoding model (i.e. not trained model)\n",
-    "tokenizer = Tokenizer(BPE.empty())\n",
+    "tokenizer = Tokenizer(BPE())\n",
     "\n",
     "# Then we enable lower-casing and unicode-normalization\n",
     "# The Sequence normalizer allows us to combine multiple Normalizer that will be\n",
@@ -307,7 +307,7 @@
    ],
    "source": [
     "# Let's tokenizer a simple input\n",
-    "tokenizer.model = BPE.from_files('vocab.json', 'merges.txt')\n",
+    "tokenizer.model = BPE('vocab.json', 'merges.txt')\n",
     "encoding = tokenizer.encode(\"This is a simple input to be tokenized\")\n",
     "\n",
     "print(\"Encoded string: {}\".format(encoding.tokens))\n",