minor fix

EAST_TFLite.ipynb

@@ -179,16 +179,6 @@
         "If we export the float16 model with a fixed known input shape we can can likely accelerate its inference with TFLite GPU delegate. We can specify the `input_shapes` argument in the `tf.compat.v1.lite.TFLiteConverter.from_frozen_graph()` function to do this. We are going to follow this same principle for other quantization (i.e. int8 and dynamic-range) methods as well. "
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "ZiSFtIOBwCXW",
-        "colab_type": "text"
-      },
-      "source": [
-        "For int8 and dynamic-range quantized models we can simply use `tf.compat.v1.lite.TFLiteConverter.from_frozen_graph`. "
-      ]
-    },
     {
       "cell_type": "code",
       "metadata": {