Skip to content

Instantly share code, notes, and snippets.

@TaiToTo

TaiToTo/rnn_attention_wp_1.ipynb Secret

Created January 27, 2021 13:27
Download ZIP

Revisions

  1. TaiToTo created this gist Jan 27, 2021.
    129 changes: 129 additions & 0 deletions rnn_attention_wp_1.ipynb
    View file
    Original file line number Diff line number Diff line change
    @@ -0,0 +1,129 @@
    {
    "cells": [
    {
    "cell_type": "code",
    "execution_count": 2,
    "metadata": {},
    "outputs": [
    {
    "data": {
    "text/plain": [
    "((24000, 16), (6000, 16), (24000, 11), (6000, 11))"
    ]
    },
    "execution_count": 2,
    "metadata": {},
    "output_type": "execute_result"
    }
    ],
    "source": [
    "'''\n",
    " My article is not on NLP, so please just suppose that you get the four tensors below. \n",
    " These shapes mean that the max length of the input sentences is 16, \n",
    " and the max length of the target sentneces is 11. \n",
    "'''\n",
    "input_tensor_train.shape, input_tensor_val.shape, target_tensor_train.shape, target_tensor_val.shape"
    ]
    },
    {
    "cell_type": "code",
    "execution_count": 3,
    "metadata": {},
    "outputs": [
    {
    "data": {
    "text/plain": [
    "array([[ 1, 35, 550, ..., 0, 0, 0],\n",
    " [ 1, 17, 12, ..., 0, 0, 0],\n",
    " [ 1, 140, 32, ..., 0, 0, 0],\n",
    " ...,\n",
    " [ 1, 2803, 19, ..., 0, 0, 0],\n",
    " [ 1, 9, 8, ..., 0, 0, 0],\n",
    " [ 1, 8, 18, ..., 0, 0, 0]], dtype=int32)"
    ]
    },
    "execution_count": 3,
    "metadata": {},
    "output_type": "execute_result"
    }
    ],
    "source": [
    "'''\n",
    " Each sentence is implemented like below.\n",
    " Each row denotes a sentence, and each integer denotes a token, in this case a word.\n",
    "'''\n",
    "input_tensor_train"
    ]
    },
    {
    "cell_type": "code",
    "execution_count": 4,
    "metadata": {},
    "outputs": [
    {
    "data": {
    "text/plain": [
    "array([ 1, 35, 550, 3, 2, 0, 0, 0, 0, 0, 0, 0, 0,\n",
    " 0, 0, 0], dtype=int32)"
    ]
    },
    "execution_count": 4,
    "metadata": {},
    "output_type": "execute_result"
    }
    ],
    "source": [
    "'''\n",
    " Let's see the first Spanish sentence of training data. \n",
    " The integer '1' and '2' correspond to \"<start>\" and \"<end>\" respectively. \n",
    "'''\n",
    "input_tensor_train[0]"
    ]
    },
    {
    "cell_type": "code",
    "execution_count": 5,
    "metadata": {},
    "outputs": [
    {
    "name": "stdout",
    "output_type": "stream",
    "text": [
    "1 ----> <start>\n",
    "35 ----> soy\n",
    "550 ----> doctor\n",
    "3 ----> .\n",
    "2 ----> <end>\n"
    ]
    }
    ],
    "source": [
    "'''\n",
    " You can see what each code denotes with convert()\n",
    "'''\n",
    "convert(inp_lang, input_tensor_train[0])"
    ]
    }
    ],
    "metadata": {
    "kernelspec": {
    "display_name": "Python 3",
    "language": "python",
    "name": "python3"
    },
    "language_info": {
    "codemirror_mode": {
    "name": "ipython",
    "version": 3
    },
    "file_extension": ".py",
    "mimetype": "text/x-python",
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
    "version": "3.7.9"
    }
    },
    "nbformat": 4,
    "nbformat_minor": 4
    }