matrix-spam-ml

dataset_analysis.ipynb (37138B)

---

 {
  "cells": [
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "This notebook is mainly to debug the dataset and to see how the data is distributed. It is also used to generate the dataset statistics."
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 1,
    "metadata": {},
    "outputs": [],
    "source": [
     "import csv\n",
     "\n",
     "import numpy as np  # numerical computing\n",
     "import pandas as pd  # data analysis, working with DataFrames\n",
     "import seaborn as sns\n",
     "\n",
     "df = pd.read_csv(\"./input/MatrixData.tsv\", sep='\\t', quoting=csv.QUOTE_NONE, encoding='utf-8')\n",
     "df = df[df['message'].str.split().str.len().gt(18)]\n",
     "df.reset_index(drop=True, inplace=True)"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 2,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
        "<AxesSubplot: xlabel='label', ylabel='count'>"
       ]
      },
      "execution_count": 2,
      "metadata": {},
      "output_type": "execute_result"
     },
     {
      "data": {
       "image/png": "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",
       "text/plain": [
        "<Figure size 640x480 with 1 Axes>"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
     }
    ],
    "source": [
     "sns.countplot(x=df['label']) # countplot for label"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 3,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
        "<AxesSubplot: ylabel='count'>"
       ]
      },
      "execution_count": 3,
      "metadata": {},
      "output_type": "execute_result"
     },
     {
      "data": {
       "image/png": "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",
       "text/plain": [
        "<Figure size 640x480 with 1 Axes>"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
     }
    ],
    "source": [
     "sns.countplot(x=[len(df.loc[i]['message']) for i in range(len(df))])"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 4,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
       "we have 54912 words in our Dataframe\n",
       "the average word count in every sentence is 28\n"
      ]
     },
     {
      "data": {
       "text/plain": [
        "([20, 28, 32, 26, 26], 54912, 28)"
       ]
      },
      "execution_count": 4,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
     "text_words_lengths = [len(df.loc[i]['message'].split()) for i in range(0, len(df))]\n",
     "total_length = np.sum(text_words_lengths)\n",
     "text_words_mean = int(np.mean(text_words_lengths))\n",
     "print('we have ' + str(total_length) + ' words in our Dataframe')\n",
     "print('the average word count in every sentence is ' + str(text_words_mean))\n",
     "text_words_lengths[:5], total_length, text_words_mean"
    ]
   }
  ],
  "metadata": {
   "kernelspec": {
    "display_name": "Python 3.9.15 ('tf')",
    "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": "3.9.15"
   },
   "orig_nbformat": 4,
   "vscode": {
    "interpreter": {
     "hash": "86eece18b6898e5d361741678d0e9a4298e9b9ab2411f93d35b863e6e254e93a"
    }
   }
  },
  "nbformat": 4,
  "nbformat_minor": 2
 }