duyguAnalizi.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "history_visible": true,
      "authorship_tag": "ABX9TyOeAj+HfikL0HP5O9WHmEU/",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/rojinakbudak/kurumsal-web-sitesi/blob/main/duyguAnalizi.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "lIPx3f7mfygW"
      },
      "outputs": [],
      "source": [
        "##Downloading the Dataset\n",
        "url=\"https://ai.stanford.edu/~amaas/data/sentiment/aclImdb_v1.tar.gz\"\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import tensorflow as tf"
      ],
      "metadata": {
        "id": "2asDpfY8gh83"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "dataset=tf.keras.utils.get_file(\"aclImdb_v1\" ,\n",
        "                        url ,\n",
        "                        untar=True,\n",
        "                        cache_dir=\".\",\n",
        "                        cache_subdir=\" \")\n",
        "\n",
        "##veri seti indirildi"
      ],
      "metadata": {
        "id": "xno-gLLlg-C1"
      },
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import os\n",
        "\n",
        "dataset_dir=os.path.join(os.path.dirname(dataset),\"aclImdb\")\n",
        "##veri setinin yolunu dataset_dir değişkenine atadık"
      ],
      "metadata": {
        "id": "hMbD_is7hfKa"
      },
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "os.listdir(dataset_dir)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bUbGWY-vi9sS",
        "outputId": "728b901e-9231-40f6-e5ad-1aae153b8798"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['test', 'train', 'imdb.vocab', 'imdbEr.txt', 'README']"
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_dir=os.path.join(dataset_dir,\"train\")\n",
        "os.listdir(train_dir)\n",
        "##train klasöründeki dosyaları getirdi"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9PQwpM8jjbCk",
        "outputId": "d058303c-2d43-484c-cb1f-bb23a008c7c0"
      },
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['labeledBow.feat',\n",
              " 'neg',\n",
              " 'unsup',\n",
              " 'urls_unsup.txt',\n",
              " 'urls_neg.txt',\n",
              " 'unsupBow.feat',\n",
              " 'urls_pos.txt',\n",
              " 'pos']"
            ]
          },
          "metadata": {},
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "sample_file = os.path.join(train_dir, \"pos/0_9.txt\")\n",
        "with open(sample_file) as f:\n",
        "  print(f.read())\n",
        "  #etiketi pozitif olan yorum geldi"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kdE-L3Yej27u",
        "outputId": "633b6b4d-6a87-4463-f6d5-f82f6f708d20"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Bromwell High is a cartoon comedy. It ran at the same time as some other programs about school life, such as \"Teachers\". My 35 years in the teaching profession lead me to believe that Bromwell High's satire is much closer to reality than is \"Teachers\". The scramble to survive financially, the insightful students who can see right through their pathetic teachers' pomp, the pettiness of the whole situation, all remind me of the schools I knew and their students. When I saw the episode in which a student repeatedly tried to burn down the school, I immediately recalled ......... at .......... High. A classic line: INSPECTOR: I'm here to sack one of your teachers. STUDENT: Welcome to Bromwell High. I expect that many adults of my age think that Bromwell High is far fetched. What a pity that it isn't!\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import shutil\n",
        "remove_dir = os.path.join(train_dir, \"unsup\")\n",
        "shutil.rmtree(remove_dir)\n",
        "#unsup dosyası kaldırıldı"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 339
        },
        "id": "WXV4M2_6kJUg",
        "outputId": "d9dfc515-7c49-4c71-9bb4-568123ffa00a"
      },
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "error",
          "ename": "FileNotFoundError",
          "evalue": "[Errno 2] No such file or directory: './ /aclImdb/train/unsup'",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mFileNotFoundError\u001b[0m                         Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-23-9f950b910cb5>\u001b[0m in \u001b[0;36m<cell line: 3>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mshutil\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      2\u001b[0m \u001b[0mremove_dir\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mos\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpath\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mjoin\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrain_dir\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"unsup\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m \u001b[0mshutil\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrmtree\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mremove_dir\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      4\u001b[0m \u001b[0;31m#unsup dosyası kaldırıldı\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/lib/python3.10/shutil.py\u001b[0m in \u001b[0;36mrmtree\u001b[0;34m(path, ignore_errors, onerror)\u001b[0m\n\u001b[1;32m    713\u001b[0m             \u001b[0morig_st\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mos\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlstat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpath\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    714\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 715\u001b[0;31m             \u001b[0monerror\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mos\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlstat\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mpath\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msys\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mexc_info\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    716\u001b[0m             \u001b[0;32mreturn\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    717\u001b[0m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/lib/python3.10/shutil.py\u001b[0m in \u001b[0;36mrmtree\u001b[0;34m(path, ignore_errors, onerror)\u001b[0m\n\u001b[1;32m    711\u001b[0m         \u001b[0;31m# lstat()/open()/fstat() trick.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    712\u001b[0m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 713\u001b[0;31m             \u001b[0morig_st\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mos\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlstat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpath\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    714\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    715\u001b[0m             \u001b[0monerror\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mos\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlstat\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mpath\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msys\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mexc_info\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mFileNotFoundError\u001b[0m: [Errno 2] No such file or directory: './ /aclImdb/train/unsup'"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "batch_size=32\n",
        "seed=42\n",
        "raw_train_ds = tf.keras.utils.text_dataset_from_directory(\n",
        "    train_dir,\n",
        "    batch_size=batch_size,\n",
        "    validation_split=0.2, #verisetinin %20'sinin validasyon verisi olması isteniyor\n",
        "    subset=\"training\",\n",
        "    seed = seed)\n",
        "#eğitim veri seti oluştu"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JBLXJdXvkQfN",
        "outputId": "7b56078d-670c-430c-c2db-e2c96ad659c5"
      },
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found 25000 files belonging to 2 classes.\n",
            "Using 20000 files for training.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "for text_batch, label_batch in raw_train_ds.take(1):\n",
        "  for i in range(3):\n",
        "    print(\"Review\" , text_batch.numpy()[i])\n",
        "    print(\"Label\" , label_batch.numpy()[i])\n",
        "##datasetteki 3 yorumun metni ve etiketi"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RSSlf7OqlBlg",
        "outputId": "6732e101-7e81-4df8-80e1-0f1019b56e35"
      },
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Review b'\"Pandemonium\" is a horror movie spoof that comes off more stupid than funny. Believe me when I tell you, I love comedies. Especially comedy spoofs. \"Airplane\", \"The Naked Gun\" trilogy, \"Blazing Saddles\", \"High Anxiety\", and \"Spaceballs\" are some of my favorite comedies that spoof a particular genre. \"Pandemonium\" is not up there with those films. Most of the scenes in this movie had me sitting there in stunned silence because the movie wasn\\'t all that funny. There are a few laughs in the film, but when you watch a comedy, you expect to laugh a lot more than a few times and that\\'s all this film has going for it. Geez, \"Scream\" had more laughs than this film and that was more of a horror film. How bizarre is that?<br /><br />*1/2 (out of four)'\n",
            "Label 0\n",
            "Review b\"David Mamet is a very interesting and a very un-equal director. His first movie 'House of Games' was the one I liked best, and it set a series of films with characters whose perspective of life changes as they get into complicated situations, and so does the perspective of the viewer.<br /><br />So is 'Homicide' which from the title tries to set the mind of the viewer to the usual crime drama. The principal characters are two cops, one Jewish and one Irish who deal with a racially charged area. The murder of an old Jewish shop owner who proves to be an ancient veteran of the Israeli Independence war triggers the Jewish identity in the mind and heart of the Jewish detective.<br /><br />This is were the flaws of the film are the more obvious. The process of awakening is theatrical and hard to believe, the group of Jewish militants is operatic, and the way the detective eventually walks to the final violent confrontation is pathetic. The end of the film itself is Mamet-like smart, but disappoints from a human emotional perspective.<br /><br />Joe Mantegna and William Macy give strong performances, but the flaws of the story are too evident to be easily compensated.\"\n",
            "Label 0\n",
            "Review b'Great documentary about the lives of NY firefighters during the worst terrorist attack of all time.. That reason alone is why this should be a must see collectors item.. What shocked me was not only the attacks, but the\"High Fat Diet\" and physical appearance of some of these firefighters. I think a lot of Doctors would agree with me that,in the physical shape they were in, some of these firefighters would NOT of made it to the 79th floor carrying over 60 lbs of gear. Having said that i now have a greater respect for firefighters and i realize becoming a firefighter is a life altering job. The French have a history of making great documentary\\'s and that is what this is, a Great Documentary.....'\n",
            "Label 1\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "raw_val_ds = tf.keras.utils.text_dataset_from_directory(\n",
        "    train_dir,\n",
        "    batch_size=batch_size,\n",
        "    validation_split=0.2, #verisetinin %20'sinin validasyon verisi olması isteniyor\n",
        "    subset=\"validation\",\n",
        "    seed = seed)\n",
        "##validasyon verisi oluşturuldu"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PIwbHLAvmQlA",
        "outputId": "18d52d97-a5cd-456d-89df-f112d804b830"
      },
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found 25000 files belonging to 2 classes.\n",
            "Using 5000 files for validation.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_dir=os.path.join(dataset_dir,\"test\")\n",
        "raw_test_ds = tf.keras.utils.text_dataset_from_directory(\n",
        "    test_dir,\n",
        "    batch_size=batch_size,\n",
        "    )\n",
        "##test veriseti oluşturuldu"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NDjA1IiHmZlu",
        "outputId": "0301fbf2-acfe-4c30-fbb2-3c01c026e14f"
      },
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found 25000 files belonging to 2 classes.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import re\n",
        "import string\n",
        "def custom_standardization(input_data):\n",
        "  lowercase = tf.strings.lower(input_data)\n",
        "  stripped_html=tf.strings.regex_replace(lowercase, '<br />', '')\n",
        "  return tf.strings.regex_replace(\n",
        "      stripped_html, '[%s]' % re.escape(string.punctuation),'')\n",
        ""
      ],
      "metadata": {
        "id": "7Bz4WkHqm0A_"
      },
      "execution_count": 49,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.layers.experimental.preprocessing import TextVectorization\n",
        "max_features=10000\n",
        "sequence_length=250\n",
        "vectorize_layer=tf.keras.layers.TextVectorization(\n",
        "    standardize=custom_standardization,\n",
        "    max_tokens = max_features,\n",
        "    output_mode=\"int\",\n",
        "    output_sequence_length=250\n",
        ")\n",
        "#özelleştirdiğimiz TextVectorization katmanı oluşturuldu"
      ],
      "metadata": {
        "id": "jo8gBKKRoCm3"
      },
      "execution_count": 50,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_text = raw_train_ds.map(lambda x, y: x)\n",
        "vectorize_layer.adapt(train_text)"
      ],
      "metadata": {
        "id": "PCt_HRfZpsJq"
      },
      "execution_count": 51,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def vectorize_text(text, label):\n",
        "  text = tf.expand_dims(text, -1)\n",
        "  return vectorize_layer(text), label\n",
        "  #yukarıdaki katmandan çıkan sonuçları görmek için fonksiyonu yazdık"
      ],
      "metadata": {
        "id": "vofCqdeVp3eB"
      },
      "execution_count": 52,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "text_batch, label_batch = next(iter(raw_train_ds))\n",
        "#metinleri ve etiketleri barındıran değişkenleri oluşturduk\n",
        ""
      ],
      "metadata": {
        "id": "gqD6A0e2rNei"
      },
      "execution_count": 53,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "first_review, first_label= text_batch[0], label_batch[0]\n",
        "print(\"Review: \", first_review)\n",
        "print(\"Label: \", raw_train_ds.class_names[first_label])\n",
        "print(\"Processed data: \", vectorize_text(first_review, first_label))\n",
        "#veri setinden bir grup yorumu ve etiketi ilk değerleri alacak şekilde getirdik"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2m1PGeRAwQ6b",
        "outputId": "6a49b95d-8a9b-44a9-a53a-5e8fc0c69289"
      },
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Review:  tf.Tensor(b\"This is really terrible.<br /><br />The only redeeming feature about this movie is that the next time people ask me what is the worst vampire movie I have ever watched, I would have a suitable reply.<br /><br />I think it is filmed on 35 mm so it is already tacky like hell. I wouldn't have bothered commenting but I noticed some fanboys (probably connected to the movie) had claimed that this was the best movie since the Matrix. Let me debunk the myths and lies.<br /><br />There is nothing good in the movie. Everything yells tacky. The actress is ugly. The fight choreography is the worst I have ever seen. The fight scenes are unbelievably amateurish. Imagine a girl flailing her arms around in a circle helplessly and delivering weak kicks which wouldn't hurt a kitten. Obviously, the director just pulled people off the street to give them roles in the movie.<br /><br />I know the director did not have much budget for the movie but still better movies have been made on smaller budget before. Unforgivable.\", shape=(), dtype=string)\n",
            "Label:  neg\n",
            "Processed data:  (<tf.Tensor: shape=(1, 250), dtype=int64, numpy=\n",
            "array([[  10,    7,   62,    1,   61, 1538,  776,   42,   10,   17,    7,\n",
            "          12,    2,  362,   60,   78,  906,   69,   47,    7,    2,  240,\n",
            "        1241,   17,   11,   25,  121,  281,   11,   57,   25,    4, 4341,\n",
            "           1,  101,    9,    7,  817,   20, 5284, 8966,   38,    9,    7,\n",
            "         449, 5087,   37,  547,   11,  545,   25, 2459, 5403,   18,   11,\n",
            "        1978,   46,    1,  234, 3080,    6,    2,   17,   66, 4236,   12,\n",
            "          10,   13,    2,  114,   17,  233,    2, 2598,  368,   69,    1,\n",
            "           2, 9663,    3,    1,    7,  154,   49,    8,    2,   17,  276,\n",
            "        7632, 5087,    2,  509,    7, 1508,    2,  531, 3961,    7,    2,\n",
            "         240,   11,   25,  121,  105,    2,  531,  137,   22, 3512, 2264,\n",
            "         835,    4,  247,    1,   39, 2830,  183,    8,    4, 4285,    1,\n",
            "           3, 4281,  791, 3405,   59,  545, 1462,    4,    1,  519,    2,\n",
            "         172,   40, 1852,   78,  126,    2,  896,    6,  190,   98,  555,\n",
            "           8,    2, 3404,  117,    2,  172,  116,   21,   25,   72,  415,\n",
            "          15,    2,   17,   18,  124,  123,   91,   25,   73,   90,   20,\n",
            "        4114,  415,  152, 9854,    0,    0,    0,    0,    0,    0,    0,\n",
            "           0,    0,    0,    0,    0,    0,    0,    0,    0,    0,    0,\n",
            "           0,    0,    0,    0,    0,    0,    0,    0,    0,    0,    0,\n",
            "           0,    0,    0,    0,    0,    0,    0,    0,    0,    0,    0,\n",
            "           0,    0,    0,    0,    0,    0,    0,    0,    0,    0,    0,\n",
            "           0,    0,    0,    0,    0,    0,    0,    0,    0,    0,    0,\n",
            "           0,    0,    0,    0,    0,    0,    0,    0,    0,    0,    0,\n",
            "           0,    0,    0,    0,    0,    0,    0,    0]])>, <tf.Tensor: shape=(), dtype=int32, numpy=0>)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_ds = raw_train_ds.map(vectorize_text)\n",
        "val_ds = raw_val_ds.map(vectorize_text)\n",
        "test_ds = raw_test_ds.map(vectorize_text)\n",
        "\n",
        "#TextVectorization katmanını daha önce oluşturduğumuz test, validasyon ve eğitim setlerine uyguladık\n",
        "#Böylece validasyon test ve eğitim verileri temizlenmiş oldu"
      ],
      "metadata": {
        "id": "qt5CPILtwcLS"
      },
      "execution_count": 55,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#Girdi ve çıktıların birbirini engellemediğine emin olmak için iki metot kullanılabilir\n",
        "#cache metodu verileri diskten yükledikten sonra bellekte tutar. Böylece eğitim esnasında verisetinin darboğaz haline gelmesini engeller.\n",
        "#veriseti belleğe sığmayacak kadar büyükse verileri küçük dosyalar halinde okunmasını sağlar\n",
        "#prefetch metodu ise eğitim esnasında veri ön işlemeyi ve model uygulamayı beraber kılar\n",
        "\n"
      ],
      "metadata": {
        "id": "Sj7O-g9Wxez_"
      },
      "execution_count": 56,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "AUTOTUNE=tf.data.AUTOTUNE\n",
        "train_ds = train_ds.cache().prefetch(buffer_size=tf.data.AUTOTUNE)\n",
        "val_ds = val_ds.cache().prefetch(buffer_size=tf.data.AUTOTUNE)\n",
        "test_ds = test_ds.cache().prefetch(buffer_size=tf.data.AUTOTUNE)\n",
        "\n",
        "#veri işleme süreci bitti . model kurulabilir"
      ],
      "metadata": {
        "id": "4eK8xsboyiDw"
      },
      "execution_count": 57,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#metin verilerini sinir ağlarında kullanmak için embedding katmanı kullanılır"
      ],
      "metadata": {
        "id": "4UvpEKWWy20G"
      },
      "execution_count": 58,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "embedding_dim=16\n",
        "model = tf.keras.Sequential([\n",
        "    tf.keras.layers.Embedding(max_features+1, 16),\n",
        "    #tamsayı ile kodlanmış metinleri alır ve bu vektörler model eğitilirken kullanılır\n",
        "    tf.keras.layers.Dropout(0.2),\n",
        "    #regülerleştirme için dropout kullanıldı\n",
        "    tf.keras.layers.GlobalAveragePooling1D(),\n",
        "    #bu katman dizi boyutunun ortalamasını alıp her bir örnek için sabit uzunlukta çıktı vektörü döndürür\n",
        "    tf.keras.layers.Dropout(0.2),\n",
        "    tf.keras.layers.Dense(1),\n",
        "])\n"
      ],
      "metadata": {
        "id": "77Ue1ypszEJQ"
      },
      "execution_count": 59,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.summary()\n",
        "#modelin özet bilgisini yazdırdı"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Kj5S4mWdzV-F",
        "outputId": "ae492ed2-901d-4a83-c5a7-4905f85bd093"
      },
      "execution_count": 60,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " embedding (Embedding)       (None, None, 16)          160016    \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, None, 16)          0         \n",
            "                                                                 \n",
            " global_average_pooling1d (  (None, 16)                0         \n",
            " GlobalAveragePooling1D)                                         \n",
            "                                                                 \n",
            " dropout_1 (Dropout)         (None, 16)                0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 1)                 17        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 160033 (625.13 KB)\n",
            "Trainable params: 160033 (625.13 KB)\n",
            "Non-trainable params: 0 (0.00 Byte)\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#modeli kurduktan sonra modeli derliyoruz"
      ],
      "metadata": {
        "id": "qs7ZPGe5zX86"
      },
      "execution_count": 61,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.compile(\n",
        "    loss = tf.keras.losses.BinaryCrossentropy(from_logits = True),\n",
        "    optimizer = \"adam\",\n",
        "    metrics = tf.metrics.BinaryAccuracy(threshold=0.0)\n",
        ")"
      ],
      "metadata": {
        "id": "HzzcxTmz0JOr"
      },
      "execution_count": 62,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#modeli eğitmeye başladık"
      ],
      "metadata": {
        "id": "pEHvzQvI0W7Q"
      },
      "execution_count": 63,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "history = model.fit(\n",
        "    train_ds,\n",
        "    validation_data=val_ds,\n",
        "    epochs=10\n",
        ")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8JZPqgyS0Zvl",
        "outputId": "c9176924-8ef3-4f88-f279-4bdb3b46a86d"
      },
      "execution_count": 64,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "625/625 [==============================] - 10s 14ms/step - loss: 0.6663 - binary_accuracy: 0.6886 - val_loss: 0.6192 - val_binary_accuracy: 0.7718\n",
            "Epoch 2/10\n",
            "625/625 [==============================] - 6s 9ms/step - loss: 0.5531 - binary_accuracy: 0.7983 - val_loss: 0.5029 - val_binary_accuracy: 0.8228\n",
            "Epoch 3/10\n",
            "625/625 [==============================] - 5s 8ms/step - loss: 0.4499 - binary_accuracy: 0.8426 - val_loss: 0.4250 - val_binary_accuracy: 0.8488\n",
            "Epoch 4/10\n",
            "625/625 [==============================] - 6s 10ms/step - loss: 0.3831 - binary_accuracy: 0.8625 - val_loss: 0.3785 - val_binary_accuracy: 0.8602\n",
            "Epoch 5/10\n",
            "625/625 [==============================] - 5s 9ms/step - loss: 0.3399 - binary_accuracy: 0.8764 - val_loss: 0.3498 - val_binary_accuracy: 0.8658\n",
            "Epoch 6/10\n",
            "625/625 [==============================] - 7s 11ms/step - loss: 0.3103 - binary_accuracy: 0.8868 - val_loss: 0.3306 - val_binary_accuracy: 0.8710\n",
            "Epoch 7/10\n",
            "625/625 [==============================] - 5s 7ms/step - loss: 0.2857 - binary_accuracy: 0.8956 - val_loss: 0.3172 - val_binary_accuracy: 0.8728\n",
            "Epoch 8/10\n",
            "625/625 [==============================] - 6s 10ms/step - loss: 0.2670 - binary_accuracy: 0.9022 - val_loss: 0.3080 - val_binary_accuracy: 0.8774\n",
            "Epoch 9/10\n",
            "625/625 [==============================] - 5s 8ms/step - loss: 0.2504 - binary_accuracy: 0.9089 - val_loss: 0.3012 - val_binary_accuracy: 0.8786\n",
            "Epoch 10/10\n",
            "625/625 [==============================] - 6s 9ms/step - loss: 0.2367 - binary_accuracy: 0.9143 - val_loss: 0.2964 - val_binary_accuracy: 0.8800\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [],
      "metadata": {
        "id": "DYpSSGz70rdv"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#her bir epoch'tan(devirden) sonra modelin eğitim verisi üzerinde doğruluğu ve kaybı ekrana yazıldı"
      ],
      "metadata": {
        "id": "gSs2m-6N0jUx"
      },
      "execution_count": 65,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "loss, accuracy = model.evaluate(test_ds)\n",
        "print(\"Loss: \", loss)\n",
        "print(\"Accuracy: \", accuracy)\n",
        "#doğruluğun ve kaybın değerini ekrana yazdırdı"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ql5YRpjI1XOb",
        "outputId": "129a7b7c-5429-4fa0-92ec-0530727f56d6"
      },
      "execution_count": 66,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "782/782 [==============================] - 11s 14ms/step - loss: 0.3144 - binary_accuracy: 0.8711\n",
            "Loss:  0.3143848180770874\n",
            "Accuracy:  0.8710799813270569\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "history_dict = history.history\n",
        "history_dict.keys()\n",
        "#history_dict bbir history nesnesi döndürür. bu nesne eğitim esnasındaki metrikleri içerir"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lw-0bzrw1fR6",
        "outputId": "a43b900f-d3a7-4869-fa25-3220592d24dc"
      },
      "execution_count": 67,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "dict_keys(['loss', 'binary_accuracy', 'val_loss', 'val_binary_accuracy'])"
            ]
          },
          "metadata": {},
          "execution_count": 67
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "acc = history_dict[\"binary_accuracy\"]\n",
        "val_acc = history_dict[\"val_binary_accuracy\"]\n",
        "loss = history_dict[\"loss\"]\n",
        "val_loss = history_dict[\"val_loss\"]"
      ],
      "metadata": {
        "id": "RXvnSV0R1zmK"
      },
      "execution_count": 68,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "epochs = range(1, len(acc)+1)\n",
        "plt.plot(epochs, loss, \"bo\", label=\"Training loss\")\n",
        "plt.plot(epochs, val_loss, \"b\", label=\"Validasyon loss\")\n",
        "plt.title(\"Training and validation loss\")\n",
        "plt.xlabel(\"Epochs\")\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.legend()\n",
        "plt.show()\n",
        "#eğitim ve validasyon için kayıpların grafiği çizildi"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "jCupYtz02KW6",
        "outputId": "eb33ceb7-cee3-4d8c-cbb4-de9419efd64d"
      },
      "execution_count": 73,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(epochs, acc, \"bo\", label=\"Training acc\")\n",
        "plt.plot(epochs, val_acc, \"b\", label=\"Validasyon acc\")\n",
        "plt.title(\"Training and validation accuracy\")\n",
        "plt.xlabel(\"Epochs\")\n",
        "plt.ylabel(\"Accuracy\")\n",
        "plt.legend()\n",
        "plt.show()\n",
        "#eğitim ve validasyon için doğrulukların grafiği çizildi\n",
        "\n",
        "#eğitim verileri üzerinde modelin doğruluğu validasyon verilerinden daha fazla . bu modelin ezberlediği anlamına gelir.\n",
        "#diğer bir ifade ile model eğitim verileni iyi tahmin ederken validasyon verileni tahmin edemiyor\n",
        "#böyle problemlerde genelleştirme problemi olur. bu problemin üstesinden gelmek için erken durdurma veya regülerleştirme teknikleri kullanılır"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "gzwof_SX2Maf",
        "outputId": "fa057caf-a065-4463-c1ee-08d45cb53272"
      },
      "execution_count": 72,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "export_model = tf.keras.Sequential([\n",
        "    vectorize_layer,\n",
        "    model,\n",
        "    tf.keras.layers.Activation(\"sigmoid\")\n",
        "])"
      ],
      "metadata": {
        "id": "BOFduDbs2XOt"
      },
      "execution_count": 74,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "export_model.compile(\n",
        "    loss = tf.keras.losses.BinaryCrossentropy(from_logits = True),\n",
        "    optimizer=\"adam\",\n",
        "    metrics = [\"accuracy\"]\n",
        ")"
      ],
      "metadata": {
        "id": "VWFfsOML3_jj"
      },
      "execution_count": 75,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "loss, accuracy = export_model.evaluate(raw_test_ds)\n",
        "print(accuracy)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "I3GbXTub4A2c",
        "outputId": "3e0fe94b-d482-4877-c876-2e182c449c73"
      },
      "execution_count": 76,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/backend.py:5818: UserWarning: \"`binary_crossentropy` received `from_logits=True`, but the `output` argument was produced by a Sigmoid activation and thus does not represent logits. Was this intended?\n",
            "  output, from_logits = _get_logits(\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "782/782 [==============================] - 5s 6ms/step - loss: 0.3144 - accuracy: 0.8711\n",
            "0.8710799813270569\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "examples = [\n",
        "  \"The movie was perfect\",\n",
        "  \"The movie was okay\",\n",
        "  \"The movie was terrible\"\n",
        "]"
      ],
      "metadata": {
        "id": "Skb6gRKU4Epd"
      },
      "execution_count": 80,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "export_model.predict(examples)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4UcUXSkG4gXh",
        "outputId": "9dc41dea-0195-49f4-ac2e-2d05d2a17d68"
      },
      "execution_count": 81,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 0s 37ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.6178846 ],\n",
              "       [0.43395373],\n",
              "       [0.35908613]], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 81
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "jFVJQFaI4g2Z"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}