![\"Open](\"https://colab.research.google.com/assets/colab-badge.svg\"){kind=icon}
"
766 | ]
767 | },
768 | {
769 | "cell_type": "code",
770 | "metadata": {
771 | "id": "txqNzM1xIPIR",
772 | "colab_type": "code",
773 | "outputId": "efffa915-6104-4a1c-c751-90e516fad837",
774 | "colab": {
775 | "base_uri": "https://localhost:8080/",
776 | "height": 120
777 | }
778 | },
779 | "source": [
780 | "from google.colab import drive\n",
781 | "drive.mount('/content/drive')"
782 | ],
783 | "execution_count": 0,
784 | "outputs": [
785 | {
786 | "output_type": "stream",
787 | "text": [
788 | "Go to this URL in a browser: https://accounts.google.com/o/oauth2/auth?client_id=947318989803-6bn6qk8qdgf4n4g3pfee6491hc0brc4i.apps.googleusercontent.com&redirect_uri=urn%3aietf%3awg%3aoauth%3a2.0%3aoob&response_type=code&scope=email%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdocs.test%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdrive%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdrive.photos.readonly%20https%3a%2f%2fwww.googleapis.com%2fauth%2fpeopleapi.readonly\n",
789 | "\n",
790 | "Enter your authorization code:\n",
791 | "··········\n",
792 | "Mounted at /content/drive\n"
793 | ],
794 | "name": "stdout"
795 | }
796 | ]
797 | },
798 | {
799 | "cell_type": "code",
800 | "metadata": {
801 | "id": "3m6sGbnGeXHk",
802 | "colab_type": "code",
803 | "outputId": "21365dc2-00df-4fcd-8b66-f1aee796f89f",
804 | "colab": {
805 | "base_uri": "https://localhost:8080/",
806 | "height": 351
807 | }
808 | },
809 | "source": [
810 | "!nvidia-smi"
811 | ],
812 | "execution_count": 0,
813 | "outputs": [
814 | {
815 | "output_type": "stream",
816 | "text": [
817 | "Sun Jun 14 03:02:22 2020 \n",
818 | "+-----------------------------------------------------------------------------+\n",
819 | "| NVIDIA-SMI 450.36.06 Driver Version: 418.67 CUDA Version: 10.1 |\n",
820 | "|-------------------------------+----------------------+----------------------+\n",
821 | "| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |\n",
822 | "| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |\n",
823 | "| | | MIG M. |\n",
824 | "|===============================+======================+======================|\n",
825 | "| 0 Tesla P100-PCIE... Off | 00000000:00:04.0 Off | 0 |\n",
826 | "| N/A 34C P0 27W / 250W | 0MiB / 16280MiB | 0% Default |\n",
827 | "| | | ERR! |\n",
828 | "+-------------------------------+----------------------+----------------------+\n",
829 | " \n",
830 | "+-----------------------------------------------------------------------------+\n",
831 | "| Processes: |\n",
832 | "| GPU GI CI PID Type Process name GPU Memory |\n",
833 | "| ID ID Usage |\n",
834 | "|=============================================================================|\n",
835 | "| No running processes found |\n",
836 | "+-----------------------------------------------------------------------------+\n"
837 | ],
838 | "name": "stdout"
839 | }
840 | ]
841 | },
842 | {
843 | "cell_type": "code",
844 | "metadata": {
845 | "id": "HqQKcukJAicL",
846 | "colab_type": "code",
847 | "outputId": "548f009b-9187-47f5-fc78-8275907438b5",
848 | "colab": {
849 | "base_uri": "https://localhost:8080/",
850 | "height": 605
851 | }
852 | },
853 | "source": [
854 | "!pip install transformers"
855 | ],
856 | "execution_count": 0,
857 | "outputs": [
858 | {
859 | "output_type": "stream",
860 | "text": [
861 | "Collecting transformers\n",
862 | "\u001b[?25l Downloading https://files.pythonhosted.org/packages/48/35/ad2c5b1b8f99feaaf9d7cdadaeef261f098c6e1a6a2935d4d07662a6b780/transformers-2.11.0-py3-none-any.whl (674kB)\n",
863 | "\u001b[K |████████████████████████████████| 675kB 2.8MB/s \n",
864 | "\u001b[?25hRequirement already satisfied: tqdm>=4.27 in /usr/local/lib/python3.6/dist-packages (from transformers) (4.41.1)\n",
865 | "Requirement already satisfied: requests in /usr/local/lib/python3.6/dist-packages (from transformers) (2.23.0)\n",
866 | "Collecting sacremoses\n",
867 | "\u001b[?25l Downloading https://files.pythonhosted.org/packages/7d/34/09d19aff26edcc8eb2a01bed8e98f13a1537005d31e95233fd48216eed10/sacremoses-0.0.43.tar.gz (883kB)\n",
868 | "\u001b[K |████████████████████████████████| 890kB 15.5MB/s \n",
869 | "\u001b[?25hCollecting tokenizers==0.7.0\n",
870 | "\u001b[?25l Downloading https://files.pythonhosted.org/packages/14/e5/a26eb4716523808bb0a799fcfdceb6ebf77a18169d9591b2f46a9adb87d9/tokenizers-0.7.0-cp36-cp36m-manylinux1_x86_64.whl (3.8MB)\n",
871 | "\u001b[K |████████████████████████████████| 3.8MB 12.9MB/s \n",
872 | "\u001b[?25hRequirement already satisfied: dataclasses; python_version < \"3.7\" in /usr/local/lib/python3.6/dist-packages (from transformers) (0.7)\n",
873 | "Collecting sentencepiece\n",
874 | "\u001b[?25l Downloading https://files.pythonhosted.org/packages/d4/a4/d0a884c4300004a78cca907a6ff9a5e9fe4f090f5d95ab341c53d28cbc58/sentencepiece-0.1.91-cp36-cp36m-manylinux1_x86_64.whl (1.1MB)\n",
875 | "\u001b[K |████████████████████████████████| 1.1MB 26.3MB/s \n",
876 | "\u001b[?25hRequirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.6/dist-packages (from transformers) (2019.12.20)\n",
877 | "Requirement already satisfied: numpy in /usr/local/lib/python3.6/dist-packages (from transformers) (1.18.5)\n",
878 | "Requirement already satisfied: packaging in /usr/local/lib/python3.6/dist-packages (from transformers) (20.4)\n",
879 | "Requirement already satisfied: filelock in /usr/local/lib/python3.6/dist-packages (from transformers) (3.0.12)\n",
880 | "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.6/dist-packages (from requests->transformers) (2020.4.5.1)\n",
881 | "Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.6/dist-packages (from requests->transformers) (2.9)\n",
882 | "Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.6/dist-packages (from requests->transformers) (1.24.3)\n",
883 | "Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from requests->transformers) (3.0.4)\n",
884 | "Requirement already satisfied: six in /usr/local/lib/python3.6/dist-packages (from sacremoses->transformers) (1.12.0)\n",
885 | "Requirement already satisfied: click in /usr/local/lib/python3.6/dist-packages (from sacremoses->transformers) (7.1.2)\n",
886 | "Requirement already satisfied: joblib in /usr/local/lib/python3.6/dist-packages (from sacremoses->transformers) (0.15.1)\n",
887 | "Requirement already satisfied: pyparsing>=2.0.2 in /usr/local/lib/python3.6/dist-packages (from packaging->transformers) (2.4.7)\n",
888 | "Building wheels for collected packages: sacremoses\n",
889 | " Building wheel for sacremoses (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
890 | " Created wheel for sacremoses: filename=sacremoses-0.0.43-cp36-none-any.whl size=893260 sha256=97710e3620e53f263dc38f0e295f2e16da22030a818999315a002e1088be2cf9\n",
891 | " Stored in directory: /root/.cache/pip/wheels/29/3c/fd/7ce5c3f0666dab31a50123635e6fb5e19ceb42ce38d4e58f45\n",
892 | "Successfully built sacremoses\n",
893 | "Installing collected packages: sacremoses, tokenizers, sentencepiece, transformers\n",
894 | "Successfully installed sacremoses-0.0.43 sentencepiece-0.1.91 tokenizers-0.7.0 transformers-2.11.0\n"
895 | ],
896 | "name": "stdout"
897 | }
898 | ]
899 | },
900 | {
901 | "cell_type": "code",
902 | "metadata": {
903 | "id": "U8FRjwBo15uT",
904 | "colab_type": "code",
905 | "colab": {}
906 | },
907 | "source": [
908 | "!pip install -q -U watermark"
909 | ],
910 | "execution_count": 0,
911 | "outputs": []
912 | },
913 | {
914 | "cell_type": "code",
915 | "metadata": {
916 | "id": "3DNdmQhn2D43",
917 | "colab_type": "code",
918 | "outputId": "ab9b076b-e566-4104-ccb7-95d16d360c85",
919 | "colab": {
920 | "base_uri": "https://localhost:8080/",
921 | "height": 134
922 | }
923 | },
924 | "source": [
925 | "%reload_ext watermark\n",
926 | "%watermark -v -p numpy,pandas,torch,transformers"
927 | ],
928 | "execution_count": 0,
929 | "outputs": [
930 | {
931 | "output_type": "stream",
932 | "text": [
933 | "CPython 3.6.9\n",
934 | "IPython 5.5.0\n",
935 | "\n",
936 | "numpy 1.18.5\n",
937 | "pandas 1.0.4\n",
938 | "torch 1.5.0+cu101\n",
939 | "transformers 2.11.0\n"
940 | ],
941 | "name": "stdout"
942 | }
943 | ]
944 | },
945 | {
946 | "cell_type": "markdown",
947 | "metadata": {
948 | "id": "Mk0GnvUQIdxc",
949 | "colab_type": "text"
950 | },
951 | "source": [
952 | "### Making the necessary imports"
953 | ]
954 | },
955 | {
956 | "cell_type": "code",
957 | "metadata": {
958 | "id": "dOpc8w_12D1f",
959 | "colab_type": "code",
960 | "outputId": "2163b72f-d590-48e3-ab75-d956198150a8",
961 | "colab": {
962 | "base_uri": "https://localhost:8080/",
963 | "height": 87
964 | }
965 | },
966 | "source": [
967 | "import transformers\n",
968 | "from transformers import XLNetTokenizer, XLNetModel, AdamW, get_linear_schedule_with_warmup\n",
969 | "import torch\n",
970 | "\n",
971 | "import numpy as np\n",
972 | "import pandas as pd\n",
973 | "import seaborn as sns\n",
974 | "import matplotlib.pyplot as plt\n",
975 | "from matplotlib import rc\n",
976 | "from sklearn.model_selection import train_test_split\n",
977 | "from sklearn.metrics import confusion_matrix, classification_report, accuracy\n",
978 | "from collections import defaultdict\n",
979 | "from textwrap import wrap\n",
980 | "from pylab import rcParams\n",
981 | "\n",
982 | "from torch import nn, optim\n",
983 | "from keras.preprocessing.sequence import pad_sequences\n",
984 | "from torch.utils.data import TensorDataset,RandomSampler,SequentialSampler\n",
985 | "from torch.utils.data import Dataset, DataLoader\n",
986 | "import torch.nn.functional as F"
987 | ],
988 | "execution_count": 0,
989 | "outputs": [
990 | {
991 | "output_type": "stream",
992 | "text": [
993 | "/usr/local/lib/python3.6/dist-packages/statsmodels/tools/_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n",
994 | " import pandas.util.testing as tm\n",
995 | "Using TensorFlow backend.\n"
996 | ],
997 | "name": "stderr"
998 | }
999 | ]
1000 | },
1001 | {
1002 | "cell_type": "code",
1003 | "metadata": {
1004 | "id": "qAICzZNo2Dyw",
1005 | "colab_type": "code",
1006 | "outputId": "dd0b17dd-a7d5-4f20-9244-38cd860de614",
1007 | "colab": {
1008 | "base_uri": "https://localhost:8080/",
1009 | "height": 33
1010 | }
1011 | },
1012 | "source": [
1013 | "%matplotlib inline\n",
1014 | "%config InlineBackend.figure_format='retina'\n",
1015 | "\n",
1016 | "sns.set(style='whitegrid', palette='muted', font_scale=1.2)\n",
1017 | "\n",
1018 | "HAPPY_COLORS_PALETTE = [\"#01BEFE\", \"#FFDD00\", \"#FF7D00\", \"#FF006D\", \"#ADFF02\", \"#8F00FF\"]\n",
1019 | "\n",
1020 | "sns.set_palette(sns.color_palette(HAPPY_COLORS_PALETTE))\n",
1021 | "\n",
1022 | "rcParams['figure.figsize'] = 12, 8\n",
1023 | "\n",
1024 | "RANDOM_SEED = 42\n",
1025 | "np.random.seed(RANDOM_SEED)\n",
1026 | "torch.manual_seed(RANDOM_SEED)\n",
1027 | "\n",
1028 | "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n",
1029 | "device"
1030 | ],
1031 | "execution_count": 0,
1032 | "outputs": [
1033 | {
1034 | "output_type": "execute_result",
1035 | "data": {
1036 | "text/plain": [
1037 | "device(type='cuda', index=0)"
1038 | ]
1039 | },
1040 | "metadata": {
1041 | "tags": []
1042 | },
1043 | "execution_count": 7
1044 | }
1045 | ]
1046 | },
1047 | {
1048 | "cell_type": "markdown",
1049 | "metadata": {
1050 | "id": "gC-WS1hYIxkr",
1051 | "colab_type": "text"
1052 | },
1053 | "source": [
1054 | "### Data Preprocessing"
1055 | ]
1056 | },
1057 | {
1058 | "cell_type": "code",
1059 | "metadata": {
1060 | "id": "jJ83RaRG2Dv8",
1061 | "colab_type": "code",
1062 | "outputId": "69fd3bb8-b85f-42f7-9185-e26fddbf2d11",
1063 | "colab": {
1064 | "base_uri": "https://localhost:8080/",
1065 | "height": 196
1066 | }
1067 | },
1068 | "source": [
1069 | "df = pd.read_csv('/content/drive/My Drive/NLP/Sentiment Analysis Series/imdb.csv')\n",
1070 | "df.head()"
1071 | ],
1072 | "execution_count": 0,
1073 | "outputs": [
1074 | {
1075 | "output_type": "execute_result",
1076 | "data": {
1077 | "text/html": [
1078 | "| \n", 1096 | " | review | \n", 1097 | "sentiment | \n", 1098 | "
|---|---|---|
| 0 | \n", 1103 | "One of the other reviewers has mentioned that ... | \n", 1104 | "positive | \n", 1105 | "
| 1 | \n", 1108 | "A wonderful little production. <br /><br />The... | \n", 1109 | "positive | \n", 1110 | "
| 2 | \n", 1113 | "I thought this was a wonderful way to spend ti... | \n", 1114 | "positive | \n", 1115 | "
| 3 | \n", 1118 | "Basically there's a family where a little boy ... | \n", 1119 | "negative | \n", 1120 | "
| 4 | \n", 1123 | "Petter Mattei's \"Love in the Time of Money\" is... | \n", 1124 | "positive | \n", 1125 | "
| \n", 1186 | " | review | \n", 1187 | "sentiment | \n", 1188 | "
|---|---|---|
| 33553 | \n", 1193 | "I really liked this Summerslam due to the look... | \n", 1194 | "positive | \n", 1195 | "
| 9427 | \n", 1198 | "Not many television shows appeal to quite as m... | \n", 1199 | "positive | \n", 1200 | "
| 199 | \n", 1203 | "The film quickly gets to a major chase scene w... | \n", 1204 | "negative | \n", 1205 | "
| 12447 | \n", 1208 | "Jane Austen would definitely approve of this o... | \n", 1209 | "positive | \n", 1210 | "
| 39489 | \n", 1213 | "Expectations were somewhat high for me when I ... | \n", 1214 | "negative | \n", 1215 | "
| 42724 | \n", 1218 | "I've watched this movie on a fairly regular ba... | \n", 1219 | "positive | \n", 1220 | "
| 10822 | \n", 1223 | "For once a story of hope highlighted over the ... | \n", 1224 | "positive | \n", 1225 | "
| 49498 | \n", 1228 | "Okay, I didn't get the Purgatory thing the fir... | \n", 1229 | "positive | \n", 1230 | "
| 4144 | \n", 1233 | "I was very disappointed with this series. It h... | \n", 1234 | "negative | \n", 1235 | "
| 36958 | \n", 1238 | "The first 30 minutes of Tinseltown had my fing... | \n", 1239 | "negative | \n", 1240 | "
| 43106 | \n", 1243 | "jeez, this was immensely boring. the leading m... | \n", 1244 | "negative | \n", 1245 | "
| 38695 | \n", 1248 | "Great just great! The West Coast got \"Dirty\" H... | \n", 1249 | "positive | \n", 1250 | "
| 6188 | \n", 1253 | "It's made in 2007 and the CG is bad for a movi... | \n", 1254 | "negative | \n", 1255 | "
| 1414 | \n", 1258 | "This movie stinks majorly. The only reason I g... | \n", 1259 | "negative | \n", 1260 | "
| 18471 | \n", 1263 | "We can start with the wooden acting but this f... | \n", 1264 | "negative | \n", 1265 | "
| 29282 | \n", 1268 | "This movie starts off somewhat slowly and gets... | \n", 1269 | "positive | \n", 1270 | "
| 15177 | \n", 1273 | "This is a slightly uneven entry with one stand... | \n", 1274 | "positive | \n", 1275 | "
| 34304 | \n", 1278 | "I was first introduced to John Waters films by... | \n", 1279 | "positive | \n", 1280 | "
| 12609 | \n", 1283 | "This movie has very good acting by virtually a... | \n", 1284 | "positive | \n", 1285 | "
| 12144 | \n", 1288 | "I can't help but notice the negative reviews t... | \n", 1289 | "positive | \n", 1290 | "