TaylorSwift_Lyrics.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "TaylorSwift_Lyrics.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ByemEPRi4hOz"
      },
      "source": [
        "# Lyrics Generation with Deep Learning\r\n",
        "\r\n",
        "In this notebook we will use the dataset: [Taylor Swift Song Lyrics from all the albums](https://www.kaggle.com/PromptCloudHQ/taylor-swift-song-lyrics-from-all-the-albums) from Kaggle.\r\n",
        "\r\n",
        "And will try to generate new lines of lyrics with a few words as a jumpstart.\r\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "U3mZiZHs4e24"
      },
      "source": [
        "# This data is stored in my drive and link has been used here.\r\n",
        "# Feel free to download or use this raw link directly as well\r\n",
        "\r\n",
        "data = 'https://raw.githubusercontent.com/aravind-naidu/Lyrics-Generator/main/taylor_swift_lyrics.csv'"
      ],
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zNKW3aQq6Y5U"
      },
      "source": [
        "# Importing Libraries"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FqHes0Y_5N3W"
      },
      "source": [
        "import pandas as pd\r\n",
        "import numpy as np\r\n",
        "\r\n",
        "from keras.preprocessing.text import Tokenizer\r\n",
        "from keras.preprocessing.sequence import pad_sequences\r\n",
        "import tensorflow.keras.utils as ku \r\n",
        "\r\n",
        "from keras.models import Sequential\r\n",
        "from keras.layers import Embedding, LSTM, Dense, Dropout, Bidirectional\r\n",
        "from keras.regularizers import Regularizer\r\n",
        "\r\n",
        "import matplotlib.pyplot as plt\r\n",
        "import seaborn as sns"
      ],
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "W5U5b0VkALiq"
      },
      "source": [
        "# Importing Dataset\r\n",
        "I faced some encoding errors while importing data without the \"encoding\" type perameter. \r\n",
        "\r\n",
        "And had to use 'unicode_escape'."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "pIhDCMlf5VOv",
        "outputId": "12412abc-e0c9-4feb-a17d-e0a52d3e9bba"
      },
      "source": [
        "df = pd.read_csv(data, encoding= 'unicode_escape')\r\n",
        "df.head()"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>artist</th>\n",
              "      <th>album</th>\n",
              "      <th>track_title</th>\n",
              "      <th>track_n</th>\n",
              "      <th>lyric</th>\n",
              "      <th>line</th>\n",
              "      <th>year</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Taylor Swift</td>\n",
              "      <td>Taylor Swift</td>\n",
              "      <td>Tim McGraw</td>\n",
              "      <td>1</td>\n",
              "      <td>He said the way my blue eyes shined</td>\n",
              "      <td>1</td>\n",
              "      <td>2006</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Taylor Swift</td>\n",
              "      <td>Taylor Swift</td>\n",
              "      <td>Tim McGraw</td>\n",
              "      <td>1</td>\n",
              "      <td>Put those Georgia stars to shame that night</td>\n",
              "      <td>2</td>\n",
              "      <td>2006</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Taylor Swift</td>\n",
              "      <td>Taylor Swift</td>\n",
              "      <td>Tim McGraw</td>\n",
              "      <td>1</td>\n",
              "      <td>I said, \"That's a lie\"</td>\n",
              "      <td>3</td>\n",
              "      <td>2006</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Taylor Swift</td>\n",
              "      <td>Taylor Swift</td>\n",
              "      <td>Tim McGraw</td>\n",
              "      <td>1</td>\n",
              "      <td>Just a boy in a Chevy truck</td>\n",
              "      <td>4</td>\n",
              "      <td>2006</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Taylor Swift</td>\n",
              "      <td>Taylor Swift</td>\n",
              "      <td>Tim McGraw</td>\n",
              "      <td>1</td>\n",
              "      <td>That had a tendency of gettin' stuck</td>\n",
              "      <td>5</td>\n",
              "      <td>2006</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "         artist         album  ... line  year\n",
              "0  Taylor Swift  Taylor Swift  ...    1  2006\n",
              "1  Taylor Swift  Taylor Swift  ...    2  2006\n",
              "2  Taylor Swift  Taylor Swift  ...    3  2006\n",
              "3  Taylor Swift  Taylor Swift  ...    4  2006\n",
              "4  Taylor Swift  Taylor Swift  ...    5  2006\n",
              "\n",
              "[5 rows x 7 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XlQoVizoAkoe"
      },
      "source": [
        "# Knowing the dataset more..."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hIwZLM-y6h69",
        "outputId": "e094586a-efa6-4b93-f161-0af200221b50"
      },
      "source": [
        "df.shape"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(4862, 7)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DpVY-AtJ6kI-",
        "outputId": "f7a5a4ac-8dff-46a9-9e23-fb638d44bc09"
      },
      "source": [
        "df.info()"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 4862 entries, 0 to 4861\n",
            "Data columns (total 7 columns):\n",
            " #   Column       Non-Null Count  Dtype \n",
            "---  ------       --------------  ----- \n",
            " 0   artist       4862 non-null   object\n",
            " 1   album        4862 non-null   object\n",
            " 2   track_title  4862 non-null   object\n",
            " 3   track_n      4862 non-null   int64 \n",
            " 4   lyric        4862 non-null   object\n",
            " 5   line         4862 non-null   int64 \n",
            " 6   year         4862 non-null   int64 \n",
            "dtypes: int64(3), object(4)\n",
            "memory usage: 266.0+ KB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 401
        },
        "id": "hUR9BX1y8KKG",
        "outputId": "aafd42fd-b9e2-4db9-8ff0-c772c48156ba"
      },
      "source": [
        "sns.catplot(x=\"album\",data=df,kind=\"count\")"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<seaborn.axisgrid.FacetGrid at 0x7fde502e5f90>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 29
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 360x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IX7u4Mdg8OgG",
        "outputId": "802e5532-0241-496e-e0fa-3dd85bd3e4e4"
      },
      "source": [
        "df['track_title'].nunique()"
      ],
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "94"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XrVApWea8cg-",
        "outputId": "e7529f4c-46b5-4626-f222-9cd330f1a0d4"
      },
      "source": [
        "df['track_n'].unique()"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([ 1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16, 17,\n",
              "       18, 19])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 401
        },
        "id": "ipTRJjsG8gbu",
        "outputId": "706e8b9a-2ff0-4b26-a9c7-7400e92500d7"
      },
      "source": [
        "sns.catplot(x=\"year\",data=df,kind=\"count\")"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<seaborn.axisgrid.FacetGrid at 0x7fde5024a2d0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 32
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 360x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BCV9A8NZAo8O"
      },
      "source": [
        "# Making a Corpus\r\n",
        "Making use of the column with lyrics and making a corpus of all the lines."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GVKEdRps7ZiO"
      },
      "source": [
        "corpus = []\r\n",
        "for line in df['lyric']:\r\n",
        "  lowered_line = line.lower()\r\n",
        "  corpus.append(lowered_line)"
      ],
      "execution_count": 33,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uCiFCbr17uMl",
        "outputId": "5161328f-efff-4275-b4ac-89dcbbc21a66"
      },
      "source": [
        "corpus[:20]"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['he said the way my blue eyes shined',\n",
              " 'put those georgia stars to shame that night',\n",
              " 'i said, \"that\\'s a lie\"',\n",
              " 'just a boy in a chevy truck',\n",
              " \"that had a tendency of gettin' stuck\",\n",
              " 'on backroads at night',\n",
              " 'and i was right there beside him all summer long',\n",
              " 'and then the time we woke up to find that summer gone',\n",
              " 'but when you think tim mcgraw',\n",
              " 'i hope you think my favorite song',\n",
              " 'the one we danced to all night long',\n",
              " 'the moon like a spotlight on the lake',\n",
              " 'when you think happiness',\n",
              " 'i hope you think that little black dress',\n",
              " 'think of my head on your chest',\n",
              " 'and my old faded blue jeans',\n",
              " 'when you think tim mcgraw',\n",
              " 'i hope you think of me',\n",
              " 'september saw a month of tears',\n",
              " \"and thankin' god that you weren't here\"]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 34
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XFQmKaI19lPd"
      },
      "source": [
        "# Data Cleaning"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TWugMFG6ofq6"
      },
      "source": [
        "tok = Tokenizer()\r\n",
        "\r\n",
        "tok.fit_on_texts(corpus)\r\n",
        "total_words = len(tok.word_index) + 1"
      ],
      "execution_count": 35,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MYCl5eidogMz"
      },
      "source": [
        "seq = []\r\n",
        "for line in corpus:\r\n",
        "    token_list = tok.texts_to_sequences([line])[0]\r\n",
        "    for i in range(1, len(token_list)):\r\n",
        "        n_gram_sequence = token_list[:i+1]\r\n",
        "        seq.append(n_gram_sequence)"
      ],
      "execution_count": 36,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ESWD44mzolAj"
      },
      "source": [
        "max_sequence_len = max([len(x) for x in seq])\r\n",
        "seq = np.array(pad_sequences(seq, maxlen=max_sequence_len, padding='pre'))"
      ],
      "execution_count": 37,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "28Zd_dOqomHz"
      },
      "source": [
        "predictors, label = seq[:,:-1],seq[:,-1]"
      ],
      "execution_count": 38,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "h_j9MumXonJP"
      },
      "source": [
        "label = ku.to_categorical(label, num_classes=total_words)"
      ],
      "execution_count": 39,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eaHanc9C_BR1",
        "outputId": "09435ad4-62fd-4f6c-84ed-70765c1a85ff"
      },
      "source": [
        "print(\"Size of Input Sequences\", seq.shape[1])"
      ],
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Size of Input Sequences 18\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YeEbdS69A6Y2"
      },
      "source": [
        "# Making a Model\r\n",
        "We are using a Embedding layer with Bidirectional LSTM and LSTM.\r\n",
        "\r\n",
        "We also add, Dropout layer and output layers. And use Adam Optimizer."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "M1TrVzE0_cfN",
        "outputId": "3e83d649-8db9-43d6-ca72-26d5255a4170"
      },
      "source": [
        "model = Sequential()\r\n",
        "model.add(Embedding(total_words, 50, input_length=max_sequence_len-1))  \r\n",
        "model.add(Bidirectional(LSTM(150, return_sequences=True)))  \r\n",
        "model.add(Dropout(0.2))  \r\n",
        "model.add(LSTM(100))  \r\n",
        "model.add(Dense(total_words/2, activation='relu'))  \r\n",
        "model.add(Dense(total_words, activation='softmax'))  \r\n",
        "model.compile(loss='categorical_crossentropy', optimizer='adam', metrics='accuracy')  \r\n",
        "print(model.summary())"
      ],
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_1\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "embedding_1 (Embedding)      (None, 17, 50)            120400    \n",
            "_________________________________________________________________\n",
            "bidirectional_1 (Bidirection (None, 17, 300)           241200    \n",
            "_________________________________________________________________\n",
            "dropout_1 (Dropout)          (None, 17, 300)           0         \n",
            "_________________________________________________________________\n",
            "lstm_3 (LSTM)                (None, 100)               160400    \n",
            "_________________________________________________________________\n",
            "dense_2 (Dense)              (None, 1204)              121604    \n",
            "_________________________________________________________________\n",
            "dense_3 (Dense)              (None, 2408)              2901640   \n",
            "=================================================================\n",
            "Total params: 3,545,244\n",
            "Trainable params: 3,545,244\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "None\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "h_-4BJng_oxk",
        "outputId": "33cae38b-5230-403d-ad25-77ca012d4fd5"
      },
      "source": [
        "hist = model.fit(predictors, label, epochs= 100, verbose=1)"
      ],
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/100\n",
            "957/957 [==============================] - 60s 28ms/step - loss: 6.1463 - accuracy: 0.0496\n",
            "Epoch 2/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 5.3330 - accuracy: 0.0702\n",
            "Epoch 3/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 4.9708 - accuracy: 0.0985\n",
            "Epoch 4/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 4.6442 - accuracy: 0.1274\n",
            "Epoch 5/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 4.3698 - accuracy: 0.1558\n",
            "Epoch 6/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 4.1250 - accuracy: 0.1835\n",
            "Epoch 7/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 3.8661 - accuracy: 0.2074\n",
            "Epoch 8/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 3.6563 - accuracy: 0.2328\n",
            "Epoch 9/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 3.4578 - accuracy: 0.2629\n",
            "Epoch 10/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 3.2071 - accuracy: 0.2982\n",
            "Epoch 11/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 2.9996 - accuracy: 0.3296\n",
            "Epoch 12/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 2.8123 - accuracy: 0.3554\n",
            "Epoch 13/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 2.6257 - accuracy: 0.3865\n",
            "Epoch 14/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 2.4272 - accuracy: 0.4213\n",
            "Epoch 15/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 2.2708 - accuracy: 0.4509\n",
            "Epoch 16/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 2.0949 - accuracy: 0.4843\n",
            "Epoch 17/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 1.9532 - accuracy: 0.5088\n",
            "Epoch 18/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 1.8597 - accuracy: 0.5317\n",
            "Epoch 19/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 1.7403 - accuracy: 0.5548\n",
            "Epoch 20/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 1.6299 - accuracy: 0.5782\n",
            "Epoch 21/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 1.5325 - accuracy: 0.6015\n",
            "Epoch 22/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 1.4371 - accuracy: 0.6239\n",
            "Epoch 23/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 1.3776 - accuracy: 0.6337\n",
            "Epoch 24/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 1.2860 - accuracy: 0.6583\n",
            "Epoch 25/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 1.2453 - accuracy: 0.6682\n",
            "Epoch 26/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 1.1833 - accuracy: 0.6812\n",
            "Epoch 27/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 1.1311 - accuracy: 0.6946\n",
            "Epoch 28/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 1.0844 - accuracy: 0.7057\n",
            "Epoch 29/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 1.0359 - accuracy: 0.7173\n",
            "Epoch 30/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 1.0163 - accuracy: 0.7220\n",
            "Epoch 31/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.9690 - accuracy: 0.7342\n",
            "Epoch 32/100\n",
            "957/957 [==============================] - 27s 29ms/step - loss: 0.9449 - accuracy: 0.7392\n",
            "Epoch 33/100\n",
            "957/957 [==============================] - 27s 29ms/step - loss: 0.9454 - accuracy: 0.7386\n",
            "Epoch 34/100\n",
            "957/957 [==============================] - 27s 29ms/step - loss: 0.9084 - accuracy: 0.7469\n",
            "Epoch 35/100\n",
            "957/957 [==============================] - 28s 29ms/step - loss: 0.8752 - accuracy: 0.7567\n",
            "Epoch 36/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.8514 - accuracy: 0.7633\n",
            "Epoch 37/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.8291 - accuracy: 0.7665\n",
            "Epoch 38/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.8175 - accuracy: 0.7700\n",
            "Epoch 39/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.8003 - accuracy: 0.7785\n",
            "Epoch 40/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.8092 - accuracy: 0.7740\n",
            "Epoch 41/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.7889 - accuracy: 0.7789\n",
            "Epoch 42/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.7892 - accuracy: 0.7766\n",
            "Epoch 43/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.7575 - accuracy: 0.7876\n",
            "Epoch 44/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.7618 - accuracy: 0.7860\n",
            "Epoch 45/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.7430 - accuracy: 0.7932\n",
            "Epoch 46/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.7279 - accuracy: 0.7975\n",
            "Epoch 47/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.7444 - accuracy: 0.7879\n",
            "Epoch 48/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.7415 - accuracy: 0.7882\n",
            "Epoch 49/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.7159 - accuracy: 0.7969\n",
            "Epoch 50/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.7164 - accuracy: 0.7939\n",
            "Epoch 51/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.7111 - accuracy: 0.7974\n",
            "Epoch 52/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.7007 - accuracy: 0.7980\n",
            "Epoch 53/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6967 - accuracy: 0.8023\n",
            "Epoch 54/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.7031 - accuracy: 0.8015\n",
            "Epoch 55/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6840 - accuracy: 0.8068\n",
            "Epoch 56/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6876 - accuracy: 0.8044\n",
            "Epoch 57/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6897 - accuracy: 0.8011\n",
            "Epoch 58/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6825 - accuracy: 0.8053\n",
            "Epoch 59/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6740 - accuracy: 0.8063\n",
            "Epoch 60/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6649 - accuracy: 0.8079\n",
            "Epoch 61/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6730 - accuracy: 0.8079\n",
            "Epoch 62/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6535 - accuracy: 0.8105\n",
            "Epoch 63/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6710 - accuracy: 0.8068\n",
            "Epoch 64/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6543 - accuracy: 0.8117\n",
            "Epoch 65/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6777 - accuracy: 0.8054\n",
            "Epoch 66/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6655 - accuracy: 0.8071\n",
            "Epoch 67/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6517 - accuracy: 0.8101\n",
            "Epoch 68/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6606 - accuracy: 0.8114\n",
            "Epoch 69/100\n",
            "957/957 [==============================] - 27s 29ms/step - loss: 0.6462 - accuracy: 0.8100\n",
            "Epoch 70/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6552 - accuracy: 0.8070\n",
            "Epoch 71/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6319 - accuracy: 0.8166\n",
            "Epoch 72/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6467 - accuracy: 0.8100\n",
            "Epoch 73/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6577 - accuracy: 0.8065\n",
            "Epoch 74/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6425 - accuracy: 0.8120\n",
            "Epoch 75/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6218 - accuracy: 0.8192\n",
            "Epoch 76/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6437 - accuracy: 0.8115\n",
            "Epoch 77/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6435 - accuracy: 0.8119\n",
            "Epoch 78/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6272 - accuracy: 0.8186\n",
            "Epoch 79/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6422 - accuracy: 0.8098\n",
            "Epoch 80/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6314 - accuracy: 0.8119\n",
            "Epoch 81/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6421 - accuracy: 0.8132\n",
            "Epoch 82/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6337 - accuracy: 0.8108\n",
            "Epoch 83/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6166 - accuracy: 0.8146\n",
            "Epoch 84/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6217 - accuracy: 0.8161\n",
            "Epoch 85/100\n",
            "957/957 [==============================] - 27s 29ms/step - loss: 0.6248 - accuracy: 0.8170\n",
            "Epoch 86/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6402 - accuracy: 0.8113\n",
            "Epoch 87/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6326 - accuracy: 0.8107\n",
            "Epoch 88/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6304 - accuracy: 0.8114\n",
            "Epoch 89/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6296 - accuracy: 0.8123\n",
            "Epoch 90/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6126 - accuracy: 0.8187\n",
            "Epoch 91/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6195 - accuracy: 0.8178\n",
            "Epoch 92/100\n",
            "957/957 [==============================] - 28s 29ms/step - loss: 0.6342 - accuracy: 0.8117\n",
            "Epoch 93/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6181 - accuracy: 0.8158\n",
            "Epoch 94/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6003 - accuracy: 0.8205\n",
            "Epoch 95/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6279 - accuracy: 0.8137\n",
            "Epoch 96/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6166 - accuracy: 0.8181\n",
            "Epoch 97/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6126 - accuracy: 0.8199\n",
            "Epoch 98/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6102 - accuracy: 0.8191\n",
            "Epoch 99/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.6005 - accuracy: 0.8220\n",
            "Epoch 100/100\n",
            "957/957 [==============================] - 27s 28ms/step - loss: 0.5986 - accuracy: 0.8192\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7YoXdGiqBUV2"
      },
      "source": [
        "# Plotting Results\r\n",
        "Graphs of:\r\n",
        "* Accuracy \r\n",
        "* Loss"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0GAwSFQeDXTR"
      },
      "source": [
        "import matplotlib.pyplot as plt\r\n",
        "\r\n",
        "def plot_graph(history,string):\r\n",
        "    plt.plot(history.history[string])\r\n",
        "    plt.xlabel(\"Epochs\")\r\n",
        "    plt.ylabel(string)\r\n",
        "    plt.show()"
      ],
      "execution_count": 43,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "XNjeu2HYDYrd",
        "outputId": "dad807f0-9433-4a2c-f2ca-2282451cb237"
      },
      "source": [
        "plot_graph(hist,'accuracy')"
      ],
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "htXuExrPDd-d",
        "outputId": "2b2743b2-0162-4c45-9f10-2fde07109817"
      },
      "source": [
        "plot_graph(hist,'loss')"
      ],
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1_kL-dkCBhbO"
      },
      "source": [
        "# Generating Lyrics\r\n",
        "From the trained model we will now generate the lyrics.\r\n",
        "\r\n",
        "We would need to pass a word(s) and the total size of expected prediction in the **gen_lyrics()** function."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "H5Gs-VuRDgih"
      },
      "source": [
        "def gen_lyrics(seed_text, total_words):\r\n",
        "    for _ in range(total_words-1):\r\n",
        "        token_list = tok.texts_to_sequences([seed_text])[0]\r\n",
        "        token_list = pad_sequences([token_list], maxlen=max_sequence_len-1, padding='pre')\r\n",
        "        predicted = np.argmax(model.predict(token_list, verbose=0), axis=-1)\r\n",
        "        output_word = \"\"\r\n",
        "        for word, index in tok.word_index.items():\r\n",
        "            if index == predicted:\r\n",
        "                output_word = word\r\n",
        "                break\r\n",
        "        seed_text += \" \" + output_word\r\n",
        "    print(seed_text)"
      ],
      "execution_count": 46,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NClvzFMCDie9",
        "outputId": "f2babae6-ab9a-42db-b284-a5508376a2b0"
      },
      "source": [
        "gen_lyrics(\"Hello\",10)"
      ],
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Hello to a redhead named abigail covers flying in your\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gzcXtexvEda4",
        "outputId": "bb4f675e-19da-416a-ee0e-7ee697af31c4"
      },
      "source": [
        "gen_lyrics(\"Singing\",10)"
      ],
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Singing long live all the mountains we moved down you\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ddpzig5EEgqs",
        "outputId": "5ea867f4-fb68-45cd-ab90-1b37e9496f37"
      },
      "source": [
        "gen_lyrics(\"Singing Alone\",15)"
      ],
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Singing Alone yeah we can shake shake you cab to go and i did too gonna\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "a3Fd2HKrElLE",
        "outputId": "453b8035-0b88-400e-9084-780ca3cf2a11"
      },
      "source": [
        "gen_lyrics(\"I am going to\",12)"
      ],
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "I am going to see you back at my front door trying to block you\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Vms-3ktgEvnE",
        "outputId": "1278199e-9926-48db-c605-a323e689d11c"
      },
      "source": [
        "gen_lyrics(\"Marry me Juliet\",12)"
      ],
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Marry me Juliet like i don't know you do i don't think you do\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AkCEMLcrFE8V",
        "outputId": "3b14bfbc-3c35-4f2c-eb35-0c3683a05a3b"
      },
      "source": [
        "gen_lyrics(\"He said the\",12)"
      ],
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "He said the way my blue eyes shine from my mind i tried to\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FXAL3M-1FRc9",
        "outputId": "f95b1f57-c8d7-4b0e-934c-6bd161bb2581"
      },
      "source": [
        "gen_lyrics(\"Georgia stars\",12)"
      ],
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Georgia stars to lose your girl at you do like we're jaguars if\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EeLSmLLfCWvP"
      },
      "source": [
        "# Future Vision\r\n",
        "\r\n",
        "This is a very basic approach to start with, we can add more data to increase the prediction quality. Moreover, we can use some different architecture and work more intensively as well.\r\n",
        "\r\n",
        "I did this project as a 'Weekend Skills Brush-up\". I will be trying to push this ahead someday later.\r\n",
        "\r\n",
        "Feel free to reproduce and make it better. Also, do let me know about it.\r\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cp7LlNzsEG2T"
      },
      "source": [
        "# Ways to reach me: \r\n",
        "\r\n",
        "<p align=\"left\">\r\n",
        "<a href=\"www.linkedin.com/in/aravind-naidu\" target=\"blank\"><img align=\"center\" src=\"https://cdn.jsdelivr.net/npm/simple-icons@3.0.1/icons/linkedin.svg\" alt=\"aravindnaidulinkedin\" height=\"40\" width=\"40\" /></a>\r\n",
        "<a href=\"https://www.instagram.com/aravind.naidu/\" target=\"blank\"><img align=\"center\" src=\"https://cdn.jsdelivr.net/npm/simple-icons@3.0.1/icons/instagram.svg\" alt=\"aravindnaiduinstagram\" height=\"40\" width=\"40\" /></a>\r\n",
        "<a href=\"https://medium.com/@aravind.ai\" target=\"blank\"><img align=\"center\" src=\"https://cdn.jsdelivr.net/npm/simple-icons@3.0.1/icons/medium.svg\" alt=\"aravindnaidumedium\" height=\"40\" width=\"40\" /></a>\r\n",
        "<a href=\"https://twitter.com/aravind_ml\" target=\"blank\"><img align=\"center\" src=\"https://cdn.jsdelivr.net/npm/simple-icons@3.0.1/icons/twitter.svg\" alt=\"aravindnaidutwitter\" height=\"40\" width=\"40\" /></a>\r\n",
        "</p>\r\n"
      ]
    }
  ]
}