Skip to content

Instantly share code, notes, and snippets.

@skilfoy

skilfoy/project5-creating-an-rnn.ipynb

Last active June 11, 2023 21:36
Show Gist options
  • Save skilfoy/9cac271821bd8e30cf654aeef2ac94e4 to your computer and use it in GitHub Desktop.
Save skilfoy/9cac271821bd8e30cf654aeef2ac94e4 to your computer and use it in GitHub Desktop.
Download ZIP
project5-creating-an-rnn.ipynb
Display the source blob
Display the rendered blob
Raw
project5-creating-an-rnn.ipynb
{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"provenance": [],
"machine_shape": "hm",
"gpuType": "A100",
"mount_file_id": "1ct-tr7-5_9Blq-l2JBt8OQ7Kwps4BYCd",
"authorship_tag": "ABX9TyM5JQD7EutsXGDzmtGPyrGf",
"include_colab_link": true
},
"kernelspec": {
"name": "python3",
"display_name": "Python 3"
},
"language_info": {
"name": "python"
},
"accelerator": "GPU"
},
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "view-in-github",
"colab_type": "text"
},
"source": [
"<a href=\"https://colab.research.google.com/gist/skilfoy/9cac271821bd8e30cf654aeef2ac94e4/project5-creating-an-rnn.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
]
},
{
"cell_type": "markdown",
"source": [
"https://gist.github.com/skilfoy/9cac271821bd8e30cf654aeef2ac94e4"
],
"metadata": {
"id": "qzeW8vDfUGo_"
}
},
{
"cell_type": "markdown",
"source": [
"# Project 5 - Creating an RNN\n",
"\n",
"In this notebook, I perform a detailed analysis and modeling of a sample dataset of women's clothing e-commerce reviews. The objective is to predict whether a customer will recommend a product based on their reviews. To achieve this, a Recurrent Neural Network (RNN) is built using TensorFlow, which leverages both GRU (Gated Recurrent Unit) and LSTM (Long Short-Term Memory) layers.\n",
"\n",
"The model achieves an accuracy of 89.27% on the test data. Additionally, it demonstrates a precision of 95.10%, a recall of 91.61%, and an area under the receiver operating characteristic curve (AUC) of 0.9396. These performance metrics suggest that the model is quite successful in predicting whether a clothing item would be recommended based on the reviews. The confusion matrix reveals that the model performed better in identifying positive reviews than negative ones."
],
"metadata": {
"id": "Gxf0ctkIl3bb"
}
},
{
"cell_type": "markdown",
"source": [
"Import libraries:"
],
"metadata": {
"id": "7IaloIL7l6TV"
}
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "03o5jbwaevNm"
},
"outputs": [],
"source": [
"import os\n",
"import pandas as pd\n",
"import re\n",
"import random\n",
"import numpy as np\n",
"import matplotlib.pyplot as plt\n",
"import tensorflow as tf\n",
"from tensorflow.keras.preprocessing.text import Tokenizer\n",
"from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
"from tensorflow.keras.models import Sequential\n",
"from tensorflow.keras.layers import Embedding, LSTM, Dense, Dropout, GRU\n",
"from tensorflow.keras.optimizers import Adam\n",
"from tensorflow.keras.callbacks import EarlyStopping\n",
"from sklearn.model_selection import train_test_split\n",
"from sklearn.metrics import confusion_matrix, classification_report"
]
},
{
"cell_type": "markdown",
"source": [
"Set seeds for reproducibility."
],
"metadata": {
"id": "LpwRK8u7l21j"
}
},
{
"cell_type": "code",
"source": [
"os.environ['PYTHONHASHSEED']='7'\n",
"random.seed(7)\n",
"np.random.seed(7)\n",
"tf.random.set_seed(7)\n",
"tf.keras.utils.set_random_seed(7)"
],
"metadata": {
"id": "ei17a5oImHaI"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"## Data Preparation"
],
"metadata": {
"id": "GMuq1QdIm8l9"
}
},
{
"cell_type": "markdown",
"source": [
"Import data."
],
"metadata": {
"id": "rIeJKIzsm0sQ"
}
},
{
"cell_type": "code",
"source": [
"df = pd.read_csv('/content/drive/MyDrive/Colab Notebooks/DSCI619/Project 5/Womens-Clothing-E-Commerce-Reviews.csv')"
],
"metadata": {
"id": "6enZqvzUmpiY"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"Drop rows with missing values."
],
"metadata": {
"id": "uF04BYTdq3ol"
}
},
{
"cell_type": "code",
"source": [
"df.dropna(inplace=True)"
],
"metadata": {
"id": "a6ynwQHFq5xP"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"Clean up the data by concatenating Title, Review Text, Division Name, Department Name, and Class Name into a new feature called Reviews, and remove non-alphabetic characters with regular expressions."
],
"metadata": {
"id": "RRHtMBKnpKTU"
}
},
{
"cell_type": "code",
"source": [
"df[\"Reviews\"] = df[\"Title\"].astype(str) + \" \" + df[\"Review Text\"].astype(str) + \" \" + df[\"Division Name\"].astype(str) + \" \" + df[\"Department Name\"].astype(str) + \" \" + df[\"Class Name\"].astype(str)\n",
"df[\"Reviews\"] = df[\"Reviews\"].apply(lambda x: re.sub(r\"[^a-zA-Z\\s]\", \"\", x.lower()))"
],
"metadata": {
"id": "FmoV84-FpK0M"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"df.head()"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 843
},
"id": "MdPIDObkp_jY",
"outputId": "5256a785-b260-4bea-d4a3-0658b4f1ff0e"
},
"execution_count": null,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
" Unnamed: 0 Clothing ID Age Title \\\n",
"2 2 1077 60 Some major design flaws \n",
"3 3 1049 50 My favorite buy! \n",
"4 4 847 47 Flattering shirt \n",
"5 5 1080 49 Not for the very petite \n",
"6 6 858 39 Cagrcoal shimmer fun \n",
"\n",
" Review Text Rating Recommended IND \\\n",
"2 I had such high hopes for this dress and reall... 3 0 \n",
"3 I love, love, love this jumpsuit. it's fun, fl... 5 1 \n",
"4 This shirt is very flattering to all due to th... 5 1 \n",
"5 I love tracy reese dresses, but this one is no... 2 0 \n",
"6 I aded this in my basket at hte last mintue to... 5 1 \n",
"\n",
" Positive Feedback Count Division Name Department Name Class Name \\\n",
"2 0 General Dresses Dresses \n",
"3 0 General Petite Bottoms Pants \n",
"4 6 General Tops Blouses \n",
"5 4 General Dresses Dresses \n",
"6 1 General Petite Tops Knits \n",
"\n",
" Reviews \n",
"2 some major design flaws i had such high hopes ... \n",
"3 my favorite buy i love love love this jumpsuit... \n",
"4 flattering shirt this shirt is very flattering... \n",
"5 not for the very petite i love tracy reese dre... \n",
"6 cagrcoal shimmer fun i aded this in my basket ... "
],
"text/html": [
"\n",
" <div id=\"df-74ab786f-b806-4078-b58f-a96c0cd62907\">\n",
" <div class=\"colab-df-container\">\n",
" <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>Unnamed: 0</th>\n",
" <th>Clothing ID</th>\n",
" <th>Age</th>\n",
" <th>Title</th>\n",
" <th>Review Text</th>\n",
" <th>Rating</th>\n",
" <th>Recommended IND</th>\n",
" <th>Positive Feedback Count</th>\n",
" <th>Division Name</th>\n",
" <th>Department Name</th>\n",
" <th>Class Name</th>\n",
" <th>Reviews</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>2</td>\n",
" <td>1077</td>\n",
" <td>60</td>\n",
" <td>Some major design flaws</td>\n",
" <td>I had such high hopes for this dress and reall...</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>General</td>\n",
" <td>Dresses</td>\n",
" <td>Dresses</td>\n",
" <td>some major design flaws i had such high hopes ...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>3</td>\n",
" <td>1049</td>\n",
" <td>50</td>\n",
" <td>My favorite buy!</td>\n",
" <td>I love, love, love this jumpsuit. it's fun, fl...</td>\n",
" <td>5</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>General Petite</td>\n",
" <td>Bottoms</td>\n",
" <td>Pants</td>\n",
" <td>my favorite buy i love love love this jumpsuit...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>4</td>\n",
" <td>847</td>\n",
" <td>47</td>\n",
" <td>Flattering shirt</td>\n",
" <td>This shirt is very flattering to all due to th...</td>\n",
" <td>5</td>\n",
" <td>1</td>\n",
" <td>6</td>\n",
" <td>General</td>\n",
" <td>Tops</td>\n",
" <td>Blouses</td>\n",
" <td>flattering shirt this shirt is very flattering...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5</th>\n",
" <td>5</td>\n",
" <td>1080</td>\n",
" <td>49</td>\n",
" <td>Not for the very petite</td>\n",
" <td>I love tracy reese dresses, but this one is no...</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>4</td>\n",
" <td>General</td>\n",
" <td>Dresses</td>\n",
" <td>Dresses</td>\n",
" <td>not for the very petite i love tracy reese dre...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>6</th>\n",
" <td>6</td>\n",
" <td>858</td>\n",
" <td>39</td>\n",
" <td>Cagrcoal shimmer fun</td>\n",
" <td>I aded this in my basket at hte last mintue to...</td>\n",
" <td>5</td>\n",
" <td>1</td>\n",
" <td>1</td>\n",
" <td>General Petite</td>\n",
" <td>Tops</td>\n",
" <td>Knits</td>\n",
" <td>cagrcoal shimmer fun i aded this in my basket ...</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>\n",
" <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-74ab786f-b806-4078-b58f-a96c0cd62907')\"\n",
" title=\"Convert this dataframe to an interactive table.\"\n",
" style=\"display:none;\">\n",
" \n",
" <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
" width=\"24px\">\n",
" <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
" <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
" </svg>\n",
" </button>\n",
" \n",
" <style>\n",
" .colab-df-container {\n",
" display:flex;\n",
" flex-wrap:wrap;\n",
" gap: 12px;\n",
" }\n",
"\n",
" .colab-df-convert {\n",
" background-color: #E8F0FE;\n",
" border: none;\n",
" border-radius: 50%;\n",
" cursor: pointer;\n",
" display: none;\n",
" fill: #1967D2;\n",
" height: 32px;\n",
" padding: 0 0 0 0;\n",
" width: 32px;\n",
" }\n",
"\n",
" .colab-df-convert:hover {\n",
" background-color: #E2EBFA;\n",
" box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
" fill: #174EA6;\n",
" }\n",
"\n",
" [theme=dark] .colab-df-convert {\n",
" background-color: #3B4455;\n",
" fill: #D2E3FC;\n",
" }\n",
"\n",
" [theme=dark] .colab-df-convert:hover {\n",
" background-color: #434B5C;\n",
" box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
" filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
" fill: #FFFFFF;\n",
" }\n",
" </style>\n",
"\n",
" <script>\n",
" const buttonEl =\n",
" document.querySelector('#df-74ab786f-b806-4078-b58f-a96c0cd62907 button.colab-df-convert');\n",
" buttonEl.style.display =\n",
" google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
"\n",
" async function convertToInteractive(key) {\n",
" const element = document.querySelector('#df-74ab786f-b806-4078-b58f-a96c0cd62907');\n",
" const dataTable =\n",
" await google.colab.kernel.invokeFunction('convertToInteractive',\n",
" [key], {});\n",
" if (!dataTable) return;\n",
"\n",
" const docLinkHtml = 'Like what you see? Visit the ' +\n",
" '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
" + ' to learn more about interactive tables.';\n",
" element.innerHTML = '';\n",
" dataTable['output_type'] = 'display_data';\n",
" await google.colab.output.renderOutput(dataTable, element);\n",
" const docLink = document.createElement('div');\n",
" docLink.innerHTML = docLinkHtml;\n",
" element.appendChild(docLink);\n",
" }\n",
" </script>\n",
" </div>\n",
" </div>\n",
" "
]
},
"metadata": {},
"execution_count": 6
}
]
},
{
"cell_type": "markdown",
"source": [
"## Build, Compile, and Train the RNN model."
],
"metadata": {
"id": "-JeCRgnYsr0N"
}
},
{
"cell_type": "markdown",
"source": [
"Preprocess the text data."
],
"metadata": {
"id": "XJ5VUWjVRf6z"
}
},
{
"cell_type": "code",
"source": [
"tokenizer = Tokenizer()\n",
"tokenizer.fit_on_texts(df['Reviews'])\n",
"sequences = tokenizer.texts_to_sequences(df['Reviews'])\n",
"data = pad_sequences(sequences, maxlen=1000)"
],
"metadata": {
"id": "bawSIp1j1dsR"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"Split the data into train and test.\n"
],
"metadata": {
"id": "nXzWe7qb1rNs"
}
},
{
"cell_type": "code",
"source": [
"X = data\n",
"y = df['Recommended IND'].values\n",
"X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=7)"
],
"metadata": {
"id": "migA3RyF1jIw"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"Build the model."
],
"metadata": {
"id": "CHuSAXozGrh2"
}
},
{
"cell_type": "code",
"source": [
"model = Sequential()\n",
"model.add(Embedding(input_dim=len(tokenizer.word_index)+1, output_dim=128, input_length=1000))\n",
"model.add(GRU(128, return_sequences=True))\n",
"model.add(LSTM(64))\n",
"model.add(Dense(64, activation='relu'))\n",
"model.add(Dropout(0.5))\n",
"model.add(Dense(1, activation='sigmoid'))"
],
"metadata": {
"id": "HwLrzXws1jtE"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"Compile the model."
],
"metadata": {
"id": "OgOiqrttGskH"
}
},
{
"cell_type": "code",
"source": [
"model.compile(optimizer='adam', \n",
" loss='binary_crossentropy', \n",
" metrics=['accuracy', \n",
" tf.keras.metrics.Precision(), \n",
" tf.keras.metrics.Recall(), \n",
" tf.keras.metrics.AUC()])"
],
"metadata": {
"id": "uhv2W7XA1jaU"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"Train the model."
],
"metadata": {
"id": "NcPhWjLgGuT1"
}
},
{
"cell_type": "code",
"source": [
"history = model.fit(X_train,\n",
" y_train,\n",
" validation_data=(X_test, y_test),\n",
" epochs=10,\n",
" callbacks=[EarlyStopping(monitor='val_loss', patience=3)])"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "qCVjnDdo1Oai",
"outputId": "3c7ac86c-0047-4c15-90d1-0adba642a4af"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Epoch 1/10\n",
"492/492 [==============================] - 73s 135ms/step - loss: 0.3421 - accuracy: 0.8543 - precision: 0.8904 - recall: 0.9372 - auc: 0.8547 - val_loss: 0.2371 - val_accuracy: 0.9044 - val_precision: 0.9378 - val_recall: 0.9459 - val_auc: 0.9366\n",
"Epoch 2/10\n",
"492/492 [==============================] - 39s 79ms/step - loss: 0.2103 - accuracy: 0.9149 - precision: 0.9488 - recall: 0.9471 - auc: 0.9494 - val_loss: 0.2385 - val_accuracy: 0.8963 - val_precision: 0.9680 - val_recall: 0.9031 - val_auc: 0.9478\n",
"Epoch 3/10\n",
"492/492 [==============================] - 35s 70ms/step - loss: 0.1554 - accuracy: 0.9409 - precision: 0.9663 - recall: 0.9614 - auc: 0.9712 - val_loss: 0.2509 - val_accuracy: 0.9115 - val_precision: 0.9424 - val_recall: 0.9500 - val_auc: 0.9333\n",
"Epoch 4/10\n",
"492/492 [==============================] - 35s 72ms/step - loss: 0.1329 - accuracy: 0.9524 - precision: 0.9745 - recall: 0.9672 - auc: 0.9787 - val_loss: 0.2568 - val_accuracy: 0.8927 - val_precision: 0.9510 - val_recall: 0.9161 - val_auc: 0.9396\n"
]
}
]
},
{
"cell_type": "markdown",
"source": [
"Display the model architecture."
],
"metadata": {
"id": "cjTzwqgTOvXT"
}
},
{
"cell_type": "code",
"source": [
"tf.keras.utils.plot_model(model, show_shapes=True)"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 754
},
"id": "EZIWMowdOuvZ",
"outputId": "dc59aa76-8245-419c-ee34-a2d315d9249d"
},
"execution_count": null,
"outputs": [
{
"output_type": "execute_result",
"data": {
"image/png": "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\n",
"text/plain": [
"<IPython.core.display.Image object>"
]
},
"metadata": {},
"execution_count": 12
}
]
},
{
"cell_type": "markdown",
"source": [
"## Evaluate the Model"
],
"metadata": {
"id": "iePXXV_bud80"
}
},
{
"cell_type": "code",
"source": [
"loss, accuracy, precision, recall, auc = model.evaluate(X_test, y_test)\n",
"print(\"Loss: \", loss)\n",
"print(\"Accuracy: \", accuracy)"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "pwVL8Jzs2B9N",
"outputId": "175a4954-83d7-4654-b13b-0fc43176d136"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"123/123 [==============================] - 4s 28ms/step - loss: 0.2568 - accuracy: 0.8927 - precision: 0.9510 - recall: 0.9161 - auc: 0.9396\n",
"Loss: 0.2568265199661255\n",
"Accuracy: 0.8927027583122253\n"
]
}
]
},
{
"cell_type": "markdown",
"source": [
"Forecast the label."
],
"metadata": {
"id": "VCRr40gXOIFD"
}
},
{
"cell_type": "code",
"source": [
"y_pred = (model.predict(X_test)> 0.5).astype(int)"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "FFbthTLVOJho",
"outputId": "8284cc3f-368a-48cf-b69e-a4ed8c59e3cf"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"123/123 [==============================] - 4s 26ms/step\n"
]
}
]
},
{
"cell_type": "markdown",
"source": [
"Get the confusion matrix."
],
"metadata": {
"id": "DOZ8afHBNZmL"
}
},
{
"cell_type": "code",
"source": [
"confusion_matrix(y_test, y_pred)"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "iqNj-PapNZ2S",
"outputId": "5571bb39-7e82-4b01-c889-e5bd717979eb"
},
"execution_count": null,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"array([[ 562, 152],\n",
" [ 270, 2949]])"
]
},
"metadata": {},
"execution_count": 15
}
]
},
{
"cell_type": "markdown",
"source": [
"Print out the classification report."
],
"metadata": {
"id": "-jAyAD99NgNS"
}
},
{
"cell_type": "code",
"source": [
"label_names = ['negative', 'positive']\n",
"print(classification_report(y_test, y_pred, target_names=label_names))"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "ltWjEv04NiuK",
"outputId": "676df559-e673-4d74-bd63-af73ac0ccf45"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
" precision recall f1-score support\n",
"\n",
" negative 0.68 0.79 0.73 714\n",
" positive 0.95 0.92 0.93 3219\n",
"\n",
" accuracy 0.89 3933\n",
" macro avg 0.81 0.85 0.83 3933\n",
"weighted avg 0.90 0.89 0.90 3933\n",
"\n"
]
}
]
},
{
"cell_type": "markdown",
"source": [
"Plot accuracy and loss."
],
"metadata": {
"id": "W5huhEvgGybp"
}
},
{
"cell_type": "code",
"source": [
"# Plot training and validation accuracy\n",
"plt.figure(figsize=(12, 4))\n",
"plt.subplot(1, 2, 1)\n",
"plt.plot(history.history['accuracy'])\n",
"plt.plot(history.history['val_accuracy'])\n",
"plt.title('Model accuracy')\n",
"plt.ylabel('Accuracy')\n",
"plt.xlabel('Epoch')\n",
"plt.legend(['Train', 'Test'], loc='upper left')\n",
"\n",
"# Plot training and validation loss\n",
"plt.subplot(1, 2, 2)\n",
"plt.plot(history.history['loss'])\n",
"plt.plot(history.history['val_loss'])\n",
"plt.title('Model loss')\n",
"plt.ylabel('Loss')\n",
"plt.xlabel('Epoch')\n",
"plt.legend(['Train', 'Test'], loc='upper left')\n",
"\n",
"plt.tight_layout()\n",
"plt.show()"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 295
},
"id": "_anoUcU32D7l",
"outputId": "7cc7af06-4f90-4a72-ac2d-7ed36b4dc9c4"
},
"execution_count": null,
"outputs": [
{
"output_type": "display_data",
"data": {
"text/plain": [
"<Figure size 1200x400 with 2 Axes>"
],
"image/png": "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\n"
},
"metadata": {}
}
]
},
{
"cell_type": "markdown",
"source": [
"## Recommendation"
],
"metadata": {
"id": "ExhoRIvOI6nR"
}
},
{
"cell_type": "markdown",
"source": [
"The RNN model we built has shown strong performance, achieving an accuracy of 89.27% on the test data, alongside strong precision, recall, and AUC-ROC metrics. This indicates its capability to effectively predict customer recommendations based on reviews. However, before deploying this model, consider the following points:\n",
"\n",
"1. **Interpretability**: The model uses complex GRU and LSTM layers, which can make interpretations challenging. If your business requires interpretable models, consider simpler models or use techniques like LIME or SHAP for interpretations.\n",
"\n",
"2. **Generalization**: Test the model's performance on unseen data to ensure it generalizes well beyond the test data.\n",
"\n",
"3. **Efficiency**: Given its complexity, the model could be computationally intensive. If accuracy gains don't justify resource expenditure, consider simpler architectures.\n",
"\n",
"4. **Maintenance**: Post-deployment, continuous monitoring is necessary to ensure consistent performance. Regular retraining or updates may also be needed with fresh data.\n",
"\n",
"Overall, given its high accuracy and solid performance metrics, the model is promising. However, the final decision should consider your specific business needs and resources."
],
"metadata": {
"id": "VbicMbkQI9a7"
}
}
]
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment