Upload 5 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+alldata_1_for_kaggle.csv filter=lfs diff=lfs merge=lfs -text

alldata_1_for_kaggle.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c3b56772f9f0940ab5ba4c7c8c85afff6e75b06463428483867278e03b614d9
+size 189221599

app.py

@@ -0,0 +1,209 @@
+import streamlit as st
+import pandas as pd
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+import numpy as np
+import os
+from PIL import Image
+from sklearn.preprocessing import StandardScaler, LabelEncoder
+from tokenizers import Tokenizer
+from tokenizers.models import WordLevel
+from tokenizers.pre_tokenizers import Whitespace
+from collections import Counter
+import torch
+from torch.nn.utils.rnn import pad_sequence
+
+
+
+st.set_page_config(layout="centered")
+
+# Add custom CSS for background image and styling
+# Add custom CSS for background image and styling
+st.markdown("""
+    <style>
+        .stApp {
+            background-image: url("");
+            background-size: cover;
+            background-position: center;
+            background-repeat: no-repeat;
+            height: auto;  /* Allows the page to expand for scrolling */
+            overflow: auto;  /* Enables scrolling if the page content overflows */
+            # position : relative
+        }
+
+        /* Adjust opacity of overlay to make content more visible */
+        .stApp::before {
+            content: "";
+            position: absolute;
+            top: 0;
+            left: 0;
+            width: 100%;
+            height: 100%;
+            background-color: rgba(255, 255, 255, 0.8);  /* Slightly higher opacity */
+            z-index: -1;
+        }
+
+        /* Ensure content appears above the overlay */
+        .stApp > * {
+            position: relative;
+            z-index: 2;
+        }
+
+        /* Ensure the dataframe is visible */
+        .dataframe {
+            background-color: rgba(255, 255, 255, 0.9) !important;
+            z-index: 3;
+        }
+
+        /* Style text elements for better visibility */
+        h1, h3, span, div {
+            text-shadow: 1px 1px 2px rgba(255, 255, 255, 0.2);
+        }
+            
+        /* Custom CSS for select box heading */
+        div.stSelectbox > label {
+            color: #000000 !important;  /* Change to your desired color */
+            # background-color: black !important;  /* Background color of the dropdown */
+            font-size: 24px !important;  /* Change font size */
+            font-weight: bold !important;  /* Make text bold */
+        }
+
+        /* Custom CSS for image caption */
+        .custom-caption {
+            color: #000000 !important;  /* Change to your desired color */
+            font-size: 24px !important;  /* Optional: Change font size */
+            text-align: center;  /* Center-align the caption */
+        }
+            
+        .stMainBlockContainer {
+            background-color: white !important;  /* Background color of the dropdown */
+        }
+
+        .stTextArea{
+            color: #000000 !important
+        }
+            
+    </style>
+""", unsafe_allow_html=True)
+
+
+# Custom title styling functions
+def colored_title(text, color):
+    st.markdown(f"<h1 style='color: {color};'>{text}</h1>", unsafe_allow_html=True)
+
+def colored_subheader(text, color):
+    st.markdown(f"<h3 style='color: {color};'>{text}</h3>", unsafe_allow_html=True)
+
+def colored_text(text, color):
+    st.markdown(f"<span style='color: {color};'>{text}</span>", unsafe_allow_html=True)
+
+
+embedding_dim = 128                        
+hidden_units = 128                         
+num_classes = 3                             
+
+class RNNModel(nn.Module):
+    def __init__(self, vocab_size, embedding_dim, hidden_units, num_classes):
+        super(RNNModel, self).__init__()
+        self.embedding = nn.Embedding(vocab_size, embedding_dim)
+        self.rnn = nn.RNN(embedding_dim, hidden_units, batch_first=True, dropout=0.2)
+        self.fc = nn.Linear(hidden_units, num_classes)
+
+    def forward(self, x):
+        x = self.embedding(x)
+        output, _ = self.rnn(x)
+        x = output[:, -1, :]  # Use last timestep output
+        x = self.fc(x)
+        return F.softmax(x, dim=1)
+
+@st.cache_resource
+def load_model(vocab_size):
+
+    model = RNNModel(vocab_size,embedding_dim,hidden_units,num_classes)
+    try:
+        state_dict = torch.load('rnn_classification_model_weights.pth', map_location=torch.device('cpu'))
+        model.load_state_dict(state_dict)
+        model.eval()
+        return model
+    except Exception as e:
+        st.error(f"Error loading model: {str(e)}")
+        return None
+
+@st.cache_data
+def load_data():
+
+    df=pd.read_csv("alldata_1_for_kaggle.csv",encoding='latin1')
+    df = df.rename(columns={'0': 'labels', 'a': 'text'})
+    texts = df['text'].values
+    unique_classes = (df['labels'].unique())  # Ensure consistent order
+    class_mapping = {i: f"{idx}" for i,idx in enumerate(unique_classes)}
+
+    # Tokenization (basic whitespace tokenizer)
+    def tokenize(text):
+        return text.lower().split()
+
+    # Build vocabulary based on word frequency (similar to Keras Tokenizer)
+    word_counts = Counter()
+    for text in texts:
+        word_counts.update(tokenize(text))
+
+    # Sort words by frequency (most common words get lower indices)
+    sorted_words = [word for word, _ in word_counts.most_common()]
+
+    # Create vocabulary mapping with <pad> and <unk> tokens
+    vocab = {"<pad>": 0, "<unk>": 1}
+    vocab.update({word: idx + 2 for idx, word in enumerate(sorted_words)})
+
+    # Initialize tokenizer
+    tokenizer = Tokenizer(WordLevel(vocab, unk_token="<unk>"))
+    tokenizer.pre_tokenizer = Whitespace()
+
+    # Convert texts to sequences
+    def text_to_sequence(texts):
+        return [tokenizer.encode(text.lower()).ids for text in texts]
+
+    X_train_seq = text_to_sequence(texts)
+    max_len = max([len(seq) for seq in X_train_seq])
+    # Convert each sequence to a tensor individually
+    X_train_seq = [torch.tensor(seq) for seq in X_train_seq]
+
+    # Padding the sequences correctly
+    X_train_seq_reversed = [seq.flip(0) for seq in X_train_seq]
+    X_train_pad_reversed = pad_sequence(X_train_seq_reversed, batch_first=True, padding_value=0)
+    X_train_pad = X_train_pad_reversed.flip(1)
+    vocab_size = len(tokenizer.get_vocab()) 
+
+    return X_train_pad, texts, class_mapping, vocab_size
+
+
+def main():
+    colored_title("Text Classification using RNN", "black")
+
+    # Load data
+    X_test,texts,class_mapping, vocab_size = load_data()
+
+
+    # Display test images for selection
+    colored_subheader("Select a Row for Prediction:", "black")
+    selected_index = st.selectbox("Select a row", options=range(len(texts)), index=0)
+
+    colored_text("Selected Text:","black")
+    st.text_area("Text Content:", value=texts[selected_index], height=150, disabled=True)
+
+    # Predict button
+    if st.button("Predict"):
+        model = load_model(vocab_size)
+        if model is not None:
+            with torch.no_grad():
+                output = model(X_test[[selected_index]])
+                predicted_class = torch.argmax(output, dim=1).item()
+
+            # Display prediction result
+            colored_subheader("Prediction Results:", "green")
+            colored_text(f"Predicted Class: {class_mapping[predicted_class]}", "green") 
+    
+
+            
+if __name__ == "__main__":
+    main()

requirements.txt

@@ -0,0 +1,7 @@
+streamlit==1.30.0
+pandas==2.1.4
+torch==2.2.0
+torchvision==0.17.0
+numpy==1.26.3
+scikit-learn==1.3.2
+tokenizers==0.15.1

rnn_classification_model_weights.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a89b017c0a640b6cc429d433a47a01801da0e66d284219fbf77123048ad74e
+size 115119704

text_classification_using_rnn.ipynb

@@ -0,0 +1,1039 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "7c413f6e-62bf-43e5-91b5-f68761a90bef",
+   "metadata": {},
+   "source": [
+    "# 1. Objective"
+   ]
+  },
+  {
+   "cell_type": "raw",
+   "id": "be82cf3f-4671-480d-aaca-c4f238e3810d",
+   "metadata": {},
+   "source": [
+    "Text Classification using RNN"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "65506456-317f-4ef6-af5c-99ffa1006551",
+   "metadata": {},
+   "source": [
+    "# 2. Imports"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "id": "11b56a45",
+   "metadata": {
+    "colab": {
+     "base_uri": "https://localhost:8080/",
+     "height": 206
+    },
+    "id": "11b56a45",
+    "outputId": "e0a3e575-bbac-4482-bb49-14d1168be7c7",
+    "papermill": {
+     "duration": 5.049477,
+     "end_time": "2024-08-08T15:25:12.643518",
+     "exception": false,
+     "start_time": "2024-08-08T15:25:07.594041",
+     "status": "completed"
+    },
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "#read the data\n",
+    "import pandas as pd\n",
+    "import time\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "from tokenizers import Tokenizer\n",
+    "from tokenizers.models import WordLevel\n",
+    "from tokenizers.pre_tokenizers import Whitespace\n",
+    "from collections import Counter\n",
+    "import torch\n",
+    "from torch.nn.utils.rnn import pad_sequence\n",
+    "import torch\n",
+    "import torch.nn.functional as F\n",
+    "from sklearn.preprocessing import LabelEncoder\n",
+    "import torch\n",
+    "import torch.nn as nn\n",
+    "import torch.nn.functional as F\n",
+    "import torch.optim as optim\n",
+    "from torch.utils.data import DataLoader, TensorDataset\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "5a54d59b-e63a-4272-80f3-28bbacee8fef",
+   "metadata": {},
+   "source": [
+    "## GPU"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "a7725648-ea0f-464e-a643-a5540ab96550",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "'NVIDIA GeForce RTX 4050 Laptop GPU'"
+      ]
+     },
+     "execution_count": 2,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "torch.cuda.get_device_name()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "5cb79f2d",
+   "metadata": {
+    "id": "5cb79f2d",
+    "papermill": {
+     "duration": 0.009874,
+     "end_time": "2024-08-08T15:25:12.665020",
+     "exception": false,
+     "start_time": "2024-08-08T15:25:12.655146",
+     "status": "completed"
+    },
+    "tags": []
+   },
+   "source": [
+    "## 3.Initial Data Exploration and Cleaning"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "48f9564d-b1b7-4c91-a0c8-603b3f253576",
+   "metadata": {},
+   "outputs": [
+    {
+     "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>Unnamed: 0</th>\n",
+       "      <th>0</th>\n",
+       "      <th>a</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0</td>\n",
+       "      <td>Thyroid_Cancer</td>\n",
+       "      <td>Thyroid surgery in  children in a single insti...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>1</td>\n",
+       "      <td>Thyroid_Cancer</td>\n",
+       "      <td>\" The adopted strategy was the same as that us...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>2</td>\n",
+       "      <td>Thyroid_Cancer</td>\n",
+       "      <td>coronary arterybypass grafting thrombosis fib...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>3</td>\n",
+       "      <td>Thyroid_Cancer</td>\n",
+       "      <td>Solitary plasmacytoma SP of the skull is an u...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>4</td>\n",
+       "      <td>Thyroid_Cancer</td>\n",
+       "      <td>This study aimed to investigate serum matrix ...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   Unnamed: 0               0  \\\n",
+       "0           0  Thyroid_Cancer   \n",
+       "1           1  Thyroid_Cancer   \n",
+       "2           2  Thyroid_Cancer   \n",
+       "3           3  Thyroid_Cancer   \n",
+       "4           4  Thyroid_Cancer   \n",
+       "\n",
+       "                                                   a  \n",
+       "0  Thyroid surgery in  children in a single insti...  \n",
+       "1  \" The adopted strategy was the same as that us...  \n",
+       "2  coronary arterybypass grafting thrombosis fib...  \n",
+       "3   Solitary plasmacytoma SP of the skull is an u...  \n",
+       "4   This study aimed to investigate serum matrix ...  "
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df=pd.read_csv(\"alldata_1_for_kaggle.csv\",encoding='latin1')\n",
+    "df.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "b525ffae",
+   "metadata": {
+    "colab": {
+     "base_uri": "https://localhost:8080/"
+    },
+    "id": "b525ffae",
+    "outputId": "dc7c2843-0900-426d-afdc-a667cdaec5eb",
+    "papermill": {
+     "duration": 0.037312,
+     "end_time": "2024-08-08T15:25:12.711939",
+     "exception": false,
+     "start_time": "2024-08-08T15:25:12.674627",
+     "status": "completed"
+    },
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "<class 'pandas.core.frame.DataFrame'>\n",
+      "RangeIndex: 7570 entries, 0 to 7569\n",
+      "Data columns (total 3 columns):\n",
+      " #   Column      Non-Null Count  Dtype \n",
+      "---  ------      --------------  ----- \n",
+      " 0   Unnamed: 0  7570 non-null   int64 \n",
+      " 1   0           7570 non-null   object\n",
+      " 2   a           7570 non-null   object\n",
+      "dtypes: int64(1), object(2)\n",
+      "memory usage: 177.6+ KB\n"
+     ]
+    }
+   ],
+   "source": [
+    "df.info()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "40635443",
+   "metadata": {
+    "colab": {
+     "base_uri": "https://localhost:8080/"
+    },
+    "id": "40635443",
+    "outputId": "de366877-cc8d-4366-bd77-b92645ffc2b4",
+    "papermill": {
+     "duration": 0.287576,
+     "end_time": "2024-08-08T15:25:13.030383",
+     "exception": false,
+     "start_time": "2024-08-08T15:25:12.742807",
+     "status": "completed"
+    },
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "np.int64(0)"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "#Check Duplicated vals\n",
+    "df.duplicated().sum()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "f48819da",
+   "metadata": {
+    "id": "f48819da",
+    "papermill": {
+     "duration": 0.019039,
+     "end_time": "2024-08-08T15:25:13.061571",
+     "exception": false,
+     "start_time": "2024-08-08T15:25:13.042532",
+     "status": "completed"
+    },
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "#rename cols\n",
+    "df = df.rename(columns={'0': 'labels', 'a': 'text'})"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "ecb539f3",
+   "metadata": {
+    "colab": {
+     "base_uri": "https://localhost:8080/",
+     "height": 209
+    },
+    "id": "ecb539f3",
+    "outputId": "9d7a45f0-840d-4185-8c72-0caf92e261e5",
+    "papermill": {
+     "duration": 0.023086,
+     "end_time": "2024-08-08T15:25:13.095199",
+     "exception": false,
+     "start_time": "2024-08-08T15:25:13.072113",
+     "status": "completed"
+    },
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "labels\n",
+       "Thyroid_Cancer    2810\n",
+       "Colon_Cancer      2580\n",
+       "Lung_Cancer       2180\n",
+       "Name: count, dtype: int64"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df['labels'].value_counts()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "e85ba39f-6029-47eb-8c43-d68870e73be7",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "{0: 'Thyroid_Cancer', 1: 'Colon_Cancer', 2: 'Lung_Cancer'}\n"
+     ]
+    }
+   ],
+   "source": [
+    "unique_classes = (df['labels'].unique())  # Ensure consistent order\n",
+    "class_mapping = {i: f\"{idx}\" for i,idx in enumerate(unique_classes)}\n",
+    "\n",
+    "print(class_mapping)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "86b8333a",
+   "metadata": {
+    "id": "86b8333a",
+    "papermill": {
+     "duration": 0.017608,
+     "end_time": "2024-08-08T15:25:13.143258",
+     "exception": false,
+     "start_time": "2024-08-08T15:25:13.125650",
+     "status": "completed"
+    },
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "texts = df['text'].values\n",
+    "labels = df['labels'].values"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "796aa449",
+   "metadata": {
+    "id": "796aa449",
+    "papermill": {
+     "duration": 0.009915,
+     "end_time": "2024-08-08T15:25:13.163996",
+     "exception": false,
+     "start_time": "2024-08-08T15:25:13.154081",
+     "status": "completed"
+    },
+    "tags": []
+   },
+   "source": [
+    "## 4.Spliting the data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "78cd4ab3",
+   "metadata": {
+    "colab": {
+     "base_uri": "https://localhost:8080/"
+    },
+    "id": "78cd4ab3",
+    "outputId": "14a16d84-74ef-43be-cc22-1adcb1d5eea3",
+    "papermill": {
+     "duration": 1.105941,
+     "end_time": "2024-08-08T15:25:14.280223",
+     "exception": false,
+     "start_time": "2024-08-08T15:25:13.174282",
+     "status": "completed"
+    },
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Dimensions of X_train : (6056,)\n",
+      "Dimensions of X_test  : (1514,)\n",
+      "Dimensions of y_train : (6056,)\n",
+      "Dimensions of y_test  : (1514,)\n"
+     ]
+    }
+   ],
+   "source": [
+    "X_train, X_test, y_train, y_test = train_test_split(texts, labels, test_size=0.2, random_state=42,shuffle=True,stratify=labels)\n",
+    "\n",
+    "print(\"Dimensions of X_train :\", X_train.shape)\n",
+    "print(\"Dimensions of X_test  :\", X_test.shape)\n",
+    "print(\"Dimensions of y_train :\", y_train.shape)\n",
+    "print(\"Dimensions of y_test  :\", y_test.shape)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "6597d366",
+   "metadata": {
+    "id": "6597d366",
+    "papermill": {
+     "duration": 0.010665,
+     "end_time": "2024-08-08T15:25:14.321468",
+     "exception": false,
+     "start_time": "2024-08-08T15:25:14.310803",
+     "status": "completed"
+    },
+    "tags": []
+   },
+   "source": [
+    "## 5.Text Tokenization and Sequence Conversion\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "heM2tow_QwB-",
+   "metadata": {
+    "colab": {
+     "base_uri": "https://localhost:8080/"
+    },
+    "id": "heM2tow_QwB-",
+    "outputId": "95c07961-806f-4061-9b00-16c0854369f2"
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[1373, 327, 6, 33362, 8237, 50, 7, 865, 199, 8, 2, 167, 12848, 5989, 16, 1929, 4, 12, 2, 17591, 106, 278, 16, 5, 768, 4, 2, 6365, 10870, 16, 5, 308, 1500, 644, 5989, 16, 64577, 22, 4619, 105, 204996, 2576, 211, 1239, 124, 20, 91, 64, 25, 1436, 2, 2277, 1593, 26178, 1955, 4, 55, 26, 79, 8, 632, 483, 1304, 72611, 25, 3, 28127, 35898, 4, 4, 1474, 128, 178, 64578, 237, 2, 160, 2, 2576, 1593, 12, 419, 1955, 4, 12, 204997, 13, 570, 64, 25, 3, 7940, 7153, 204998, 1639, 4, 10926, 2086, 59, 3555, 5765, 30, 33, 340, 115, 204999, 5989, 17, 2877, 1923, 7, 2277, 1593, 3, 1929, 1634, 39818, 8, 632, 483, 30, 2725, 14, 2, 211, 1593, 501, 2877, 17, 129, 26, 1159, 13, 1851, 2, 25, 3, 1639, 822, 1134, 7738, 3, 2, 17, 40968, 4, 2877, 17, 11, 185, 8, 205000, 39819, 4017, 15679, 1775, 75, 22721, 1039, 29, 342, 444, 15679, 1775, 75, 19, 7, 349, 32430, 4, 1039, 9, 239, 2, 60, 260, 1368, 14, 46, 3, 2, 2877, 12168, 1702, 521, 2, 25, 3, 1639, 4, 5765, 663, 46, 8, 205001, 2, 25, 3, 213, 3, 2, 2277, 28127, 107, 7, 536, 46, 32430, 4, 1039, 995, 5, 7, 704, 560, 3, 255, 198, 79, 8, 2, 79287, 3, 1572, 240, 35898, 21086, 31027, 3876, 4, 19718, 74, 32, 14472, 228, 3, 1639, 179, 1134, 24, 172, 3183, 2, 333, 3, 77744, 5, 2720, 2, 2576, 1593, 3, 1929, 5, 434, 673, 57, 22, 1074, 24327, 2, 587, 3, 37, 1639, 3818, 822, 172, 547, 6, 760, 205002, 3, 205003, 8319, 79288, 1199, 205004, 78, 31028, 205005, 76, 1403, 24328, 34776, 3681, 228, 3, 1639, 4, 318, 3, 5306, 1496, 442, 198, 79, 2663, 2277, 1593, 3, 1572, 16, 5, 24689, 8001, 17, 72, 205006, 1781, 205007, 31029, 253, 16162, 154, 3, 818, 205008, 15475, 67, 844, 1266, 321, 455, 1047, 79288, 23, 12, 29, 9481, 773, 180, 2, 586, 3, 9764, 254, 869, 807, 7176, 3489, 108, 283, 7821, 5, 224, 354, 430, 2, 8095, 4, 619, 22, 7669, 1749, 3155, 49, 899, 32, 11547, 2, 1233, 4, 117, 1959, 23206, 205009, 205010, 40969, 1016, 3, 29394, 139, 1038, 3196, 1674, 10137, 205011, 119, 4, 2131, 6232, 205012, 5714, 1573, 205013, 205014, 21, 14079, 717, 3, 612, 164, 1038, 7472, 72, 1781, 18511, 455, 72, 42493, 1918, 2, 158, 33, 8418, 56, 1076, 1918, 205015, 3, 2576, 131, 5, 2277, 1572, 205016, 16, 552, 12, 2, 21719, 106, 278, 16, 601, 1169, 4, 5, 17441, 16, 771, 774, 14, 377, 252, 121, 3, 552, 502, 26, 196, 697, 971, 4, 377, 970, 502, 1562, 21, 1572, 16, 161, 7237, 205017, 1718, 1929, 22, 2, 106, 278, 635, 2, 557, 106, 601, 16, 3, 205018, 1084, 5504, 743, 16, 90, 118, 53, 12, 2, 17591, 106, 278, 16, 5, 19911, 2, 6365, 106, 278, 16, 5, 308, 5989, 240, 22, 1495, 15, 1500, 644, 64577, 19, 205019, 205020, 5989, 240, 33, 7, 246, 5617, 6, 2425, 205021, 33, 7, 99, 152, 3, 183, 4, 7, 881, 82, 152, 3, 842, 205022, 4, 205023, 33, 52, 75559, 105, 315, 169, 8, 2, 2277, 4, 2576, 1593, 635, 2, 106, 2641, 2, 2576, 1593, 3, 88, 1929, 1607, 2, 1283, 501, 381, 229, 2577, 32431, 4, 7, 109, 116, 3, 79289, 23, 1593, 205024, 9, 1507, 2351, 12, 419, 1955, 4, 50, 7, 83, 2971, 113, 588, 90, 118, 2277, 1495, 160, 33, 7, 3544, 216, 82, 73, 6, 2576, 160, 181, 740, 1923, 632, 483, 22, 1955, 4, 87, 4285, 205025, 14, 9270, 4, 1622, 6, 2824, 1052, 377, 3, 79289, 6264, 387, 3, 2, 1507, 131, 33, 405, 406, 4, 29, 216, 881, 82, 152, 3, 1373, 327, 6, 33362, 8237, 50, 7, 865, 199, 8, 2, 7782, 88, 1929, 2, 102, 81, 3, 1929, 3162, 6, 2, 338, 7238, 1201, 3, 2, 205026, 1277, 33, 52, 196, 8, 79290, 606, 16, 309, 30, 205027, 7, 24, 763, 136, 3, 79290, 5989, 16, 13, 16435, 2, 9530, 205028, 5989, 24, 259, 8001, 6, 33362, 47, 4480, 24, 280, 1612, 160, 5, 17592, 110, 14, 2299, 54, 9, 21422, 205029, 4, 205030, 987, 506, 320, 6, 2, 2277, 1593, 3, 88, 64577, 90, 118, 2, 38701, 7, 2351, 320, 6, 205031, 24, 1718, 8, 2718, 1201, 3, 205032, 90, 118, 34, 12, 79, 8, 405, 266, 90, 118, 2, 211, 439, 5664, 7, 36, 2065, 7, 35899, 1593, 12, 205033, 5, 7, 378, 31, 11852, 571, 14, 386, 178, 205034, 2028, 19718, 12, 238, 27, 99, 71, 5, 35899, 1572, 205035, 5, 1201, 3, 632, 483, 4, 12, 79, 8, 247, 358, 12849, 4, 1182, 555, 3, 2, 58, 2, 31, 44, 603, 14, 102, 25, 3, 19718, 55, 2457, 2, 25, 3, 66, 386, 159, 85, 15, 36, 59, 3876, 205036, 44, 402, 15, 3876, 34, 12, 402, 6, 26, 79, 8, 2, 79287, 3, 1929, 205037, 234, 14, 3876, 1536, 6, 2, 1120, 751, 921, 3, 2, 64, 75560, 63, 86, 31912, 64, 205038, 38702, 15, 25741, 4, 21086, 276, 711, 91, 25, 2, 31, 143, 5146, 101, 21086, 4, 25741, 647, 2, 304, 3, 2, 2277, 1593, 3, 1929, 5, 443, 4, 16964, 2, 2576, 1593, 3, 1929, 12, 79, 8, 2, 25, 3, 2, 1120, 205039, 2086, 59, 3555, 5765, 34777, 257, 59, 8002, 4, 7940, 7153, 245, 2900, 1639, 90, 118, 2, 228, 3, 1639, 8, 29, 205040, 7822, 2, 25, 3, 19718, 4, 783, 632, 483, 5, 443, 2, 1623, 114, 3, 1639, 142, 5, 1844, 12, 7670, 2176, 107, 2600, 3, 1639, 5, 1572, 240, 12, 7, 2971, 305, 745, 101, 5, 23553, 5, 443, 57, 472, 14, 1639, 2485, 85, 15, 27459, 55, 647, 483, 783, 367, 172, 764, 1080, 4, 1480, 222, 5, 1572, 16, 72, 1781, 18511, 455, 72, 18133, 3, 2576, 131, 5, 2277, 1572, 16, 309, 8419, 2818, 367, 172, 8, 1474, 128, 178, 64578, 547, 50, 52, 48, 44332, 6, 760, 10719, 1500, 644, 205041, 128, 314, 2, 342, 547, 301, 35, 33, 2, 338, 587, 5, 205042, 2, 3742, 3, 23, 31, 10, 6, 526, 226, 2, 228, 3, 1639, 4, 318, 3, 5306, 5, 2, 8001, 3187, 4, 2, 24689, 8001, 16436, 502, 205043, 25, 3, 54, 275, 5, 2720, 2, 35899, 204, 3, 1572, 6997, 764, 54, 14, 11, 79, 8, 2, 205044, 1005, 3, 1572, 24329, 4, 205045, 2486, 79291, 110, 1415, 21, 44333, 11, 48, 5, 47, 57, 26548, 11, 9410, 332, 6, 7, 12023, 27, 1, 1, 68, 180, 7, 205046, 7460, 480, 747, 4, 1519, 72612, 1323, 10772, 356, 205047, 3, 2, 8001, 4480, 24, 280, 8237, 79292, 4, 2, 35900, 24, 259, 11, 1090, 13, 2583, 1025, 27, 7, 400, 3, 253, 5773, 2, 6000, 2162, 16041, 3, 17592, 2486, 79291, 110, 23, 31, 15031, 6, 205048, 17271, 5, 2, 3159, 9, 2, 227, 4, 108, 3, 666, 924, 3, 205049, 36, 2370, 11, 503, 4109, 4, 2, 924, 11, 4855, 162, 2, 269, 3, 5584, 10, 842, 28, 545, 19, 162, 17271, 13, 62, 477, 205050, 1447, 13, 2487, 40971, 4, 11225, 6, 2, 975, 5827, 164, 199, 3081, 20, 10720, 251, 10, 468, 13, 27460, 1097, 4, 205051, 236, 34, 2, 1052, 4, 36, 11, 2291, 227, 3, 1030, 6, 2113, 32432, 10, 56, 1743, 6, 2, 924, 145, 2, 1240, 2, 755, 10, 438, 13, 22722, 3, 2754, 64579, 673, 227, 860, 205052, 13630, 8001, 3187, 17, 4, 93, 3, 2, 24689, 16436, 8237, 4, 11091, 11, 964, 5, 5, 1, 5364, 76, 1766, 6213, 5364, 76, 354, 2819, 2170, 8, 8956, 79293, 258, 15, 391, 340, 2, 17, 11, 22185, 27, 7, 421, 205054, 4, 2, 2946, 11, 3710, 3273, 128, 354, 925, 166, 239, 2, 64580, 24, 259, 50, 52, 23207, 37, 1152, 6294, 5334, 19719, 42, 2, 24689, 16436, 48, 5, 2, 369, 31, 33, 52, 340, 1304, 9, 117, 569, 6, 652, 2780, 5, 3590, 4207, 652, 7872, 162, 3563, 5, 30488, 6424, 205055, 652, 160, 162, 1642, 5307, 5, 40972, 110, 90, 2722, 2, 8237, 55, 44, 652, 160, 935, 5788, 1025, 9, 303, 205056, 3187, 4, 2, 24689, 16436, 8237, 4, 11091, 11, 3563, 6, 10611, 23208, 590, 354, 236, 34, 2, 17, 11, 671, 8, 7, 258, 205057, 3382, 3, 7, 1596, 3, 2819, 4, 1, 5364, 76, 19394, 128, 354, 9, 77, 14886, 11, 182, 1377, 6, 549, 8363, 10773, 2845, 2, 303, 4567, 205059, 15679, 1775, 75, 7, 1639, 4017, 22721, 1039, 15679, 1775, 75, 29, 1474, 128, 178, 64578, 444, 19, 7, 349, 3, 1702, 4, 1039, 12480, 9, 4, 1428, 2, 496, 3, 2, 573, 11, 2776, 124, 20, 3208, 205060, 3, 79294, 205061, 141, 3, 1702, 4, 1039, 20, 2, 8001, 17, 10, 392, 13, 3979, 3, 205062, 79294, 2029, 15, 391, 340, 13, 23, 666, 4046, 203, 133, 1329, 2, 24, 21423, 11, 1793, 2288, 8, 3400, 7177, 72, 1781, 18511, 455, 72, 18133, 3, 2576, 131, 5, 2277, 1572, 205063, 1146, 236, 34, 2, 17, 11, 1295, 9, 466, 8, 2952, 4, 205064, 52551, 1146, 2, 15476, 17, 11, 1013, 8, 1, 1775, 192, 3606, 1, 1775, 1, 5, 9067, 205067, 2595, 4, 1631, 475, 17846, 215, 10, 1008, 37, 34778, 1802, 211, 119, 409, 15, 391, 340, 2, 25, 3, 315, 54, 10, 503, 8, 1631, 1000, 28818, 1351, 847, 1841, 37, 1608, 14, 11, 1415, 5178, 205068, 2054, 2, 54, 1157, 8, 2, 13926, 116, 3, 2, 1608, 22, 1776, 64581, 310, 1057, 92, 157, 215, 15679, 1775, 165, 10, 4504, 6, 1741, 37, 205069, 1741, 765, 700, 4135, 2054, 2, 1257, 3, 2, 1741, 10, 103, 37, 2, 52552, 4221, 4398, 1119, 20545, 4135, 2054, 8, 15679, 1775, 1, 4, 15679, 1775, 75, 1608, 5, 7, 157, 334, 3, 15679, 1775, 192, 5, 7, 17847, 1631, 416, 68039, 2054, 1257, 10, 3011, 13, 4398, 1119, 1644, 205071, 2750, 3, 7, 76, 72614, 451, 13, 1398, 3, 6967, 27, 1, 1, 68, 9, 205072, 27, 1, 1, 68, 9, 76, 4, 2558, 27, 1, 1, 68, 9, 76, 34, 3471, 1852, 205073, 2, 873, 71, 11, 1449, 6, 15679, 3087, 2209, 25, 503, 13, 2, 205074, 2, 2051, 1398, 29395, 9, 15679, 3087, 2209, 4, 2, 873, 7628, 29395, 9, 2, 218, 54, 27461, 5, 1829, 205075, 762, 64582, 762, 42, 10, 103, 15, 391, 340, 2, 24, 10288, 11, 5002, 5, 7370, 77745, 9961, 194, 3980, 3298, 1161, 2170, 205076, 2588, 444, 8213, 4, 2046, 205077, 3141, 3582, 2331, 9902, 4603, 10, 19148, 4, 2, 6968, 10, 3358, 6, 3943, 367, 12686, 205078, 11, 2001, 37, 2, 10774, 2542, 59, 194, 700, 205079, 205080, 6, 3642, 122, 11, 296, 4, 16601, 2, 59, 2648, 11, 2387, 205081, 205082, 77746, 6487, 1415, 21, 4135, 2054, 4, 2204, 11548, 15679, 1775, 75, 205083, 6858, 12582, 640, 37, 205084, 2464, 2, 205085, 2799, 5, 22723, 2254, 17593, 590, 20313, 9669, 4, 205086, 258, 3418, 7322, 9, 466, 27, 1219, 931, 2, 1931, 11, 205087, 27, 1, 1, 68, 8, 2, 188, 411, 3151, 5, 9326, 2172, 258, 3418, 205088, 11, 1415, 21, 1729, 18512, 4, 11, 1242, 390, 402, 202, 4, 298, 1634, 24, 280, 13, 446, 42, 721, 6, 108, 5, 618, 1949, 2, 619, 3, 2, 411, 1157, 8, 91, 13926, 994, 22, 98, 5, 2407, 205089, 4428, 661, 9, 2604, 96, 8957, 5, 9669, 2, 1931, 11, 671, 2663, 9327, 19, 8958, 793, 411, 9, 466, 27, 1219, 72615, 5699, 11, 3568, 37, 2, 10786, 56245, 446, 7018, 1364, 2337, 205091, 205092, 1054, 11, 2851, 27, 28, 15679, 1775, 75, 4, 28819, 5, 40973, 218, 6620, 205093, 5, 7, 205094, 9, 466, 2, 1054, 11, 2659, 6, 22724, 6, 1219, 931, 205095, 105, 9669, 9, 466, 2, 188, 562, 48, 5, 23, 31, 1157, 8, 205096, 4, 12766, 994, 22, 1776, 5, 2407, 92, 2, 188, 562, 11, 205097, 8067, 4060, 4877, 17848, 923, 9, 2559, 19, 356, 188, 562, 11853, 931, 9, 466, 2483, 11149, 8067, 10, 48, 9, 205098, 323, 10, 103, 9, 28, 5585, 27, 1219, 931, 37, 205099, 8067, 2404, 11, 10612, 5, 9067, 1519, 10927, 5, 6412, 2952, 72, 1781, 18511, 455, 72, 18133, 3, 2576, 131, 5, 2277, 1572, 205100, 5, 13146, 4, 205101, 93, 4961, 23209, 2, 401, 4, 1335, 205102, 5, 2, 205103, 7473, 450, 11, 103, 5, 3954, 4, 2, 60, 22, 238, 15, 2, 209, 65, 205104, 341, 11, 103, 37, 2811, 205105, 564, 724, 37, 3012, 1382, 8, 8772, 5364, 76, 19, 37069, 5364, 76, 6052, 795, 9, 64, 25, 1022, 205106, 20, 2, 7628, 480, 2051, 15679, 1, 464, 250, 14, 11, 804, 21, 2676, 6, 15679, 3087, 2209, 71, 646, 563, 1313, 2, 123, 3, 731, 10, 205107, 3, 27459, 4, 22721, 20, 2, 799, 4, 2468, 205108, 3187, 4, 2877, 7239, 8001, 3187, 4, 2877, 17, 8237, 4, 11091, 11, 185, 8, 549, 2845, 205109, 1702, 15679, 1775, 75, 22721, 1039, 15679, 1775, 75, 19, 1702, 4, 1039, 12480, 9, 670, 15, 26179, 38, 1078, 28, 3074, 126, 10, 56, 467, 5, 2, 2468, 3, 2, 8001, 3187, 17, 8, 205111, 126, 10, 7, 467, 5, 2, 2468, 3, 2, 8237, 4, 11091, 17, 162, 6042, 1702, 38, 6457, 4, 18321, 4, 12480, 38, 2771, 4, 25742, 2, 17, 8364, 83, 205112, 205113, 4, 24330, 2, 3523, 17, 3, 2, 1572, 126, 10, 7, 560, 697, 116, 3, 8001, 3187, 17, 185, 8, 1039, 4, 12480, 162, 73, 6, 2, 19912, 8, 1702, 676, 19, 8, 2845, 38, 30489, 126, 10, 44, 7, 560, 5, 2, 116, 205114, 17, 162, 185, 8, 1702, 4, 12480, 162, 73, 6, 2, 2845, 185, 17, 205115, 126, 10, 56, 109, 560, 5, 2, 116, 3, 8237, 17, 5, 224, 3, 2, 25743, 38, 29396, 29, 1096, 234, 2, 1103, 3, 6680, 17, 5, 2, 185, 169, 1794, 560, 5, 24, 116, 73, 6, 2, 2845, 63, 129, 26, 201, 6, 1103, 3, 205116, 102, 483, 3, 205117, 3, 1639, 228, 4, 342, 318, 20, 2, 25, 205118, 17594, 386, 159, 1639, 5765, 4, 8002, 695, 7, 114, 5, 2, 5346, 3, 52553, 1593, 3, 1572, 16, 90, 118, 57, 1447, 13, 64583, 33, 115, 205119, 228, 3, 1639, 8, 4749, 318, 3, 342, 809, 5, 144, 88, 5989, 17, 828, 2, 1685, 3, 2, 1639, 4017, 5, 13631, 31, 30, 613, 2, 156, 3, 2, 1639, 4017, 1702, 4, 29, 342, 444, 205120, 2, 25, 3, 2576, 1120, 159, 5, 2, 8001, 3187, 17, 4, 2, 205121, 25, 71, 9, 2, 218, 54, 5, 23, 31, 22, 98, 9, 7, 670, 670, 4, 670, 22725, 9, 2, 3187, 8237, 4, 11091, 17, 1057, 1829, 4, 2407, 3238, 219, 9, 33363, 2, 670, 64, 4, 59, 71, 22, 98, 5, 2, 641, 420, 3, 2, 205122, 8, 1702, 102, 2, 25, 3, 1639, 5, 2, 8001, 3187, 38, 1487, 193, 1375, 64584, 24, 259, 7, 320, 156, 10, 968, 5, 8237, 38, 1997, 193, 1375, 4, 584, 4, 11091, 38, 2606, 193, 1375, 4, 205123, 280, 46, 8, 1039, 301, 35, 764, 2, 25, 3, 1639, 5, 2, 8001, 205124, 1487, 1375, 4, 584, 19, 8237, 38, 1997, 1375, 4, 584, 17, 111, 126, 10, 7, 293, 247, 5, 1639, 59, 5, 2, 11091, 17, 38, 2606, 1375, 4, 584, 46, 8, 101, 1702, 4, 1039, 12480, 12408, 25, 71, 3, 1639, 268, 5, 2, 8001, 3187, 17, 111, 126, 10, 56, 247, 697, 59, 71, 126, 10, 56, 247, 5, 268, 25, 5, 2, 11091, 17, 111, 126, 10, 205125, 247, 5, 2, 59, 71, 38, 2606, 193, 1375, 4, 584, 72, 1781, 18511, 455, 72, 18133, 3, 2576, 131, 5, 2277, 1572, 12244, 2468, 4, 799, 3, 8001, 3187, 4, 2877, 17, 2, 8001, 3187, 1323, 4, 2877, 17, 8237, 19395, 4, 11091, 1650, 11, 185, 30490, 2845, 104, 27459, 1702, 15679, 1775, 75, 22721, 1039, 15679, 1775, 75, 19, 1702, 4, 1039, 12480, 9, 670, 2, 1812, 11, 103, 5, 205126, 2, 250, 98, 22, 209, 789, 3, 104, 65, 2061, 13227, 3012, 1382, 10, 103, 451, 13, 8772, 5364, 76, 6052, 170, 2227, 79295, 4136, 472, 109, 339, 40, 1781, 205127, 64, 4, 59, 25, 5, 2, 8001, 3187, 17, 185, 8, 1702, 40974, 38, 1487, 456, 1375, 4, 8159, 107, 2, 25, 10, 102, 5, 2, 8237, 4, 64585, 38, 1997, 456, 1375, 4, 8159, 4, 2606, 456, 1375, 4, 8159, 27, 2, 59, 123, 46, 8, 1039, 205128, 59, 5, 2, 3187, 17, 111, 2, 71, 11, 570, 5, 2, 8237, 17, 79296, 2, 25, 3, 5765, 5, 2, 8001, 3187, 17, 111, 53, 102, 2, 25, 205129, 2, 11091, 17, 27, 2, 59, 205130, 3, 2, 8001, 3187, 17, 8, 1702, 1039, 19, 12480, 301, 35, 247, 2, 12338, 3, 8002, 111, 126, 10, 29, 247, 5, 2, 59, 71, 51, 46, 8, 1702, 205131, 162, 73, 6, 2, 2845, 185, 63, 38, 1487, 720, 1375, 4, 8520, 5, 8237, 4, 64585, 126, 10, 29, 7258, 578, 3, 8002, 59, 13, 37, 2, 536, 12480, 46, 38, 1997, 1375, 4, 8520, 4, 2606, 1375, 4, 8520, 72, 1781, 18511, 455, 72, 18133, 3, 2576, 131, 5, 2277, 1572, 16, 72, 1781, 18511, 455, 72, 18133, 3, 2576, 131, 5, 2277, 1572, 12244, 25, 3, 2576, 541, 5, 8001, 3187, 4, 2877, 17, 2, 8001, 3187, 205132, 8237, 79297, 4, 11091, 79298, 14700, 8, 549, 2845, 104, 27459, 1702, 15679, 1775, 75, 22721, 1039, 15679, 1775, 75, 19, 1702, 4, 1039, 12480, 9, 670, 2832, 133, 79299, 10, 103, 6, 2693, 64, 25, 7, 70, 205133, 446, 762, 42, 10, 48, 6, 1200, 59, 71, 7, 70, 16965, 4, 205134, 4670, 1023, 36974, 3, 96, 59, 2974, 10, 422, 7, 70, 205135, 64, 25, 10, 1449, 6, 15679, 3087, 2209, 4, 64, 11639, 22, 3981, 15, 4559, 374, 6, 2, 2845, 104, 1257, 10, 1755, 3189, 71, 9, 1639, 5, 2845, 8237, 696, 205136, 480, 2051, 15679, 1, 464, 225, 3, 10, 2326, 34, 12, 132, 69, 2, 918, 7628, 464, 249, 9, 1639, 5, 14, 24, 205137, 1812, 3, 64, 4, 59, 32, 11, 103, 4, 22, 98, 15, 209, 65, 2061, 13227, 3012, 1382, 5044, 451, 13, 8772, 5364, 76, 6052, 170, 2227, 8, 9765, 4136, 472, 109, 339, 40, 1781, 205138, 3, 1702, 4, 1039, 20, 2, 809, 4, 25, 71, 3, 205139, 156, 3, 1702, 4, 2, 342, 444, 1039, 10, 392, 20, 2, 25, 4, 809, 3, 342, 5, 2, 8001, 3187, 4, 2877, 17, 125, 2, 349, 3, 79296, 2, 25, 3, 342, 268, 5, 2, 8001, 3187, 17, 107, 49, 166, 926, 270, 2, 59, 71, 38, 1342, 193, 456, 4, 720, 126, 10, 56, 2277, 809, 40975, 5, 2, 8001, 3187, 17, 4, 2143, 3, 2, 926, 84, 224, 156, 20, 2, 809, 71, 38, 1342, 456, 2, 25, 3, 342, 5, 2, 2877, 17, 2377, 8, 7, 560, 205140, 17, 4, 29, 247, 5, 11091, 17, 38, 2362, 193, 456, 4, 1375, 4, 38, 3275, 193, 456, 4, 1375, 68040, 2, 4480, 24, 280, 84, 2277, 809, 3, 2, 342, 39820, 4, 46, 8, 1039, 270, 2, 39820, 71, 5, 8237, 38, 2362, 456, 4, 720, 4, 11091, 38, 3275, 456, 31913, 34, 792, 14, 2, 1039, 46, 10, 205141, 3, 2, 1639, 4017, 4, 318, 342, 809, 20, 2600, 3, 205142, 103, 13, 64583, 33, 115, 14, 318, 3, 2, 342, 8, 4749, 228, 3, 1639, 142, 18972, 144, 88, 5989, 17, 21, 7, 632, 167038, 1132, 6, 7, 11854, 1422, 1005, 8, 2, 5374, 205143, 4, 2, 1029, 3, 31030, 4, 35901, 241, 30, 18513, 6, 68041, 2, 8001, 3187, 4, 2, 2877, 17, 502, 31914, 83, 21, 7, 2277, 1132, 6, 10613, 205144, 1132, 162, 185, 8, 1702, 4, 1039, 2, 10289, 32, 12, 115, 5, 38, 4, 2, 59, 25, 32, 12, 115, 5, 38, 9, 79300, 3187, 17, 126, 10, 7, 560, 5, 25, 3, 35898, 20546, 4, 21086, 8, 205145, 38, 2005, 193, 456, 4, 8159, 4, 38, 2510, 193, 456, 4, 1375, 2, 59, 9, 35898, 10, 205146, 2, 8001, 3187, 17, 2, 25, 3, 205147, 4, 31027, 102, 8, 1702, 205148, 8, 1039, 4, 12480, 38, 2005, 720, 1375, 584, 4, 8520, 4, 38, 2510, 720, 4, 584, 2, 205149, 956, 35, 4013, 39, 2, 205150, 57426, 4, 205151, 2114, 4, 8838, 59, 372, 13, 47, 166, 54, 276, 53, 12, 35, 402, 34, 2310, 12, 205152, 27, 2, 59, 123, 126, 10, 7, 560, 5, 25, 3, 31030, 5, 2, 64580, 17, 38, 2005, 15896, 4, 38, 2510, 193, 4, 8159, 5, 2, 8237, 17, 2, 25, 71, 3, 205153, 64586, 8, 315, 926, 10, 320, 6, 2, 8001, 3187, 17, 38, 2601, 79301, 14080, 3190, 205154, 8, 2, 6366, 3, 35898, 59, 34, 10, 238, 5, 2, 8237, 17, 10397, 25, 270, 8, 1702, 4, 12480, 46, 5, 2, 11091, 17, 2, 25, 501, 64586, 10, 320, 6, 8237, 8, 2, 6366, 3, 31027, 38, 3148, 205155, 46, 79302, 270, 2, 25, 3, 31027, 2, 59, 71, 9, 2, 49, 2, 64586, 10, 11701, 2, 268, 123, 8, 2, 6366, 3, 20546, 31030, 4, 31027, 20546, 89, 9996, 5, 25, 8, 1039, 4, 12480, 46, 4, 31027, 34, 89, 9996, 5, 25, 8, 1039, 38, 5242, 79301, 5, 2, 8237, 4, 11091, 17, 31030, 4047, 10, 296, 107, 2, 59, 71, 11, 570, 21, 49, 166, 205156, 2601, 15896, 38, 3148, 15896, 4, 38, 3190, 193, 8520, 38, 5242, 193, 8520, 72, 1781, 18511, 455, 72, 18133, 3, 2576, 131, 5, 2277, 1572, 12244, 25, 4, 809, 3, 1474, 128, 178, 245, 2, 8001, 3187, 205157, 8237, 205158, 4, 11091, 205159, 14700, 8, 549, 2845, 104, 27459, 1702, 15679, 1775, 75, 22721, 1039, 15679, 1775, 75, 19, 1702, 4, 1039, 12480, 9, 670, 2832, 133, 79299, 10, 103, 6, 2693, 64, 25, 7, 70, 271, 446, 762, 42, 10, 48, 6, 1200, 59, 71, 7, 70, 205160, 4, 2, 79303, 1023, 36974, 3, 96, 59, 2974, 10, 422, 205161, 70, 4, 205162, 4, 1375, 205163, 10, 35, 249, 5, 2, 8001, 205164, 259, 64, 25, 10, 1449, 6, 15679, 3087, 2209, 4, 64, 4, 59, 22, 3981, 15, 4559, 374, 6, 2, 2845, 104, 205165, 3, 64, 4, 59, 32, 11, 103, 4, 22, 98, 15, 209, 65, 2061, 13227, 3012, 1382, 10, 103, 451, 1, 5364, 76, 6052, 170, 2227, 8, 9765, 4136, 472, 109, 339, 40, 1781, 205167, 3, 1639, 228, 4, 342, 318, 20, 25, 38703, 159, 3876, 4, 19718, 4, 2, 1824, 205168, 8, 632, 205169, 19718, 50, 52, 2524, 5, 2, 167, 3, 632, 483, 205170, 240, 4, 228, 3, 1639, 12, 115, 6, 7822, 2, 25, 3, 19718, 30, 241, 613, 2, 141, 3, 1639, 228, 4, 342, 318, 20, 2, 25, 3, 19718, 5, 8001, 3187, 4, 2877, 17, 74, 60, 1368, 14, 1702, 205171, 59, 237, 2, 3187, 4, 8237, 17, 179, 2, 268, 4, 59, 10, 33364, 2, 11091, 17, 21, 1702, 327, 46, 8, 1039, 15, 184, 15, 12480, 270, 2, 72, 1781, 18511, 455, 72, 18133, 3, 2576, 131, 5, 2277, 1572, 16, 72, 1781, 18511, 455, 72, 18133, 3, 2576, 131, 5, 2277, 1572, 12244, 64, 25, 3, 28127, 2, 8001, 3187, 205172, 8237, 205173, 4, 11091, 79304, 11, 185, 30490, 2845, 104, 27459, 1702, 15679, 1775, 75, 22721, 1039, 15679, 1775, 75, 19, 1702, 4, 1039, 12480, 9, 670, 2832, 205174, 42, 10, 103, 6, 2693, 64, 25, 7, 70, 79304, 64, 25, 10, 1449, 6, 15679, 3087, 2209, 13331, 3981, 15, 4559, 374, 6, 2, 2845, 104, 3954, 1812, 3, 64, 71, 11, 103, 4, 27461, 15, 209, 65, 2061, 13227, 3012, 1382, 10, 103, 451, 13, 8772, 5364, 76, 6052, 170, 2227, 79295, 4136, 472, 109, 339, 40, 1781, 205175, 3, 19718, 5, 2, 8001, 3187, 38, 3749, 193, 1375, 4, 584, 8237, 38, 5161, 193, 1375, 4, 584, 79292, 17, 38, 5324, 193, 1375, 4, 205176, 178, 3876, 12, 568, 59, 79, 8, 88, 1572, 205177, 5, 35899, 541, 90, 118, 126, 10, 7, 560, 5, 2, 25, 3, 3876, 697, 8001, 3187, 17, 8, 49, 926, 38, 3749, 456, 1375, 4, 8159, 5, 2, 8237, 17, 2, 25, 3, 3876, 10, 186, 4, 46, 8, 1702, 102, 117, 25, 107, 46, 79305, 4, 12480, 270, 117, 25, 38, 5161, 456, 1375, 4, 8159, 2, 25, 3, 3876, 5, 64585, 102, 27, 2, 268, 123, 8, 1702, 4, 12480, 926, 107, 2, 59, 12687, 270, 162, 73, 6, 2, 2845, 104, 38, 5324, 456, 1375, 4, 205178, 25, 3, 25741, 7, 64, 79, 8, 2, 2277, 64, 25, 1573, 10, 44, 392, 5, 2, 8001, 3187, 4, 2, 24689, 17, 9, 2, 64580, 4, 8237, 17, 126, 10, 7, 560, 5, 2, 25, 3, 25741, 5, 17, 185, 79305, 4, 12480, 38, 3749, 4, 5161, 720, 1375, 4, 8520, 9, 11091, 1039, 4, 12480, 926, 205179, 59, 38, 5324, 1375, 4, 205180, 3, 25741, 19718, 4, 3876, 237, 160, 2706, 205181, 8237, 11091, 4, 35900, 205182, 8237, 4, 11091, 22, 291, 6, 26, 3, 2, 2277, 211, 1593, 179, 35900, 26549, 291, 6, 26, 3, 2, 2576, 211, 1593, 3, 1572, 16, 17, 241, 205183, 6, 1845, 2, 5, 434, 25, 3, 25741, 19718, 4, 3876, 237, 2, 79306, 35899, 1201, 3, 160, 9615, 21, 8001, 8237, 4, 11091, 17, 39821, 6, 2, 25, 5, 160, 9615, 21, 2, 2576, 35900, 17, 49, 166, 4856, 198, 11, 955, 237, 2, 79306, 1201, 3, 160, 598, 21, 79300, 8237, 4, 11091, 17, 38, 7941, 9566, 12024, 13810, 23554, 4, 18322, 2016, 7, 246, 25, 7598, 5, 2, 9903, 1201, 3, 2, 8001, 8237, 4, 11091, 160, 38, 7941, 205184, 13810, 23554, 4, 1, 1, 8839, 126, 10, 186, 6, 56, 323, 5, 2, 35900, 17, 9, 205186, 4, 3876, 38, 10186, 15343, 4, 205187, 634, 3, 1929, 105, 315, 1239, 50, 2381, 5, 2, 216, 113, 575, 3, 2, 58, 8, 2, 2277, 1593, 1234, 7, 1835, 588, 162, 73, 6, 52553, 1593, 2, 211, 428, 275, 5, 2, 167, 3, 47, 205188, 35, 1694, 907, 107, 2, 114, 3, 213, 1120, 159, 12, 907, 205189, 7873, 29, 223, 114, 5, 2, 228, 3, 2, 2576, 281, 54, 90, 118, 205190, 7873, 7, 114, 5, 2, 167, 3, 2, 2277, 1593, 3, 1929, 5, 207, 30491, 144, 88, 5989, 17, 571, 14, 228, 3, 1639, 8, 29, 4017, 1157, 68042, 3, 367, 172, 231, 2, 342, 131, 55, 6425, 17, 21, 7, 205191, 1132, 6, 7, 83, 11854, 1422, 1132, 349, 205192, 342, 547, 4, 1639, 3818, 571, 1438, 60, 390, 213, 205193, 5, 434, 15, 184, 15, 390, 66, 240, 124, 20, 47, 57, 30, 4176, 9904, 226, 2, 8001, 3187, 4, 2, 24689, 8001, 24, 280, 14, 22, 72, 1781, 18511, 455, 72, 18133, 3, 2576, 131, 5, 2277, 1572, 16, 72, 1781, 18511, 455, 72, 18133, 3, 2576, 131, 5, 2277, 1572, 12244, 59, 25, 3, 28127, 2, 8001, 3187, 205194, 8237, 79297, 4, 11091, 79298, 11, 185, 8, 549, 2845, 104, 27459, 1702, 15679, 1775, 75, 22721, 1039, 15679, 1775, 75, 19, 1702, 4, 1039, 12480, 9, 670, 446, 762, 42, 10, 48, 6, 1200, 59, 71, 4845, 8214, 271, 4, 2, 79303, 1023, 36974, 3, 96, 59, 2974, 10, 422, 205195, 70, 4, 205196, 59, 71, 22, 3981, 15, 4559, 374, 6, 2, 2845, 205197, 1812, 3, 59, 32, 11, 103, 4, 22, 98, 15, 209, 65, 2061, 13227, 3012, 1382, 10, 103, 451, 13, 8772, 5364, 1, 170, 2227, 8, 9765, 4136, 472, 109, 339, 40, 1781, 205199, 1304, 15, 7, 2277, 1593, 3, 552, 129, 11226, 6, 7, 2576, 205200, 204, 162, 185, 8, 7, 1639, 4017, 822, 29, 342, 444, 15, 23, 129, 922, 17272, 588, 3, 2, 14570, 8001, 3187, 17, 162, 3563, 5, 258, 9037, 319, 54, 14, 22, 2231, 2, 2277, 1593, 5, 23, 31, 46, 3, 47, 17, 8, 2, 1639, 4017, 205201, 2, 25, 3, 1639, 15, 184, 15, 8002, 111, 35, 5765, 5, 19720, 88, 5989, 17, 46, 8, 2, 1639, 4017, 603, 29, 247, 5, 2, 25, 3, 2, 386, 178, 5765, 4, 23, 12, 5423, 6, 1983, 12, 968, 5, 74, 31, 47, 72616, 26, 201, 6, 2, 24, 204, 645, 30, 22, 37, 9530, 17, 2, 205202, 17, 89, 29, 997, 5, 2, 25, 3, 1639, 4, 5765, 162, 185, 79302, 107, 2, 25, 3, 8002, 5, 47, 4480, 280, 10, 35, 704, 5, 40976, 280, 57, 13, 1512, 33, 115, 14, 33, 115, 14, 8002, 4, 5765, 40977, 1639, 228, 6, 3260, 2, 483, 3, 7, 2277, 1572, 16, 1593, 6, 7, 205203, 2576, 1593, 66, 57, 33, 44, 1154, 2, 25, 3, 5765, 1635, 167, 3, 2, 2576, 1593, 3, 5989, 16, 309, 5, 74, 57, 9182, 1702, 301, 433, 5, 2, 483, 3, 2, 24689, 17, 124, 935, 3528, 295, 4, 380, 2, 25, 3, 1639, 15, 184, 15, 5765, 101, 3, 34, 22, 205204, 6, 3260, 2576, 483, 3, 1572, 16, 24, 205205, 231, 2, 1639, 131, 12, 992, 9, 2, 128, 2757, 4, 3734, 483, 3, 4208, 4, 66, 144, 423, 309, 4, 16, 17, 309, 205206, 231, 2, 342, 245, 778, 7, 114, 5, 24, 172, 2, 28127, 695, 29, 42494, 5, 2, 483, 3, 423, 17, 4, 97, 21422, 1411, 22, 238, 27, 97, 946, 3, 483, 2, 144, 1572, 50, 166, 97, 946, 3, 24690, 47, 946, 22, 2350, 13, 2, 25, 3, 21086, 20546, 4, 35901, 21086, 17132, 5, 7, 2109, 3, 2277, 17, 14, 22, 2713, 6, 695, 7, 114, 5, 6070, 15, 184, 205207, 663, 66, 2277, 17, 4, 3523, 17, 5, 2, 144, 205208, 20546, 2, 25, 3, 35901, 12, 3119, 6, 2, 106, 1422, 17, 205209, 17, 47, 483, 946, 3, 2, 1572, 22, 1937, 13, 1572, 25744, 508, 2550, 55, 1562, 5, 224, 3, 2, 97, 24, 324, 3, 2, 1572, 2600, 3, 21086, 12, 968, 5, 2, 497, 3, 2, 1422, 160, 4, 117, 25, 3162, 6, 405, 266, 21086, 663, 2, 25, 3, 35901, 12, 3119, 6, 1422, 1572, 240, 4, 117, 25, 12, 79, 8, 878, 266, 7, 31, 77747, 64583, 89, 14, 144, 88, 5989, 17, 5, 763, 2299, 21086, 205210, 2, 25, 3, 31030, 4, 35901, 228, 3, 1639, 5, 47, 17, 380, 2600, 3, 31030, 4, 270, 2, 25, 3, 21086, 23, 156, 10, 77748, 162]\n"
+     ]
+    }
+   ],
+   "source": [
+    "def tokenize(text):\n",
+    "    return text.lower().split()\n",
+    "\n",
+    "word_counts = Counter()\n",
+    "for text in X_train:\n",
+    "    word_counts.update(tokenize(text))\n",
+    "\n",
+    "sorted_words = [word for word, _ in word_counts.most_common()]\n",
+    "\n",
+    "vocab = {\"<pad>\": 0, \"<unk>\": 1}\n",
+    "vocab.update({word: idx + 2 for idx, word in enumerate(sorted_words)})\n",
+    "\n",
+    "tokenizer = Tokenizer(WordLevel(vocab, unk_token=\"<unk>\"))\n",
+    "tokenizer.pre_tokenizer = Whitespace()\n",
+    "\n",
+    "def text_to_sequence(texts):\n",
+    "    return [tokenizer.encode(text.lower()).ids for text in texts]\n",
+    "\n",
+    "X_train_seq = text_to_sequence(X_train)\n",
+    "X_test_seq = text_to_sequence(X_test)\n",
+    "\n",
+    "# Display the first sequence\n",
+    "print(X_train_seq[0])\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "a5fbedc8",
+   "metadata": {
+    "id": "a5fbedc8",
+    "papermill": {
+     "duration": 0.011214,
+     "end_time": "2024-08-08T15:26:35.539595",
+     "exception": false,
+     "start_time": "2024-08-08T15:26:35.528381",
+     "status": "completed"
+    },
+    "tags": []
+   },
+   "source": [
+    "## 6.Sequence Padding and Length Adjustment"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "id": "j8v22pCcMkKW",
+   "metadata": {
+    "colab": {
+     "base_uri": "https://localhost:8080/"
+    },
+    "id": "j8v22pCcMkKW",
+    "outputId": "c6f7c315-fe5a-436e-9b3c-de0e32aec153"
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Padded Training Sequences:\n",
+      "tensor([[    0,     0,     0,  ...,    10, 77748,   162],\n",
+      "        [    0,     0,     0,  ...,    67,  1040,  1383],\n",
+      "        [    0,     0,     0,  ...,  4300,     1,  1383],\n",
+      "        ...,\n",
+      "        [    0,     0,     0,  ...,  2942,  2742,  1383],\n",
+      "        [    0,     0,     0,  ..., 98495, 14944, 58599],\n",
+      "        [    0,     0,     0,  ..., 28107,  1040, 29357]])\n",
+      "\n",
+      "Padded Testing Sequences:\n",
+      "tensor([[     0,      0,      0,  ...,     19,   1729,   1383],\n",
+      "        [     0,      0,      0,  ...,     32,     42,   1383],\n",
+      "        [     0,      0,      0,  ...,  15124,   6281,   3539],\n",
+      "        ...,\n",
+      "        [     0,      0,      0,  ...,     86,   8647,   1383],\n",
+      "        [     0,      0,      0,  ...,      4,    217, 194767],\n",
+      "        [     0,      0,      0,  ...,    155,      4,   3127]])\n"
+     ]
+    }
+   ],
+   "source": [
+    "max_len = max([len(seq) for seq in X_train_seq])\n",
+    "\n",
+    "X_train_seq = [torch.tensor(seq) for seq in X_train_seq]\n",
+    "X_test_seq = [torch.tensor(seq) for seq in X_test_seq]\n",
+    "\n",
+    "X_train_pad = pad_sequence(X_train_seq, batch_first=True, padding_value=0,padding_side=\"left\")\n",
+    "X_test_pad = pad_sequence(X_test_seq, batch_first=True, padding_value=0,padding_side=\"left\")\n",
+    "\n",
+    "# Output\n",
+    "print(\"Padded Training Sequences:\")\n",
+    "print(X_train_pad)\n",
+    "\n",
+    "print(\"\\nPadded Testing Sequences:\")\n",
+    "print(X_test_pad)\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "f0281378",
+   "metadata": {
+    "execution": {
+     "iopub.execute_input": "2024-08-08T13:32:07.711479Z",
+     "iopub.status.busy": "2024-08-08T13:32:07.710553Z",
+     "iopub.status.idle": "2024-08-08T13:32:07.717413Z",
+     "shell.execute_reply": "2024-08-08T13:32:07.716327Z",
+     "shell.execute_reply.started": "2024-08-08T13:32:07.711437Z"
+    },
+    "id": "f0281378",
+    "papermill": {
+     "duration": 0.011855,
+     "end_time": "2024-08-08T15:26:37.668786",
+     "exception": false,
+     "start_time": "2024-08-08T15:26:37.656931",
+     "status": "completed"
+    },
+    "tags": []
+   },
+   "source": [
+    "## 7.One-Hot Encoding"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "id": "pojEi95sTJrq",
+   "metadata": {
+    "colab": {
+     "base_uri": "https://localhost:8080/"
+    },
+    "id": "pojEi95sTJrq",
+    "outputId": "6e0564f1-cf89-4208-ce3b-31cef684e8bf"
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "One-Hot Encoded Training Labels:\n",
+      "tensor([[1, 0, 0],\n",
+      "        [0, 0, 1],\n",
+      "        [0, 1, 0],\n",
+      "        ...,\n",
+      "        [0, 1, 0],\n",
+      "        [1, 0, 0],\n",
+      "        [0, 0, 1]])\n",
+      "\n",
+      "One-Hot Encoded Testing Labels:\n",
+      "tensor([[0, 1, 0],\n",
+      "        [0, 1, 0],\n",
+      "        [0, 0, 1],\n",
+      "        ...,\n",
+      "        [0, 1, 0],\n",
+      "        [0, 0, 1],\n",
+      "        [0, 0, 1]])\n"
+     ]
+    }
+   ],
+   "source": [
+    "label_encoder = LabelEncoder()\n",
+    "y_train_ = label_encoder.fit_transform(y_train)\n",
+    "y_test_ = label_encoder.transform(y_test)\n",
+    "\n",
+    "# Convert to PyTorch tensors\n",
+    "y_train_tensor = torch.tensor(y_train_)\n",
+    "y_test_tensor = torch.tensor(y_test_)\n",
+    "num_classes = 3\n",
+    "\n",
+    "y_train_cat = F.one_hot(y_train_tensor, num_classes=num_classes)\n",
+    "y_test_cat = F.one_hot(y_test_tensor, num_classes=num_classes)\n",
+    "\n",
+    "# Output\n",
+    "print(\"One-Hot Encoded Training Labels:\")\n",
+    "print(y_train_cat)\n",
+    "\n",
+    "print(\"\\nOne-Hot Encoded Testing Labels:\")\n",
+    "print(y_test_cat)\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "a8eb740b",
+   "metadata": {
+    "id": "a8eb740b",
+    "papermill": {
+     "duration": 0.012226,
+     "end_time": "2024-08-08T15:26:37.776293",
+     "exception": false,
+     "start_time": "2024-08-08T15:26:37.764067",
+     "status": "completed"
+    },
+    "tags": []
+   },
+   "source": [
+    "## 8.RNN Architecture"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "id": "a23abe59-d803-4a3d-b4be-fe07ca1aa250",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "RNNModel(\n",
+      "  (embedding): Embedding(224577, 128)\n",
+      "  (rnn): RNN(128, 128, batch_first=True, dropout=0.2)\n",
+      "  (fc): Linear(in_features=128, out_features=3, bias=True)\n",
+      ")\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/aravind/myenv/lib/python3.12/site-packages/torch/nn/modules/rnn.py:123: UserWarning: dropout option adds dropout after all but last recurrent layer, so non-zero dropout expects num_layers greater than 1, but got dropout=0.2 and num_layers=1\n",
+      "  warnings.warn(\n"
+     ]
+    }
+   ],
+   "source": [
+    "vocab_size = len(tokenizer.get_vocab())     \n",
+    "embedding_dim = 128                         \n",
+    "hidden_units = 128                          \n",
+    "num_classes = 3                             \n",
+    "max_len = max_len\n",
+    "\n",
+    "class RNNModel(nn.Module):\n",
+    "    def __init__(self, vocab_size, embedding_dim, hidden_units, num_classes):\n",
+    "        super(RNNModel, self).__init__()\n",
+    "        self.embedding = nn.Embedding(vocab_size, embedding_dim)\n",
+    "        self.rnn = nn.RNN(embedding_dim, hidden_units, batch_first=True, dropout=0.2)\n",
+    "        self.fc = nn.Linear(hidden_units, num_classes)\n",
+    "\n",
+    "    def forward(self, x):\n",
+    "        x = self.embedding(x)\n",
+    "        output, _ = self.rnn(x)\n",
+    "        x = output[:, -1, :] \n",
+    "        x = self.fc(x)\n",
+    "        return F.softmax(x, dim=1)\n",
+    "\n",
+    "\n",
+    "model = RNNModel(vocab_size, embedding_dim, hidden_units, num_classes)\n",
+    "\n",
+    "print(model)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "99b14295",
+   "metadata": {
+    "id": "99b14295",
+    "papermill": {
+     "duration": 0.012565,
+     "end_time": "2024-08-08T15:26:38.661353",
+     "exception": false,
+     "start_time": "2024-08-08T15:26:38.648788",
+     "status": "completed"
+    },
+    "tags": []
+   },
+   "source": [
+    "## 9.Compile the Model"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "id": "QGbg8QVzulyP",
+   "metadata": {
+    "id": "QGbg8QVzulyP"
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "cuda\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "RNNModel(\n",
+       "  (embedding): Embedding(224577, 128)\n",
+       "  (rnn): RNN(128, 128, batch_first=True, dropout=0.2)\n",
+       "  (fc): Linear(in_features=128, out_features=3, bias=True)\n",
+       ")"
+      ]
+     },
+     "execution_count": 16,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "optimizer = optim.Adam(model.parameters(), lr=0.001)\n",
+    "criterion = nn.CrossEntropyLoss()\n",
+    "def accuracy(y_pred, y_true):\n",
+    "    _, predicted = torch.max(y_pred, 1)  # Get class with max probability\n",
+    "    correct = (predicted == y_true).sum().item()\n",
+    "    return correct / y_true.size(0)\n",
+    "\n",
+    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
+    "print(device)\n",
+    "model.to(device)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "44bd65c1-b36a-4204-8073-cad73ee815a6",
+   "metadata": {},
+   "source": [
+    "## 10.Training"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "id": "Sdzi9huhvfK4",
+   "metadata": {
+    "id": "Sdzi9huhvfK4"
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 1/10, Loss: 0.6014, Accuracy: 0.9510\n",
+      "Epoch 2/10, Loss: 0.5979, Accuracy: 0.9541\n",
+      "Epoch 3/10, Loss: 0.5966, Accuracy: 0.9549\n",
+      "Epoch 4/10, Loss: 0.5944, Accuracy: 0.9571\n",
+      "Epoch 5/10, Loss: 0.5965, Accuracy: 0.9554\n",
+      "Epoch 6/10, Loss: 0.5992, Accuracy: 0.9521\n",
+      "Epoch 7/10, Loss: 0.5936, Accuracy: 0.9577\n",
+      "Epoch 8/10, Loss: 0.5927, Accuracy: 0.9587\n",
+      "Epoch 9/10, Loss: 0.5924, Accuracy: 0.9589\n",
+      "Epoch 10/10, Loss: 0.5924, Accuracy: 0.9589\n",
+      "Time taken for training 115.47041082382202 sec\n"
+     ]
+    }
+   ],
+   "source": [
+    "\n",
+    "y_train_labels = torch.argmax(y_train_cat, dim=1)  # Convert one-hot to class index\n",
+    "\n",
+    "# Convert to TensorDataset\n",
+    "train_dataset = TensorDataset(X_train_pad, y_train_labels)\n",
+    "train_loader = DataLoader(train_dataset, batch_size=16, shuffle=True)\n",
+    "\n",
+    "# Training loop\n",
+    "num_epochs = 10\n",
+    "t0 = time.time()\n",
+    "\n",
+    "for epoch in range(num_epochs):\n",
+    "    model.train()  # Set to training mode\n",
+    "    running_loss = 0.0\n",
+    "    correct_preds = 0\n",
+    "    total_samples = 0\n",
+    "\n",
+    "    for inputs, labels in train_loader:\n",
+    "        inputs, labels = inputs.to(device), labels.to(device)\n",
+    "\n",
+    "        optimizer.zero_grad()  # Reset gradients\n",
+    "        outputs = model(inputs)  # Forward pass\n",
+    "\n",
+    "        loss = criterion(outputs, labels)  # Compute loss\n",
+    "        loss.backward()  # Backpropagation\n",
+    "        optimizer.step()  # Update weights\n",
+    "\n",
+    "        # Compute accuracy\n",
+    "        _, predicted = torch.max(outputs, 1)\n",
+    "        correct_preds += (predicted == labels).sum().item()\n",
+    "        total_samples += labels.size(0)\n",
+    "\n",
+    "        running_loss += loss.item()\n",
+    "\n",
+    "    # Compute epoch loss and accuracy\n",
+    "    epoch_loss = running_loss / len(train_loader)\n",
+    "    epoch_acc = correct_preds / total_samples\n",
+    "\n",
+    "    print(f\"Epoch {epoch+1}/{num_epochs}, Loss: {epoch_loss:.4f}, Accuracy: {epoch_acc:.4f}\")\n",
+    "print(\"Time taken for training\",time.time()-t0,\"sec\")\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "92b2ba57",
+   "metadata": {
+    "id": "92b2ba57",
+    "papermill": {
+     "duration": 0.076899,
+     "end_time": "2024-08-08T15:32:26.322204",
+     "exception": false,
+     "start_time": "2024-08-08T15:32:26.245305",
+     "status": "completed"
+    },
+    "tags": []
+   },
+   "source": [
+    "## 11. Saving Model Weights"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "id": "44159482-1028-427d-9155-7e97473a3834",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Model weights saved to rnn_classification_model_weights.pth\n"
+     ]
+    }
+   ],
+   "source": [
+    "model_weights_path = 'rnn_classification_model_weights.pth'\n",
+    "\n",
+    "torch.save(model.state_dict(), model_weights_path)\n",
+    "\n",
+    "print(f\"Model weights saved to {model_weights_path}\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "id": "67399f88-0c26-4915-9d6f-96385e010912",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "try:\n",
+    "    state_dict = torch.load('rnn_classification_model_weights.pth', map_location=torch.device('cuda'))\n",
+    "    model.load_state_dict(state_dict)\n",
+    "    model.eval()\n",
+    "except Exception as e:\n",
+    "    st.error(f\"Error loading model: {str(e)}\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "id": "df87cdce-251a-4e77-9317-9faaf3895593",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "RNNModel(\n",
+       "  (embedding): Embedding(224577, 128)\n",
+       "  (rnn): RNN(128, 128, batch_first=True, dropout=0.2)\n",
+       "  (fc): Linear(in_features=128, out_features=3, bias=True)\n",
+       ")"
+      ]
+     },
+     "execution_count": 23,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "model.to(device)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "id": "xQB-r-EqSIQE",
+   "metadata": {
+    "colab": {
+     "base_uri": "https://localhost:8080/",
+     "height": 442
+    },
+    "id": "xQB-r-EqSIQE",
+    "outputId": "8c3514f9-06c1-4483-9ab8-21ea7dfa05bf"
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Thyroid surgery in  children in a single institution from Osama Ibrahim Almosallama Ali Aseerib Ahmed Alhumaida Ali S AlZahranic Saif Alsobhib Saud AlShanafeybFrom the  aDepartment of Surgery College of Medicine Qassim University Buraidah Al Qassim Saudi Arabia  bDepartment of Surgery King Faisal Specialist Hospital and Research Center Riyadh Saudi Arabia cDepartment of Medicine King Faisal Specialist Hospital and Research Center Riyadh Saudi Arabia Correspondence   Dr  Osama Ibrahim  Almosallam  Department of  Surgery  College  of  Medicine Qassim  University  PO  Box   Buraidah  Al  Qassim   Saudi  Arabia  osama_iaahotmailcom ORCID orcid0000000290367564        Citation  Almosallam OI Aseeri A Alhumaid A AlZahrani AS Alsobhi S AlShanafey S Thyroid surgery in  children  in  a  single  institution  from   Ann  Saudi  Med         Received January  Accepted May  Published August  Copyright Copyright   Annals of Saudi Medicine Saudi Arabia This is an   access  under the Creative Commons AttributionNonCommercialNoDerivatives  International License CC BYNCND The details of which can be accessed at httpcreativecommons licensesbyncnd40Funding NoneBACKGROUND Data on thyroid surgery in children are scarceOBJECTIVE Analyze outcome data on thyroid surgery in a pediatric populationDESIGN Medical record reviewSETTING Tertiary health care institutionPATIENTS AND METHODS We collected demographic and clinical data on patients  years or younger who had thyroid surgery in the period  to  Descriptive data are presentedMAIN OUTCOME MEASURES Indications for thyroidectomy thyroid pathology complications length of stay and radioactive iodine treatment and recurrencesSAMPLE SIZE RESULTS Of  patients who underwent  thyroidectomy procedures   were females and the mean age at operation was  years and   were associated with multiple endocrine neoplasia type  There was no history of radiation exposure Eightyone patients  had fine needle aspiration FNA which correlated with the final histopathology in  of cases Sixtysix patients  had malignant cancer  papillary   of  patients who had neck dissection had lymph node metastasis and   had distant metastases to the lung Procedures included total thyroidectomy  hemithyroidectomy  completion  and subtotal thyroidectomy  Twentythree  patients    developed  hypocalcemia    permanent and   had unilateral recurrent laryngeal nerve injury  permanent Patients were followed up for a mean duration of  months median  months Of  patients with thyroid cancer   received radioactive iodine and   had recurrence  Malignancy  is  the  commonest  indication  for  thyroid surgery  in  children  and  FNA  is  highly  diagnostic  Hypocalcemia  and recurrent laryngeal nerve injury are significant complications The recurrence rate in thyroid cancer is  LIMITATIONS RetrospectiveCONFLICT OF INTEREST Noneoriginal ANN SAUDI MED  JULYAUGUST WWWANNSAUDIMEDNET 0cThyroid  diseases  requiring  surgery  are  relatively uncommon in children compared to adults The prevalence  of  palpable  thyroid  nodules  in  children  ranges  from    to    Sporadic  welldifferentiated thyroid cancer is the most common endocrine malignancy in children accounting for  of pediatric cancers  in  the  prepubertal  age  group  and  up  to   of cancers in adolescents aged “ year2 The most common indication for thyroid surgery in children varies among published studies but thyroidectomy for malignant conditions is rising38 Data in children throughout the  world  are  relatively  scarce  The  objective  of  this study was to analyze the clinical data and outcome of thyroid  surgery  in  a  large  series  of  children  treated  at a  single  center  at  King  Faisal  Specialist  Hospital  and Research Center KFSHRC in RiyadhPATIENT AND METHODS With the approval of the Institutional Review Board IRB at KFSHRC the medical records of all patients  years old and younger who underwent a thyroid surgery between    and    were  retrospectively  reviewed We elected to include patients up to the year  to ensure a reasonable followup period Patients for the study were identified by a search of the operating room log for all procedures involving the thyroid gland for the specified age groupDemographic  data  clinical  features  and  surgical outcomes  were  collected  Specific  data  that  were  obtained included age at operation gender family history presenting  symptoms  history  of  radiation  exposure presence of multiple endocrine neoplasia type  MEN   thyroid  function  test  presence  and  size  of  thyroid nodules by ultrasound presence of lymph nodes metastasis or distant metastasis fine needle aspiration FNA cytology  surgical  procedure  final  histopathology  and length of followup Outcomes analyzed were postoperative complications including transient or permanent hypocalcemia  transient  or  permanent  recurrent  laryngeal nerve paralysis wound infection and hematoma length of stay and radioactive iodine treatment and recurrences Thyroid procedures in this series included hemithyroidectomy  subtotal  total  and  completion  thyroidectomy Surgeries were performed by either an endocrine adult surgeon or a pediatric surgeon No intraoperative nerve monitoring was used Early in the series procedures were performed by adult endocrine surgeons but lately a combined approach was adopted where pediatric surgeons and adult endocrine surgeons collaborated in such cases  proceduresthe normal range in our laboratory regardless of symptoms Transient hypocalcemia was identified if it lasted for less than  months while permanent hypocalcemia was  considered  if  the  serum  calcium  level  remained below normal range and the patient continued on calcium  supplementation  after    months  of  the  surgery All patients with a family history of MEN  underwent genetic testing of the RET protooncogene to confirm the diagnosis All patients who underwent completion thyroidectomy  had  a  preoperative  and  postoperative vocal  cords  assessment  at  the  Otolaryngology  clinic Descriptive data were generated and comparisons were conducted using the t test for continuous data and the chisquare or Fisher exact tests for proportionsRESULTSBetween    and      patients  underwent   surgical  procedures    patients  underwent  two  procedures  for  thyroid  disease  at  our  institution  Eighty  patients  were females The mean age at operation was  years median  years range  years The most common indication for thyroidectomy was thyroid nodule which was present in   of cases Table  The mean SD size of thyroid nodules was   mm There were  cases  associated with MEN  syndromes The final pathology in two patients with MEN syndrome showed medullary thyroid cancer MTC while the remaining  patients had prophylactic procedures before development of MTC None of the patients had a history of radiation exposure Eightyone patients  FNA  which  correlated  with  the  final  histopathology  in  of cases There were three cases of toxic adenoma and one case of Graves™ disease which did not require FNA  The  remaining  cases  underwent  FNA  at  another institution and FNA was not repeated at our institution or they came for completion thyroidectomy with documented  pathology  for  malignancy  after  they  had  their first surgery in another hospitalThe  most  common  diagnoses  included  papillary thyroid cancer  and multinodular goiter or colloid Table  Indications for thyroidectomy in  patients IndicationNodulen  MEN  prophylaxisHyperthyroidismMultinodular goiterCompletion thyroidectomy    Hypocalcemia was defined by calcium levels below Data are number  original PEDIATRIC THYROID SURGERYANN SAUDI MED  JULYAUGUST WWWANNSAUDIMEDNET 0cnodule  Table  Surgical procedures included total thyroidectomy  hemithyroidectomy  completion  thyroidectomy    and  subtotal  thyroidectomy  Neck dissection was performed in   patients  Operative  complications  were  observed in    patients    The  most  common  complication was hypocalcemia  transient  permanent and Table  Thyroid pathology in the  patientsPathologyn  BenignNormal thyroid tissueColloid noduleCystAdenomaThyroiditisGraves™ diseaseThyroid cancerPapillaryFollicularMedullaryHurthleAnaplasticTotalData are number              Table  Benign and malignant lesions in  patientsBenignn37Malignantn66 P value  Age meanyearsGender malefemalePresence of noduleHypocalcemiaRecurrent laryngeal nerve palsyBleedinghematomaWound infectionTracheal injuryOverall complicationsMean length of stay daysMEN recurrent laryngeal nerve palsy  transient  permanent all were unilateral Table  Of  patients with malignant lesions   had lymph node metastasis and  patients  had distant metastases to the lung  None  of  the  patients  developed  postoperative bleeding wound infection or tracheal injury Patients were followed up for a mean of  months median   range    months  radioactive  iodine  treatment  was  delivered  to    patients    with  malignant  lesions    patients    had  recurrences   were  local  recurrences  and    were  local  and  distant recurrences to the lung Three cases received radioactive iodine RAI before and after recurrence One case was low risk before recurrence so did not receive RAI until after recurrence One case had medullary thyroid cancer  so  did  not  receive  RAI  In  the  remaining  five cases  there was no clear data whether those patients received RAI before or only after a recurrence All local recurrences  underwent  resection  except  for  one  patient who was lost follow up There was no mortality in this study DISCUSSIONThe most common indication for thyroidectomy in this series was thyroid nodule which correlates with previously published reports in the pediatric population35 Children with thyroid nodules have an estimated fourfold  higher  risk  of  developing  thyroid  cancer  compared to adults910 The high incidence of malignancy in this series suggests children with a thyroid nodule should be carefully evaluatedFNA is a valuablemethod for preoperative evaluation of thyroid nodules However there are limitations on  the  routine  use  of  FNA  in  children  including  the need  for  sedation  sampling  errors  and  the  limited availability  of  experienced  cytopathologists11  Many previous studies reported high sensitivity and specificity of FNA in evaluating thyroid nodule in children1114 which correlate with our findingsOur data showed lymph node metastasis in  of  thyroid  cancer  cases  which  supports  the  notion that  children  with  thyroid  cancer  frequently  present with more extensive disease than adults Lymphnode involvement  at  diagnosis  is  seen  in    to    of children  compared  with    to    of  adults  with differentiated thyroid cancer1523 Because our hospital is the largest referral center in Saudi Arabia especially for oncology cases this may explain the large number of lymph node and distant metastasis In this cohortThe  most  common  complication  reported  after thyroidectomy in children is hypoparathyroidism with an incidence ranging between  to  which original PEDIATRIC THYROID SURGERYANN SAUDI MED  JULYAUGUST WWWANNSAUDIMEDNET 0ccorresponds with our results of  which are reported as hypocalcemia in Table    One study found that total thyroidectomy central and bilateral neck dissection Graves™ disease and malignancy were  risk  factors  for  hypocalcemia  after  thyroid  surgery3 In this cohort postoperative hypocalcemia was noted more in malignant cases but it failed to reach statistical significance Moreover there was no significant difference between benign and malignant cases in  terms  of  mean  age  gender  distribution  recurrent laryngeal nerve injury or overall complications a finding that was reported previously26 Multiple studies in recent years have found an inverse relationship between surgeon volume and complication  rates2728  but  similar  data  in  the  pediatric population is lacking One study found that highvolume  endocrine  surgeons  have  better  outcomes  and shorter lengths of stay and lower costs after thyroidectomy and parathyroidectomy in children compared to pediatric surgeons general surgeons or otolaryngologists29  Scheumann  and  colleagues  also  concluded that a collaborative approach between pediatric and endocrine surgeons would have better outcomes This has led other authors to suggest that a combined approach with endocrine and pediatric surgeons in addition to pediatric endocrinologists may optimize the care of children with surgical thyroid disease given the low number of pediatric patients4 Our data do not allow for comparisons of different approaches given the late adoption of the combined approach The  recurrence  rate  for  thyroid  cancer  in  children after  thyroidectomy  has  varied  widely  in  reported studies  ranging  from    to    while  it  was   in  this  cohort  Only  a  few  studies  explored  the predictors  of  recurrence  Lymph  node  involvement multiple nodules male gender younger age histologic subtype and advanced tumor stage were risk factors associated with recurrence17233033 In  this  study    of  patients  with  malignant  lesions received RAI Although there are conflicting data regarding  the  indications  of  postoperative  RAI  treatment in lowrisk patients the current recommendation is that lowrisk patients can be treated without RAI3436There are some limitations to this study The retrospective nature may affect the validity and quality of the data The small number of cases in some categories did not enable us to compare groups and explore predictors relative to these factors On the other hand this study adds to the scarce data on thyroid surgery in pediatric age group Malignancy is the commonest indication  for  thyroid  surgery  in  children  and  FNA  is highly  diagnostic  Hypocalcemia  and  recurrent  laryngeal nerve injury are significant complications Cancerrelated  death  is  extremely  rare  but  recurrence  is  not uncommon and a significant number of patients with malignant cases received RAI treatmentoriginal PEDIATRIC THYROID SURGERYANN SAUDI MED  JULYAUGUST WWWANNSAUDIMEDNET 0cREFERENCES  Trowbridge  FL  Matovinovic  J  McLaren GD Nichaman MZ Iodine and goiter in children Pediatrics  Ries LAG Melbert D Krapcho M Stinchcomb  DG  Howlader  N  Horner  MJ  et  al SEER  Cancer  Statistics  Review  “ Bethesda  National  Cancer  Institute   Based  on  November    SEER  data  submission Chen Y[h] Masiakos PT Gaz RD Hodin RA Parangi S Randolph GW et al Pediatric thyroidectomy in a high volume thyroid surgery center Risk factors for postoperative hypocalcemia J Pediatr Surg  Aug5081316   Wood  JH  Partrick  DA  Barham  HP  Bensard DD Travers HS Bruny JL et al Pediatric thyroidectomy  a  collaborative  surgical  approach J Pediatr Surg  May4658238  Scholz S Smith JR Chaignaud B Shamberger  RC  Huang  SA  Thyroid  surgery  at Children™s Hospital Boston a 35year singleinstitution  experience  J  Pediatr  Surg   Mar46343742   Josefson  J  Zimmerman  D  Thyroid  nodules and cancers in children Pediatr Endocrinol Rev  Sep611423 Hameed R Zacharin MR  Changing face of paediatric and adolescent thyroid cancer J Paediatr Child Health   LugoVicente  H  Ortiz  VN  Irizarry  H Camps  JI  Pagán  V  Pediatric  thyroid  nodules management in the era of fine needle aspirationJ Pediatr Surg  Mussa A De Andrea M Motta M Mormile A  Palestini  N  Corrias  A  Predictors  of  Malignancy in Children with Thyroid Nodules J Pediatr  Oct167488692 Amirazodi E Propst EJ Chung CT Parra DA  Wasserman  JD  Pediatric  thyroid  FNA biopsy  Outcomes  and  impact  on  management over  years at a tertiary care center Cancer Cytopathol    Partyka  KL  Huang  EC2  Cramer  HM Chen  S  Wu  HH  Histologic  and  clinical followup  of  thyroid  fineneedle  aspirates in  pediatric  patients  Cancer  Cytopathol  Sinha CK Decoppi P Pierro A Brain C Hindmarsh P Butler G et al Thyroid Surgery in Children Clinical Outcomes Eur J Pediatr Surg  Oct2554259 Kundel A Thompson GB Richards ML Qiu LX Cai Y Schwenk FW et al Pediatric Endocrine Surgery A 20Year Experience at the  Mayo  Clinic  J  Clin  Endocrinol  Metab February  “ Jiang W Newbury RO Newfield RS Pediatric  thyroid  surgery  and  management  of thyroid  nodulesan  institutional  experience features  and over a 10year period Int J Pediatr Endocrinol  Burke JF Sippel RS Chen H Evolution of Pediatric Thyroid Surgery at a Tertiary Medical Center Surg Res   “  AlQahtani  KH  Tunio  MA  Al  Asiri  M Aljohani NJ Bayoumi Y Riaz K et al Clinicopathological treatment outcomes  of  differentiated  thyroid  cancer in  Saudi  children  and  adults  J  Otolaryngol Head Neck Surg  Nov  Kluijfhout WP van Beek DJ Verrijn Stuart  AA  Lodewijk  L  Valk  GD  Van  der  Zee DC  et  al  Postoperative  Complications After  Prophylactic  Thyroidectomy  for  Very Young  Patients  With  Multiple  Endocrine Neoplasia  Type    Medicine  Baltimore 20159429e1108  Raval MV Browne M Chin AC Zimmerman D Angelos P Reynolds M Total thyroidectomy  for  benign  disease  in  the  pediatric patient”feasible  and  safe  J  Pediatr  Surg  Stavrakis AI Ituarte PH Ko CY Yeh MW Surgeon volume as a predictor of outcomes in  inpatient  and  outpatient  endocrine  surgery Surgery “  Sosa JA Bowman HM Tielsch JM Powe NR  Gordon  TA  Udelsman  R  The  importance of surgeon experience for clinical and economic  outcomes  from  thyroidectomy Ann Surg “  Tuggle CT  Roman SA Wang TS Boudourakis  L  Thomas  D  Udelsman  R  et  al Pediatric  endocrine  surgery  Who  is  operating  on  our  children  Surgery   Dec144686977  Park  S  Jeong  JS  Ryu  HR  Lee  C  Park JH Kang S et al Differentiated Thyroid Carcinoma of Children and Adolescents27Year Experience  in  the  Yonsei  University  Health System J Korean Med Sci   Palmer BA Zarroug AE Poley RN Kollars  JP  Moir  CR  Papillary  thyroid  carcinoma in children risk factors and complications  of  disease  recurrence  J  Pediatr  Surg  Wada N Sugino K Mimura T Nagahama M  Kitagawa  W  Shibuya  H  et  al  Pediatric differentiated  thyroid  carcinoma  in  stage  I risk  factor  analysis  for  disease  free  survival BMC Cancer    D Danese Gardini A Farsetti A Sciacchitano  S  Andreoli  M  Pontecorvi  A  Thyroid carcinoma in children and adolescents Eur J Pediatr   Astl  J  Chovanec  M  Lukes  P  Katra  R Dvorakova M Vlcek P et al Thyroid carcinoma surgery in children and adolescents “  years experience surgery of pediatric thyroid lymph  node  metastases carcinoma  Int  J  Pediatr  Otorhinolaryngol   Chaukar  DA  Rangarajan  V  Nair  N Nadkarni  MS  Pai  PS  Dcruz  AK  et  al  Pediatric  thyroid  cancer  J  Surg  Oncol     Dzodic  R  Buta  M  Markovic  I  Gavrilo D Matovic M Milovanovic Z et al Surgical management  of  welldifferentiated  thyroid carcinoma  in  children  and  adolescents   years of experience of a single institution in Serbia Endocr J   Scheumann  GF  Gimm  O  Wegener G  Hundeshagen  H  Dralle  H  Prognostic significance  and  surgical  management  of locoregional in papillary thyroid cancer World J Surg   Shi RL Qu N Yang SW Tumor size interpretation for predicting cervical lymph node metastasis using a differentiated thyroid cancer risk model Onco Targets Ther   “ Zimmerman D Hay ID Gough IR Goellner JR Ryan JJ Grant CS et al Papillary thyroid carcinoma in children and adults longterm  followup  of    patients  conservatively treated at one institution during three decades Surgery   Collini  P  Mattavelli  F  Pellegrinelli  A Barisella M Ferrari A Massimino M Papillary carcinoma of the thyroid gland of childhood and adolescence Morphologic subtypes biologic behavior and prognosis a clinicopathologic study of  sporadic cases treated at a single institution during a 30year period Am J Surg Pathol   BorsonChazot  Causeret  S  Lifante  JC Augros M Berger N Peix JL Predictive factors for recurrence from a series of  children and  adolescents  with  differentiated  thyroid cancer World J Surg   Baumgarten HD Bauer AJ Isaza A MostoufiMoab S Kazahaya K Adzick NS Surgical management of pediatric thyroid disease Complication  rates  after  thyroidectomy  at the Children™s Hospital of Philadelphia highvolume Pediatric Thyroid Center Journal of pediatric  surgery    Oct   Kurzawinski TR De Coppi P Thyroidectomy in Children InPediatric Surgery  pp  Springer Berlin Heidelberg  Francis  G  Waguespack  SG  Bauer  AJ Angelog P Benvenga S et al Management Guidelines  for  Children  with  Thyroid  Nodules  and  Differentiated  Thyroid  Cancer  The American  Thyroid  Association  Guidelines Task Force on Pediatric Thyroid Cancer THYROID Volume  Number  original PEDIATRIC THYROID SURGERYANN SAUDI MED  JULYAUGUST WWWANNSAUDIMEDNET 0c'\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(texts[0])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "7727274b-19a0-4506-be03-8f9977bff825",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "accelerator": "GPU",
+  "colab": {
+   "gpuType": "T4",
+   "provenance": []
+  },
+  "kaggle": {
+   "accelerator": "gpu",
+   "dataSources": [
+    {
+     "datasetId": 2389764,
+     "sourceId": 4033428,
+     "sourceType": "datasetVersion"
+    }
+   ],
+   "isGpuEnabled": true,
+   "isInternetEnabled": true,
+   "language": "python",
+   "sourceType": "notebook"
+  },
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.12.3"
+  },
+  "papermill": {
+   "default_parameters": {},
+   "duration": 458.547826,
+   "end_time": "2024-08-08T15:32:43.096410",
+   "environment_variables": {},
+   "exception": null,
+   "input_path": "__notebook__.ipynb",
+   "output_path": "__notebook__.ipynb",
+   "parameters": {},
+   "start_time": "2024-08-08T15:25:04.548584",
+   "version": "2.5.0"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}