641 lines
51 KiB
Plaintext
641 lines
51 KiB
Plaintext
{
|
|
"cells": [
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"[nltk_data] Downloading package punkt to /home/akeon/nltk_data...\n",
|
|
"[nltk_data] Package punkt is already up-to-date!\n",
|
|
"[nltk_data] Downloading package averaged_perceptron_tagger to\n",
|
|
"[nltk_data] /home/akeon/nltk_data...\n",
|
|
"[nltk_data] Package averaged_perceptron_tagger is already up-to-\n",
|
|
"[nltk_data] date!\n",
|
|
"[nltk_data] Downloading package maxent_ne_chunker to\n",
|
|
"[nltk_data] /home/akeon/nltk_data...\n",
|
|
"[nltk_data] Package maxent_ne_chunker is already up-to-date!\n",
|
|
"[nltk_data] Downloading package words to /home/akeon/nltk_data...\n",
|
|
"[nltk_data] Package words is already up-to-date!\n"
|
|
]
|
|
},
|
|
{
|
|
"data": {
|
|
"text/plain": [
|
|
"True"
|
|
]
|
|
},
|
|
"execution_count": 18,
|
|
"metadata": {},
|
|
"output_type": "execute_result"
|
|
}
|
|
],
|
|
"source": [
|
|
"import json\n",
|
|
"import string, re, pickle\n",
|
|
"import numpy as np\n",
|
|
"from nltk.tokenize import word_tokenize\n",
|
|
"from tensorflow.keras.preprocessing.text import Tokenizer\n",
|
|
"from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
|
|
"from sklearn.model_selection import train_test_split\n",
|
|
"from tensorflow.keras.models import Model\n",
|
|
"from tensorflow.keras.layers import Input, Embedding, LSTM, Bidirectional, TimeDistributed, Dense\n",
|
|
"import spacy\n",
|
|
"import nltk\n",
|
|
"\n",
|
|
"\n",
|
|
"nltk.download ('punkt')\n",
|
|
"nltk.download ('averaged_perceptron_tagger')\n",
|
|
"nltk.download ('maxent_ne_chunker')\n",
|
|
"nltk.download ('words')"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Kerajaan Aceh PER\n",
|
|
"Iskandar Muda PER\n",
|
|
"ke-17 MISC\n",
|
|
"Islam MISC\n",
|
|
"Nusantara LOC\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"import spacy\n",
|
|
"\n",
|
|
"# Load model multibahasa yang mendukung Indonesia\n",
|
|
"nlp = spacy.load(\"xx_ent_wiki_sm\")\n",
|
|
"\n",
|
|
"# Contoh teks\n",
|
|
"text = \"Kerajaan Aceh mencapai puncak kejayaannya di bawah pemerintahan Sultan Iskandar Muda pada abad ke-17. Aceh menjadi pusat perdagangan dan kebudayaan Islam di wilayah Nusantara.\"\n",
|
|
"\n",
|
|
"# Proses teks dengan model\n",
|
|
"doc = nlp(text)\n",
|
|
"\n",
|
|
"# Cetak entitas yang dikenali\n",
|
|
"for ent in doc.ents:\n",
|
|
" print(ent.text, ent.label_)\n",
|
|
" \n",
|
|
"\n",
|
|
"# def generate_ner_context(text):\n",
|
|
"# # Load the pretrained spaCy model (small Indo model or use multilingual model if needed)\n",
|
|
"# nlp = spacy.load(\"xx_ent_wiki_sm\") # Load multilingual model\n",
|
|
" \n",
|
|
"# # Process the text\n",
|
|
"# doc = nlp(text)\n",
|
|
" \n",
|
|
"# # Tokenization and Named Entity Recognition (NER)\n",
|
|
"# tokens = [token.text for token in doc]\n",
|
|
"# ner_tags = []\n",
|
|
"# for token in doc:\n",
|
|
"# if token.ent_type_:\n",
|
|
"# ner_tags.append(f\"B-{token.ent_type_}\")\n",
|
|
"# else:\n",
|
|
"# ner_tags.append(\"O\")\n",
|
|
" \n",
|
|
"# return tokens, ner_tags\n",
|
|
"\n",
|
|
"# # Example input context\n",
|
|
"# context = \"Perang Diponegoro berlangsung dari tahun 1825 hingga 1830. Perang ini dipimpin oleh Pangeran Diponegoro melawan pemerintah kolonial Belanda di Jawa Tengah.\"\n",
|
|
"\n",
|
|
"# # Generate NER and tokens\n",
|
|
"# tokens, ner_tags = generate_ner_context(context)\n",
|
|
"\n",
|
|
"# # Construct the JSON result\n",
|
|
"# result = {\n",
|
|
"# \"context\": context,\n",
|
|
"# \"context_tokens\": tokens,\n",
|
|
"# \"context_ner\": ner_tags,\n",
|
|
"# \"question_posibility\": [\n",
|
|
"# {\n",
|
|
"# \"type\": \"true_false\",\n",
|
|
"# \"question\": \"Perang Diponegoro berlangsung selama lima tahun.\",\n",
|
|
"# \"answer\": \"True\"\n",
|
|
"# },\n",
|
|
"# {\n",
|
|
"# \"type\": \"true_false\",\n",
|
|
"# \"question\": \"Perang Diponegoro berakhir pada tahun 1850.\",\n",
|
|
"# \"answer\": \"False\"\n",
|
|
"# }\n",
|
|
"# ]\n",
|
|
"# }\n",
|
|
"\n",
|
|
"# # Output the result\n",
|
|
"# import json\n",
|
|
"# print(json.dumps(result, indent=4, ensure_ascii=False))"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 2,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"dataset = [\n",
|
|
" {\n",
|
|
" \"context\": \"Pertempuran Surabaya terjadi pada 10 November 1945 antara pasukan Indonesia melawan pasukan sekutu Inggris yang berusaha mengambil alih kota setelah Jepang menyerah dalam Perang Dunia II. Pertempuran ini dikenang sebagai Hari Pahlawan di Indonesia.\",\n",
|
|
" \"context_tokens\": [\n",
|
|
" \"Pertempuran\", \"Surabaya\", \"terjadi\", \"pada\", \"10\", \"November\", \"1945\",\n",
|
|
" \"antara\", \"pasukan\", \"Indonesia\", \"melawan\", \"pasukan\", \"sekutu\", \"Inggris\",\n",
|
|
" \"yang\", \"berusaha\", \"mengambil\", \"alih\", \"kota\", \"setelah\", \"Jepang\", \"menyerah\",\n",
|
|
" \"dalam\", \"Perang\", \"Dunia\", \"II\", \".\", \"Pertempuran\", \"ini\", \"dikenang\", \"sebagai\",\n",
|
|
" \"Hari\", \"Pahlawan\", \"di\", \"Indonesia\", \".\"\n",
|
|
" ],\n",
|
|
" \"context_ner\": [\n",
|
|
" \"O\", \"B-LOC\", \"O\", \"O\", \"B-DATE\", \"I-DATE\", \"I-DATE\",\n",
|
|
" \"O\", \"O\", \"B-LOC\", \"O\", \"O\", \"O\", \"B-LOC\",\n",
|
|
" \"O\", \"O\", \"O\", \"O\", \"O\", \"O\", \"B-LOC\", \"O\",\n",
|
|
" \"O\", \"B-MISC\", \"I-MISC\", \"I-MISC\", \"O\", \"O\", \"O\", \"O\", \"O\",\n",
|
|
" \"O\", \"O\", \"O\", \"B-LOC\", \"O\"\n",
|
|
" ],\n",
|
|
" \"question_posibility\": [\n",
|
|
" {\n",
|
|
" \"type\": \"fill_in_the_blank\",\n",
|
|
" \"question\": \"Pertempuran Surabaya terjadi pada tanggal _______.\",\n",
|
|
" \"answer\": \"10 November 1945\"\n",
|
|
" },\n",
|
|
" {\n",
|
|
" \"type\": \"multiple_choice\",\n",
|
|
" \"question\": \"Pasukan yang dihadapi Indonesia dalam Pertempuran Surabaya berasal dari negara apa?\",\n",
|
|
" \"options\": [\"Jepang\", \"Belanda\", \"Inggris\", \"Australia\"],\n",
|
|
" \"answer\": \"Inggris\"\n",
|
|
" },\n",
|
|
" {\n",
|
|
" \"type\": \"true_false\",\n",
|
|
" \"question\": \"Pertempuran Surabaya diperingati sebagai Hari Pahlawan di Indonesia.\",\n",
|
|
" \"answer\": \"True\"\n",
|
|
" }\n",
|
|
" ]\n",
|
|
" },\n",
|
|
" {\n",
|
|
" \"context\": \"Perang Diponegoro berlangsung dari tahun 1825 hingga 1830. Perang ini dipimpin oleh Pangeran Diponegoro melawan pemerintah kolonial Belanda di Jawa Tengah.\",\n",
|
|
" \"context_tokens\": [\n",
|
|
" \"Perang\", \"Diponegoro\", \"berlangsung\", \"dari\", \"tahun\", \"1825\", \"hingga\", \"1830\", \".\",\n",
|
|
" \"Perang\", \"ini\", \"dipimpin\", \"oleh\", \"Pangeran\", \"Diponegoro\", \"melawan\",\n",
|
|
" \"pemerintah\", \"kolonial\", \"Belanda\", \"di\", \"Jawa\", \"Tengah\", \".\"\n",
|
|
" ],\n",
|
|
" \"context_ner\": [\n",
|
|
" \"O\", \"B-PER\", \"O\", \"O\", \"O\", \"B-DATE\", \"O\", \"B-DATE\", \"O\",\n",
|
|
" \"O\", \"O\", \"O\", \"O\", \"B-PER\", \"I-PER\", \"O\",\n",
|
|
" \"O\", \"O\", \"B-LOC\", \"O\", \"O\", \"B-LOC\", \"O\"\n",
|
|
" ],\n",
|
|
" \"question_posibility\": [\n",
|
|
" {\n",
|
|
" \"type\": \"true_false\",\n",
|
|
" \"question\": \"Perang Diponegoro berlangsung selama lima tahun.\",\n",
|
|
" \"answer\": \"True\"\n",
|
|
" },\n",
|
|
" {\n",
|
|
" \"type\": \"true_false\",\n",
|
|
" \"question\": \"Perang Diponegoro berakhir pada tahun 1850.\",\n",
|
|
" \"answer\": \"False\"\n",
|
|
" }\n",
|
|
" ]\n",
|
|
" }\n",
|
|
"]"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 3,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"contexts_tokens = []\n",
|
|
"contexts_ner = []\n",
|
|
"questions = []\n",
|
|
"answers = []\n",
|
|
"qtypes = []\n",
|
|
"\n",
|
|
"for entry in dataset:\n",
|
|
" contexts_tokens.append(entry[\"context_tokens\"])\n",
|
|
" contexts_ner.append(entry[\"context_ner\"])\n",
|
|
" qa = entry[\"question_posibility\"][0] # pilih soal pertama\n",
|
|
" questions.append(qa[\"question\"])\n",
|
|
" answers.append(qa[\"answer\"])\n",
|
|
" qtypes.append(qa[\"type\"]) # misalnya \"fill_in_the_blank\"\n",
|
|
"\n",
|
|
"# ----------------------------\n",
|
|
"# Tokenisasi untuk Kata\n",
|
|
"# ----------------------------\n",
|
|
"# Kita gabungkan semua teks dari context (dari tokens), pertanyaan, dan jawaban\n",
|
|
"all_texts = []\n",
|
|
"for tokens in contexts_tokens:\n",
|
|
" all_texts.append(\" \".join(tokens))\n",
|
|
"all_texts += questions\n",
|
|
"all_texts += answers\n",
|
|
"\n",
|
|
"tokenizer = Tokenizer(oov_token=\"<OOV>\")\n",
|
|
"tokenizer.fit_on_texts(all_texts)\n",
|
|
"\n",
|
|
"# Ubah context_tokens menjadi sequence angka\n",
|
|
"context_sequences = [tokenizer.texts_to_sequences([\" \".join(tokens)])[0] for tokens in contexts_tokens]\n",
|
|
"question_sequences = tokenizer.texts_to_sequences(questions)\n",
|
|
"answer_sequences = tokenizer.texts_to_sequences(answers)\n",
|
|
"\n",
|
|
"# Padding sequence ke panjang tetap\n",
|
|
"MAX_LENGTH = 50 # sesuaikan dengan panjang teks maksimum yang diinginkan\n",
|
|
"context_padded = pad_sequences(context_sequences, maxlen=MAX_LENGTH, padding=\"post\", truncating=\"post\")\n",
|
|
"question_padded = pad_sequences(question_sequences, maxlen=MAX_LENGTH, padding=\"post\", truncating=\"post\")\n",
|
|
"answer_padded = pad_sequences(answer_sequences, maxlen=MAX_LENGTH, padding=\"post\", truncating=\"post\")"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 4,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"# ----------------------------\n",
|
|
"# Tokenisasi untuk Label NER\n",
|
|
"# ----------------------------\n",
|
|
"# Kumpulkan semua label NER untuk membangun mapping label ke indeks\n",
|
|
"all_ner_labels = []\n",
|
|
"for ner_seq in contexts_ner:\n",
|
|
" all_ner_labels += ner_seq\n",
|
|
"\n",
|
|
"ner_labels_set = sorted(list(set(all_ner_labels)))\n",
|
|
"# Contoh: ['B-DATE', 'B-LOC', 'B-MISC', 'B-PER', 'I-DATE', 'I-MISC', 'I-PER', 'O']\n",
|
|
"ner2idx = {label: idx for idx, label in enumerate(ner_labels_set)}\n",
|
|
"idx2ner = {idx: label for label, idx in ner2idx.items()}\n",
|
|
"\n",
|
|
"# Ubah label NER ke dalam bentuk sequence angka\n",
|
|
"ner_sequences = []\n",
|
|
"for ner_seq in contexts_ner:\n",
|
|
" seq = [ner2idx[label] for label in ner_seq]\n",
|
|
" ner_sequences.append(seq)\n",
|
|
"\n",
|
|
"# Padding sequence label NER (gunakan nilai default misal label \"O\")\n",
|
|
"ner_padded = pad_sequences(ner_sequences, maxlen=MAX_LENGTH, padding=\"post\", truncating=\"post\", value=ner2idx[\"O\"])\n",
|
|
"\n",
|
|
"# ----------------------------\n",
|
|
"# Label Tipe Soal\n",
|
|
"# ----------------------------\n",
|
|
"qtype_dict = {\"fill_in_the_blank\": 0, \"true_false\": 1, \"multiple_choice\": 2}\n",
|
|
"qtype_labels = np.array([qtype_dict[q] for q in qtypes])"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 5,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"# ----------------------------\n",
|
|
"# Split Data Training dan Validation\n",
|
|
"# ----------------------------\n",
|
|
"(context_train, context_val, \n",
|
|
" question_train, question_val, \n",
|
|
" answer_train, answer_val, \n",
|
|
" qtype_train, qtype_val,\n",
|
|
" ner_train, ner_val) = train_test_split(\n",
|
|
" context_padded, question_padded, answer_padded, qtype_labels, ner_padded,\n",
|
|
" test_size=0.2, random_state=42\n",
|
|
")"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 6,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"2025-03-23 15:06:59.338033: E external/local_xla/xla/stream_executor/cuda/cuda_driver.cc:152] failed call to cuInit: INTERNAL: CUDA error: Failed call to cuInit: UNKNOWN ERROR (303)\n"
|
|
]
|
|
},
|
|
{
|
|
"data": {
|
|
"text/html": [
|
|
"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"functional\"</span>\n",
|
|
"</pre>\n"
|
|
],
|
|
"text/plain": [
|
|
"\u001b[1mModel: \"functional\"\u001b[0m\n"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
},
|
|
{
|
|
"data": {
|
|
"text/html": [
|
|
"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┓\n",
|
|
"┃<span style=\"font-weight: bold\"> Layer (type) </span>┃<span style=\"font-weight: bold\"> Output Shape </span>┃<span style=\"font-weight: bold\"> Param # </span>┃<span style=\"font-weight: bold\"> Connected to </span>┃\n",
|
|
"┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n",
|
|
"│ context_input │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">50</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ - │\n",
|
|
"│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>) │ │ │ │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ question_decoder_i… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">50</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ - │\n",
|
|
"│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>) │ │ │ │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ context_embedding │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">50</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">300</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">15,600</span> │ context_input[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
|
|
"│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Embedding</span>) │ │ │ │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ not_equal │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">50</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ context_input[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
|
|
"│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">NotEqual</span>) │ │ │ │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ question_embedding │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">50</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">300</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">15,600</span> │ question_decoder… │\n",
|
|
"│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Embedding</span>) │ │ │ │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ encoder_lstm (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">LSTM</span>) │ [(<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>), │ <span style=\"color: #00af00; text-decoration-color: #00af00\">570,368</span> │ context_embeddin… │\n",
|
|
"│ │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>), │ │ not_equal[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n",
|
|
"│ │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)] │ │ │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ question_lstm │ [(<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">50</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>), │ <span style=\"color: #00af00; text-decoration-color: #00af00\">570,368</span> │ question_embeddi… │\n",
|
|
"│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">LSTM</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>), │ │ encoder_lstm[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
|
|
"│ │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)] │ │ encoder_lstm[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ answer_lstm (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">LSTM</span>) │ [(<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">50</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>), │ <span style=\"color: #00af00; text-decoration-color: #00af00\">570,368</span> │ context_embeddin… │\n",
|
|
"│ │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>), │ │ encoder_lstm[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
|
|
"│ │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)] │ │ encoder_lstm[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ dense (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">32,896</span> │ encoder_lstm[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ ner_lstm │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">50</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1,140,736</span> │ context_embeddin… │\n",
|
|
"│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Bidirectional</span>) │ │ │ not_equal[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ question_output │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">50</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">52</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">13,364</span> │ question_lstm[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
|
|
"│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>) │ │ │ │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ answer_output │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">50</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">52</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">13,364</span> │ answer_lstm[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n",
|
|
"│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>) │ │ │ │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ question_type_outp… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">387</span> │ dense[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n",
|
|
"│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>) │ │ │ │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ ner_output │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">50</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">4,104</span> │ ner_lstm[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>], │\n",
|
|
"│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">TimeDistributed</span>) │ │ │ not_equal[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n",
|
|
"└─────────────────────┴───────────────────┴────────────┴───────────────────┘\n",
|
|
"</pre>\n"
|
|
],
|
|
"text/plain": [
|
|
"┏━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┓\n",
|
|
"┃\u001b[1m \u001b[0m\u001b[1mLayer (type) \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m Param #\u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mConnected to \u001b[0m\u001b[1m \u001b[0m┃\n",
|
|
"┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n",
|
|
"│ context_input │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m50\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ - │\n",
|
|
"│ (\u001b[38;5;33mInputLayer\u001b[0m) │ │ │ │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ question_decoder_i… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m50\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ - │\n",
|
|
"│ (\u001b[38;5;33mInputLayer\u001b[0m) │ │ │ │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ context_embedding │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m50\u001b[0m, \u001b[38;5;34m300\u001b[0m) │ \u001b[38;5;34m15,600\u001b[0m │ context_input[\u001b[38;5;34m0\u001b[0m]… │\n",
|
|
"│ (\u001b[38;5;33mEmbedding\u001b[0m) │ │ │ │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ not_equal │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m50\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ context_input[\u001b[38;5;34m0\u001b[0m]… │\n",
|
|
"│ (\u001b[38;5;33mNotEqual\u001b[0m) │ │ │ │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ question_embedding │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m50\u001b[0m, \u001b[38;5;34m300\u001b[0m) │ \u001b[38;5;34m15,600\u001b[0m │ question_decoder… │\n",
|
|
"│ (\u001b[38;5;33mEmbedding\u001b[0m) │ │ │ │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ encoder_lstm (\u001b[38;5;33mLSTM\u001b[0m) │ [(\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m), │ \u001b[38;5;34m570,368\u001b[0m │ context_embeddin… │\n",
|
|
"│ │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m), │ │ not_equal[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
|
|
"│ │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)] │ │ │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ question_lstm │ [(\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m50\u001b[0m, \u001b[38;5;34m256\u001b[0m), │ \u001b[38;5;34m570,368\u001b[0m │ question_embeddi… │\n",
|
|
"│ (\u001b[38;5;33mLSTM\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m), │ │ encoder_lstm[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
|
|
"│ │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)] │ │ encoder_lstm[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ answer_lstm (\u001b[38;5;33mLSTM\u001b[0m) │ [(\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m50\u001b[0m, \u001b[38;5;34m256\u001b[0m), │ \u001b[38;5;34m570,368\u001b[0m │ context_embeddin… │\n",
|
|
"│ │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m), │ │ encoder_lstm[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
|
|
"│ │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)] │ │ encoder_lstm[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ dense (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m) │ \u001b[38;5;34m32,896\u001b[0m │ encoder_lstm[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ ner_lstm │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m50\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m1,140,736\u001b[0m │ context_embeddin… │\n",
|
|
"│ (\u001b[38;5;33mBidirectional\u001b[0m) │ │ │ not_equal[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ question_output │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m50\u001b[0m, \u001b[38;5;34m52\u001b[0m) │ \u001b[38;5;34m13,364\u001b[0m │ question_lstm[\u001b[38;5;34m0\u001b[0m]… │\n",
|
|
"│ (\u001b[38;5;33mDense\u001b[0m) │ │ │ │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ answer_output │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m50\u001b[0m, \u001b[38;5;34m52\u001b[0m) │ \u001b[38;5;34m13,364\u001b[0m │ answer_lstm[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
|
|
"│ (\u001b[38;5;33mDense\u001b[0m) │ │ │ │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ question_type_outp… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m3\u001b[0m) │ \u001b[38;5;34m387\u001b[0m │ dense[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
|
|
"│ (\u001b[38;5;33mDense\u001b[0m) │ │ │ │\n",
|
|
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
|
|
"│ ner_output │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m50\u001b[0m, \u001b[38;5;34m8\u001b[0m) │ \u001b[38;5;34m4,104\u001b[0m │ ner_lstm[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m], │\n",
|
|
"│ (\u001b[38;5;33mTimeDistributed\u001b[0m) │ │ │ not_equal[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
|
|
"└─────────────────────┴───────────────────┴────────────┴───────────────────┘\n"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
},
|
|
{
|
|
"data": {
|
|
"text/html": [
|
|
"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">2,947,155</span> (11.24 MB)\n",
|
|
"</pre>\n"
|
|
],
|
|
"text/plain": [
|
|
"\u001b[1m Total params: \u001b[0m\u001b[38;5;34m2,947,155\u001b[0m (11.24 MB)\n"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
},
|
|
{
|
|
"data": {
|
|
"text/html": [
|
|
"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">2,947,155</span> (11.24 MB)\n",
|
|
"</pre>\n"
|
|
],
|
|
"text/plain": [
|
|
"\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m2,947,155\u001b[0m (11.24 MB)\n"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
},
|
|
{
|
|
"data": {
|
|
"text/html": [
|
|
"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
|
|
"</pre>\n"
|
|
],
|
|
"text/plain": [
|
|
"\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
},
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Epoch 1/10\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"2025-03-23 15:07:06.004502: E tensorflow/core/util/util.cc:131] oneDNN supports DT_BOOL only on platforms with AVX-512. Falling back to the default Eigen-based implementation if present.\n"
|
|
]
|
|
},
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 8s/step - answer_output_accuracy: 0.0000e+00 - answer_output_loss: 3.9473 - loss: 11.0828 - ner_output_accuracy: 0.0800 - ner_output_loss: 2.0766 - question_output_accuracy: 0.0400 - question_output_loss: 3.9452 - question_type_output_accuracy: 0.0000e+00 - question_type_output_loss: 1.1138 - val_answer_output_accuracy: 0.3200 - val_answer_output_loss: 3.9260 - val_loss: 11.0343 - val_ner_output_accuracy: 0.8600 - val_ner_output_loss: 2.0489 - val_question_output_accuracy: 0.0000e+00 - val_question_output_loss: 3.9441 - val_question_type_output_accuracy: 0.0000e+00 - val_question_type_output_loss: 1.1153\n",
|
|
"Epoch 2/10\n",
|
|
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 300ms/step - answer_output_accuracy: 0.6800 - answer_output_loss: 3.8844 - loss: 10.8637 - ner_output_accuracy: 0.7800 - ner_output_loss: 2.0194 - question_output_accuracy: 0.0800 - question_output_loss: 3.9047 - question_type_output_accuracy: 1.0000 - question_type_output_loss: 1.0550 - val_answer_output_accuracy: 0.5800 - val_answer_output_loss: 3.8962 - val_loss: 10.9915 - val_ner_output_accuracy: 0.8600 - val_ner_output_loss: 2.0227 - val_question_output_accuracy: 0.0000e+00 - val_question_output_loss: 3.9453 - val_question_type_output_accuracy: 0.0000e+00 - val_question_type_output_loss: 1.1273\n",
|
|
"Epoch 3/10\n",
|
|
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 300ms/step - answer_output_accuracy: 0.9000 - answer_output_loss: 3.8076 - loss: 10.6189 - ner_output_accuracy: 0.7800 - ner_output_loss: 1.9522 - question_output_accuracy: 0.0800 - question_output_loss: 3.8585 - question_type_output_accuracy: 1.0000 - question_type_output_loss: 1.0005 - val_answer_output_accuracy: 0.9800 - val_answer_output_loss: 3.8543 - val_loss: 10.9334 - val_ner_output_accuracy: 0.8600 - val_ner_output_loss: 1.9867 - val_question_output_accuracy: 0.0000e+00 - val_question_output_loss: 3.9469 - val_question_type_output_accuracy: 0.0000e+00 - val_question_type_output_loss: 1.1455\n",
|
|
"Epoch 4/10\n",
|
|
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 340ms/step - answer_output_accuracy: 0.9400 - answer_output_loss: 3.6877 - loss: 10.2657 - ner_output_accuracy: 0.7800 - ner_output_loss: 1.8489 - question_output_accuracy: 0.0600 - question_output_loss: 3.8010 - question_type_output_accuracy: 1.0000 - question_type_output_loss: 0.9281 - val_answer_output_accuracy: 0.9800 - val_answer_output_loss: 3.7881 - val_loss: 10.8457 - val_ner_output_accuracy: 0.8600 - val_ner_output_loss: 1.9324 - val_question_output_accuracy: 0.0000e+00 - val_question_output_loss: 3.9492 - val_question_type_output_accuracy: 0.0000e+00 - val_question_type_output_loss: 1.1760\n",
|
|
"Epoch 5/10\n",
|
|
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 330ms/step - answer_output_accuracy: 0.9400 - answer_output_loss: 3.4683 - loss: 9.6920 - ner_output_accuracy: 0.7800 - ner_output_loss: 1.6792 - question_output_accuracy: 0.0600 - question_output_loss: 3.7188 - question_type_output_accuracy: 1.0000 - question_type_output_loss: 0.8258 - val_answer_output_accuracy: 0.9800 - val_answer_output_loss: 3.6742 - val_loss: 10.7083 - val_ner_output_accuracy: 0.8600 - val_ner_output_loss: 1.8475 - val_question_output_accuracy: 0.0000e+00 - val_question_output_loss: 3.9535 - val_question_type_output_accuracy: 0.0000e+00 - val_question_type_output_loss: 1.2331\n",
|
|
"Epoch 6/10\n",
|
|
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 344ms/step - answer_output_accuracy: 0.9400 - answer_output_loss: 2.9986 - loss: 8.6406 - ner_output_accuracy: 0.7800 - ner_output_loss: 1.3997 - question_output_accuracy: 0.0400 - question_output_loss: 3.5829 - question_type_output_accuracy: 1.0000 - question_type_output_loss: 0.6593 - val_answer_output_accuracy: 0.9800 - val_answer_output_loss: 3.4580 - val_loss: 10.4731 - val_ner_output_accuracy: 0.8600 - val_ner_output_loss: 1.7102 - val_question_output_accuracy: 0.0000e+00 - val_question_output_loss: 3.9628 - val_question_type_output_accuracy: 0.0000e+00 - val_question_type_output_loss: 1.3420\n",
|
|
"Epoch 7/10\n",
|
|
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 319ms/step - answer_output_accuracy: 0.9400 - answer_output_loss: 1.9364 - loss: 6.7078 - ner_output_accuracy: 0.7800 - ner_output_loss: 1.0844 - question_output_accuracy: 0.0400 - question_output_loss: 3.3048 - question_type_output_accuracy: 1.0000 - question_type_output_loss: 0.3822 - val_answer_output_accuracy: 0.9800 - val_answer_output_loss: 2.9410 - val_loss: 10.0188 - val_ner_output_accuracy: 0.8600 - val_ner_output_loss: 1.5038 - val_question_output_accuracy: 0.0000e+00 - val_question_output_loss: 3.9871 - val_question_type_output_accuracy: 0.0000e+00 - val_question_type_output_loss: 1.5870\n",
|
|
"Epoch 8/10\n",
|
|
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 318ms/step - answer_output_accuracy: 0.9600 - answer_output_loss: 0.9184 - loss: 4.9883 - ner_output_accuracy: 0.7800 - ner_output_loss: 1.3771 - question_output_accuracy: 0.0400 - question_output_loss: 2.6239 - question_type_output_accuracy: 1.0000 - question_type_output_loss: 0.0690 - val_answer_output_accuracy: 0.9800 - val_answer_output_loss: 1.7714 - val_loss: 9.6522 - val_ner_output_accuracy: 0.8600 - val_ner_output_loss: 1.4667 - val_question_output_accuracy: 0.0000e+00 - val_question_output_loss: 4.0805 - val_question_type_output_accuracy: 0.0000e+00 - val_question_type_output_loss: 2.3336\n",
|
|
"Epoch 9/10\n",
|
|
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 286ms/step - answer_output_accuracy: 0.9600 - answer_output_loss: 0.4511 - loss: 3.5983 - ner_output_accuracy: 0.7800 - ner_output_loss: 1.3641 - question_output_accuracy: 0.0400 - question_output_loss: 1.7815 - question_type_output_accuracy: 1.0000 - question_type_output_loss: 0.0015 - val_answer_output_accuracy: 0.9800 - val_answer_output_loss: 0.8089 - val_loss: 11.4588 - val_ner_output_accuracy: 0.8600 - val_ner_output_loss: 1.5131 - val_question_output_accuracy: 0.0000e+00 - val_question_output_loss: 4.6062 - val_question_type_output_accuracy: 0.0000e+00 - val_question_type_output_loss: 4.5306\n",
|
|
"Epoch 10/10\n",
|
|
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 304ms/step - answer_output_accuracy: 0.9400 - answer_output_loss: 0.3244 - loss: 2.9690 - ner_output_accuracy: 0.7800 - ner_output_loss: 1.1538 - question_output_accuracy: 0.0600 - question_output_loss: 1.4906 - question_type_output_accuracy: 1.0000 - question_type_output_loss: 1.7498e-04 - val_answer_output_accuracy: 0.9800 - val_answer_output_loss: 0.3998 - val_loss: 16.2049 - val_ner_output_accuracy: 0.8600 - val_ner_output_loss: 1.5880 - val_question_output_accuracy: 0.0000e+00 - val_question_output_loss: 6.0197 - val_question_type_output_accuracy: 0.0000e+00 - val_question_type_output_loss: 8.1974\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"# ----------------------------\n",
|
|
"# Parameter Model\n",
|
|
"# ----------------------------\n",
|
|
"VOCAB_SIZE = len(tokenizer.word_index) + 1\n",
|
|
"EMBEDDING_DIM = 300\n",
|
|
"LSTM_UNITS = 256\n",
|
|
"BATCH_SIZE = 16\n",
|
|
"EPOCHS = 10\n",
|
|
"NUM_NER_TAGS = len(ner2idx)\n",
|
|
"\n",
|
|
"# ----------------------------\n",
|
|
"# Arsitektur Model Multi-Output\n",
|
|
"# ----------------------------\n",
|
|
"\n",
|
|
"# Encoder: Input context\n",
|
|
"context_input = Input(shape=(MAX_LENGTH,), name=\"context_input\")\n",
|
|
"context_embedding = Embedding(input_dim=VOCAB_SIZE, output_dim=EMBEDDING_DIM, mask_zero=True, name=\"context_embedding\")(context_input)\n",
|
|
"encoder_lstm = LSTM(LSTM_UNITS, return_state=True, name=\"encoder_lstm\")\n",
|
|
"encoder_output, state_h, state_c = encoder_lstm(context_embedding)\n",
|
|
"\n",
|
|
"# Branch untuk pembuatan soal (Question Decoder)\n",
|
|
"question_decoder_input = Input(shape=(MAX_LENGTH,), name=\"question_decoder_input\")\n",
|
|
"question_embedding = Embedding(input_dim=VOCAB_SIZE, output_dim=EMBEDDING_DIM, mask_zero=True, name=\"question_embedding\")(question_decoder_input)\n",
|
|
"question_lstm = LSTM(LSTM_UNITS, return_sequences=True, return_state=True, name=\"question_lstm\")\n",
|
|
"question_output, _, _ = question_lstm(question_embedding, initial_state=[state_h, state_c])\n",
|
|
"question_dense = Dense(VOCAB_SIZE, activation=\"softmax\", name=\"question_output\")(question_output)\n",
|
|
"\n",
|
|
"# Branch untuk pembuatan jawaban (Answer Decoder)\n",
|
|
"answer_lstm = LSTM(LSTM_UNITS, return_sequences=True, return_state=True, name=\"answer_lstm\")\n",
|
|
"answer_output, _, _ = answer_lstm(context_embedding, initial_state=[state_h, state_c])\n",
|
|
"answer_dense = Dense(VOCAB_SIZE, activation=\"softmax\", name=\"answer_output\")(answer_output)\n",
|
|
"\n",
|
|
"# Branch untuk klasifikasi tipe soal\n",
|
|
"type_dense = Dense(128, activation=\"relu\")(encoder_output)\n",
|
|
"question_type_output = Dense(3, activation=\"softmax\", name=\"question_type_output\")(type_dense)\n",
|
|
"\n",
|
|
"# Branch untuk NER: Menggunakan context_embedding untuk melakukan sequence tagging\n",
|
|
"ner_lstm = Bidirectional(LSTM(LSTM_UNITS, return_sequences=True, recurrent_dropout=0.1), name=\"ner_lstm\")(context_embedding)\n",
|
|
"ner_output = TimeDistributed(Dense(NUM_NER_TAGS, activation=\"softmax\"), name=\"ner_output\")(ner_lstm)\n",
|
|
"\n",
|
|
"# Gabungkan semua branch dalam satu model multi-output\n",
|
|
"model = Model(\n",
|
|
" inputs=[context_input, question_decoder_input],\n",
|
|
" outputs=[question_dense, answer_dense, question_type_output, ner_output]\n",
|
|
")\n",
|
|
"\n",
|
|
"model.compile(\n",
|
|
" optimizer=\"adam\",\n",
|
|
" loss={\n",
|
|
" \"question_output\": \"sparse_categorical_crossentropy\",\n",
|
|
" \"answer_output\": \"sparse_categorical_crossentropy\",\n",
|
|
" \"question_type_output\": \"sparse_categorical_crossentropy\",\n",
|
|
" \"ner_output\": \"sparse_categorical_crossentropy\"\n",
|
|
" },\n",
|
|
" metrics={\n",
|
|
" \"question_output\": [\"accuracy\"],\n",
|
|
" \"answer_output\": [\"accuracy\"],\n",
|
|
" \"question_type_output\": [\"accuracy\"],\n",
|
|
" \"ner_output\": [\"accuracy\"]\n",
|
|
" }\n",
|
|
")\n",
|
|
"\n",
|
|
"model.summary()\n",
|
|
"\n",
|
|
"# ----------------------------\n",
|
|
"# Training Model\n",
|
|
"# ----------------------------\n",
|
|
"model.fit(\n",
|
|
" [context_train, question_train],\n",
|
|
" {\n",
|
|
" \"question_output\": question_train,\n",
|
|
" \"answer_output\": answer_train,\n",
|
|
" \"question_type_output\": qtype_train,\n",
|
|
" \"ner_output\": ner_train\n",
|
|
" },\n",
|
|
" batch_size=BATCH_SIZE,\n",
|
|
" epochs=EPOCHS,\n",
|
|
" validation_data=(\n",
|
|
" [context_val, question_val],\n",
|
|
" {\n",
|
|
" \"question_output\": question_val,\n",
|
|
" \"answer_output\": answer_val,\n",
|
|
" \"question_type_output\": qtype_val,\n",
|
|
" \"ner_output\": ner_val\n",
|
|
" }\n",
|
|
" )\n",
|
|
")\n",
|
|
"\n",
|
|
"# Simpan model dan tokenizer bila diperlukan\n",
|
|
"model.save(\"lstm_multi_output_ner_model.keras\")\n",
|
|
"with open(\"tokenizer.pkl\", \"wb\") as handle:\n",
|
|
" pickle.dump(tokenizer, handle, protocol=pickle.HIGHEST_PROTOCOL)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": []
|
|
}
|
|
],
|
|
"metadata": {
|
|
"kernelspec": {
|
|
"display_name": "myenv",
|
|
"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": "3.10.16"
|
|
}
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 2
|
|
}
|