{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "c11e397b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "📂 Searching for documents in: /Users/przemo/Coding/document data retrieval/content\n",
      "✅ Created 8134 multilingual instruction pairs.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Map: 100%|██████████| 7320/7320 [00:01<00:00, 5442.12 examples/s]\n",
      "Map: 100%|██████████| 814/814 [00:00<00:00, 5879.32 examples/s]\n",
      "/var/folders/t6/7gm1y_yj5831dq3xh1b21y8w0000gp/T/ipykernel_31309/193414689.py:109: FutureWarning: `tokenizer` is deprecated and will be removed in version 5.0.0 for `Seq2SeqTrainer.__init__`. Use `processing_class` instead.\n",
      "  trainer = Seq2SeqTrainer(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "🚀 Training Multilingual Model...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/przemo/Coding/.venv/lib/python3.13/site-packages/torch/utils/data/dataloader.py:692: UserWarning: 'pin_memory' argument is set as true but not supported on MPS now, device pinned memory won't be used.\n",
      "  warnings.warn(warn_msg)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "    <div>\n",
       "      \n",
       "      <progress value='9150' max='9150' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
       "      [9150/9150 10:11:41, Epoch 10/10]\n",
       "    </div>\n",
       "    <table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       " <tr style=\"text-align: left;\">\n",
       "      <th>Epoch</th>\n",
       "      <th>Training Loss</th>\n",
       "      <th>Validation Loss</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.838600</td>\n",
       "      <td>0.698937</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.628400</td>\n",
       "      <td>0.617202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.598600</td>\n",
       "      <td>0.577280</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.575200</td>\n",
       "      <td>0.545364</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.488400</td>\n",
       "      <td>0.526966</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.489900</td>\n",
       "      <td>0.512095</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.447100</td>\n",
       "      <td>0.500422</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.450300</td>\n",
       "      <td>0.499637</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.396700</td>\n",
       "      <td>0.494340</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>0.402300</td>\n",
       "      <td>0.497277</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table><p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/przemo/Coding/.venv/lib/python3.13/site-packages/torch/utils/data/dataloader.py:692: UserWarning: 'pin_memory' argument is set as true but not supported on MPS now, device pinned memory won't be used.\n",
      "  warnings.warn(warn_msg)\n",
      "/Users/przemo/Coding/.venv/lib/python3.13/site-packages/torch/utils/data/dataloader.py:692: UserWarning: 'pin_memory' argument is set as true but not supported on MPS now, device pinned memory won't be used.\n",
      "  warnings.warn(warn_msg)\n",
      "/Users/przemo/Coding/.venv/lib/python3.13/site-packages/torch/utils/data/dataloader.py:692: UserWarning: 'pin_memory' argument is set as true but not supported on MPS now, device pinned memory won't be used.\n",
      "  warnings.warn(warn_msg)\n",
      "/Users/przemo/Coding/.venv/lib/python3.13/site-packages/torch/utils/data/dataloader.py:692: UserWarning: 'pin_memory' argument is set as true but not supported on MPS now, device pinned memory won't be used.\n",
      "  warnings.warn(warn_msg)\n",
      "/Users/przemo/Coding/.venv/lib/python3.13/site-packages/torch/utils/data/dataloader.py:692: UserWarning: 'pin_memory' argument is set as true but not supported on MPS now, device pinned memory won't be used.\n",
      "  warnings.warn(warn_msg)\n",
      "/Users/przemo/Coding/.venv/lib/python3.13/site-packages/torch/utils/data/dataloader.py:692: UserWarning: 'pin_memory' argument is set as true but not supported on MPS now, device pinned memory won't be used.\n",
      "  warnings.warn(warn_msg)\n",
      "/Users/przemo/Coding/.venv/lib/python3.13/site-packages/torch/utils/data/dataloader.py:692: UserWarning: 'pin_memory' argument is set as true but not supported on MPS now, device pinned memory won't be used.\n",
      "  warnings.warn(warn_msg)\n",
      "/Users/przemo/Coding/.venv/lib/python3.13/site-packages/torch/utils/data/dataloader.py:692: UserWarning: 'pin_memory' argument is set as true but not supported on MPS now, device pinned memory won't be used.\n",
      "  warnings.warn(warn_msg)\n",
      "/Users/przemo/Coding/.venv/lib/python3.13/site-packages/torch/utils/data/dataloader.py:692: UserWarning: 'pin_memory' argument is set as true but not supported on MPS now, device pinned memory won't be used.\n",
      "  warnings.warn(warn_msg)\n",
      "/Users/przemo/Coding/.venv/lib/python3.13/site-packages/torch/utils/data/dataloader.py:692: UserWarning: 'pin_memory' argument is set as true but not supported on MPS now, device pinned memory won't be used.\n",
      "  warnings.warn(warn_msg)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PL Headline: Licencja Ruchu - Podmiot - Data\n",
      "DE Summary: Dieses Dokument scheint ein Vertrag zwischen einem Unternehmen und einem Unternehmen für IT-Services zu sein.\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import torch\n",
    "import matplotlib.pyplot as plt\n",
    "from pathlib import Path\n",
    "from datasets import Dataset\n",
    "from transformers import (\n",
    "    AutoTokenizer, \n",
    "    AutoModelForSeq2SeqLM, \n",
    "    DataCollatorForSeq2Seq, \n",
    "    Seq2SeqTrainingArguments, \n",
    "    Seq2SeqTrainer\n",
    ")\n",
    "\n",
    "# --- 1. CONFIGURATION ---\n",
    "BASE_DIR = Path.cwd().parent\n",
    "DATA_ROOT = BASE_DIR / \"content\"\n",
    "# Roots for labels (titles and summaries)\n",
    "TITLE_ROOT = BASE_DIR / \"synthetic_dataset\" / \"titles\"\n",
    "SUMMARY_ROOT = BASE_DIR / \"synthetic_dataset\" / \"summary\"\n",
    "\n",
    "MODEL_ID = \"google/flan-t5-small\" # FLAN-T5 is excellent for instructions\n",
    "OUTPUT_MODEL_DIR = BASE_DIR / \"models\" / \"flan_t5_multilingual\"\n",
    "\n",
    "MAX_INPUT_LEN = 512\n",
    "MAX_TARGET_LEN = 128\n",
    "\n",
    "# Mapping language codes to full names for the prompt\n",
    "LANG_MAP = {\n",
    "    \"pl\": \"Polish\",\n",
    "    \"en\": \"English\",\n",
    "    \"de\": \"German\",\n",
    "    \"fr\": \"French\",\n",
    "    \"es\": \"Spanish\",\n",
    "    \"it\": \"Italian\",\n",
    "    \"uk\": \"Ukrainian\"\n",
    "}\n",
    "\n",
    "# --- 2. MULTILINGUAL DATA LOADING ---\n",
    "def load_multilingual_data():\n",
    "    dataset_dict = {\"input_text\": [], \"target_text\": []}\n",
    "    \n",
    "    print(f\"📂 Searching for documents in: {DATA_ROOT}\")\n",
    "    all_docs = list(DATA_ROOT.rglob(\"*.txt\"))\n",
    "    \n",
    "    for txt_file in all_docs:\n",
    "        rel_path = txt_file.relative_to(DATA_ROOT)\n",
    "        try:\n",
    "            content = txt_file.read_text(encoding=\"utf-8\").strip()\n",
    "        except: continue\n",
    "        if not content: continue\n",
    "\n",
    "        # Iterate through all available languages\n",
    "        for lang_code, lang_name in LANG_MAP.items():\n",
    "            \n",
    "            # 1. Look for TITLES (Headlines) in this language\n",
    "            t_file = TITLE_ROOT / lang_code / rel_path\n",
    "            if t_file.exists():\n",
    "                target_title = t_file.read_text(encoding=\"utf-8\").strip()\n",
    "                # PROMPT: \"headline in Polish: [content]\"\n",
    "                dataset_dict[\"input_text\"].append(f\"headline in {lang_name}: {content}\")\n",
    "                dataset_dict[\"target_text\"].append(target_title)\n",
    "            \n",
    "            # 2. Look for SUMMARIES in this language\n",
    "            s_file = SUMMARY_ROOT / lang_code / rel_path\n",
    "            if s_file.exists():\n",
    "                target_summary = s_file.read_text(encoding=\"utf-8\").strip()\n",
    "                # PROMPT: \"summarize in English: [content]\"\n",
    "                dataset_dict[\"input_text\"].append(f\"summarize in {lang_name}: {content}\")\n",
    "                dataset_dict[\"target_text\"].append(target_summary)\n",
    "                \n",
    "    return Dataset.from_dict(dataset_dict)\n",
    "\n",
    "# --- 3. PREPARATION ---\n",
    "raw_dataset = load_multilingual_data()\n",
    "if len(raw_dataset) == 0:\n",
    "    print(\"❌ No data found! Check your folder structure.\")\n",
    "else:\n",
    "    print(f\"✅ Created {len(raw_dataset)} multilingual instruction pairs.\")\n",
    "    dataset = raw_dataset.train_test_split(test_size=0.1)\n",
    "\n",
    "tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)\n",
    "model = AutoModelForSeq2SeqLM.from_pretrained(MODEL_ID)\n",
    "\n",
    "def preprocess(examples):\n",
    "    model_inputs = tokenizer(examples[\"input_text\"], max_length=MAX_INPUT_LEN, truncation=True, padding=\"max_length\")\n",
    "    labels = tokenizer(text_target=examples[\"target_text\"], max_length=MAX_TARGET_LEN, truncation=True, padding=\"max_length\")\n",
    "    model_inputs[\"labels\"] = labels[\"input_ids\"]\n",
    "    return model_inputs\n",
    "\n",
    "tokenized_dataset = dataset.map(preprocess, batched=True)\n",
    "\n",
    "# --- 4. TRAINING ---\n",
    "\n",
    "\n",
    "training_args = Seq2SeqTrainingArguments(\n",
    "    output_dir=\"./tmp_multilingual_results\",\n",
    "    eval_strategy=\"epoch\",\n",
    "    learning_rate=5e-4, # Slightly higher for multilingual tasks\n",
    "    per_device_train_batch_size=8,\n",
    "    per_device_eval_batch_size=8,\n",
    "    num_train_epochs=10,\n",
    "    predict_with_generate=True,\n",
    "    generation_max_length=MAX_TARGET_LEN,\n",
    "    generation_num_beams=4,\n",
    "    fp16=torch.cuda.is_available(), # Speed up if GPU is present\n",
    "    logging_steps=50\n",
    ")\n",
    "\n",
    "trainer = Seq2SeqTrainer(\n",
    "    model=model,\n",
    "    args=training_args,\n",
    "    train_dataset=tokenized_dataset[\"train\"],\n",
    "    eval_dataset=tokenized_dataset[\"test\"],\n",
    "    tokenizer=tokenizer,\n",
    "    data_collator=DataCollatorForSeq2Seq(tokenizer, model=model),\n",
    ")\n",
    "\n",
    "print(\"🚀 Training Multilingual Model...\")\n",
    "trainer.train()\n",
    "\n",
    "# --- 5. SAVING & TESTING ---\n",
    "model.save_pretrained(OUTPUT_MODEL_DIR)\n",
    "tokenizer.save_pretrained(OUTPUT_MODEL_DIR)\n",
    "\n",
    "def test_multilingual(text, task=\"summarize\", lang=\"English\"):\n",
    "    device = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n",
    "    model.to(device)\n",
    "    prompt = f\"{task} in {lang}: {text}\"\n",
    "    inputs = tokenizer(prompt, return_tensors=\"pt\", max_length=MAX_INPUT_LEN, truncation=True).to(device)\n",
    "    outputs = model.generate(inputs[\"input_ids\"], max_length=MAX_TARGET_LEN, num_beams=4)\n",
    "    return tokenizer.decode(outputs[0], skip_special_tokens=True)\n",
    "\n",
    "# Example Test\n",
    "test_text = \"This contract is between Company A and Company B for IT services.\"\n",
    "print(f\"PL Headline: {test_multilingual(test_text, 'headline', 'Polish')}\")\n",
    "print(f\"DE Summary: {test_multilingual(test_text, 'summarize', 'German')}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "5adc4c0f",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/t6/7gm1y_yj5831dq3xh1b21y8w0000gp/T/ipykernel_31309/3125303398.py:40: FutureWarning: \n",
      "\n",
      "Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the same effect.\n",
      "\n",
      "  sns.barplot(x=eval_df['epoch'].astype(float).astype(int), y=eval_df['eval_samples_per_second'], ax=axes[2], palette='viridis')\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x1800 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "==================================================\n",
      "SZCZEGÓŁOWE WYNIKI EWALUACJI\n",
      "==================================================\n",
      " epoch  step  eval_loss  eval_runtime  eval_samples_per_second\n",
      "   1.0   915   0.698937       35.1236                   23.175\n",
      "   2.0  1830   0.617202       34.4665                   23.617\n",
      "   3.0  2745   0.577280       34.1537                   23.833\n",
      "   4.0  3660   0.545364       18.8015                   43.294\n",
      "   5.0  4575   0.526966       82.4268                    9.875\n",
      "   6.0  5490   0.512095       18.8094                   43.276\n",
      "   7.0  6405   0.500422       20.3806                   39.940\n",
      "   8.0  7320   0.499637       38.3302                   21.236\n",
      "   9.0  8235   0.494340       43.7664                   18.599\n",
      "  10.0  9150   0.497277       33.9088                   24.006\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "# 1. Pobranie danych z historii trenera\n",
    "history = trainer.state.log_history\n",
    "df = pd.DataFrame(history)\n",
    "\n",
    "# Rozdzielenie logów treningowych i ewaluacyjnych\n",
    "train_df = df[df['loss'].notna()].copy()\n",
    "eval_df = df[df['eval_loss'].notna()].copy()\n",
    "\n",
    "# Konfiguracja stylu i rozmiaru - wysokość ustawiona na 18 cali dla 3 wykresów\n",
    "sns.set_theme(style=\"whitegrid\")\n",
    "fig, axes = plt.subplots(3, 1, figsize=(12, 18))\n",
    "\n",
    "# --- WYKRES 1: STRATA (LOSS) ---\n",
    "# Wyświetlamy surowe dane (lekka linia) i średnią kroczącą (wyraźna linia) dla lepszej czytelności\n",
    "axes[0].plot(train_df['step'], train_df['loss'], label='Train Loss (raw)', color='#1f77b4', alpha=0.3)\n",
    "axes[0].plot(train_df['step'], train_df['loss'].rolling(window=5).mean(), label='Train Loss (smoothed)', color='#1f77b4', linewidth=2)\n",
    "\n",
    "if not eval_df.empty:\n",
    "    axes[0].plot(eval_df['step'], eval_df['eval_loss'], label='Eval Loss', color='#d62728', marker='o', markersize=8, linewidth=3)\n",
    "\n",
    "axes[0].set_title('Krzywa Straty (Loss)', fontsize=16, fontweight='bold')\n",
    "axes[0].set_xlabel('Krok (Step)', fontsize=12)\n",
    "axes[0].set_ylabel('Loss', fontsize=12)\n",
    "axes[0].legend(fontsize=12)\n",
    "\n",
    "# --- WYKRES 2: LEARNING RATE ---\n",
    "axes[1].plot(train_df['step'], train_df['learning_rate'], color='#2ca02c', linewidth=2)\n",
    "axes[1].set_title('Zmiana Learning Rate (Scheduler)', fontsize=16, fontweight='bold')\n",
    "axes[1].set_xlabel('Krok (Step)', fontsize=12)\n",
    "axes[1].set_ylabel('Learning Rate', fontsize=12)\n",
    "axes[1].ticklabel_format(axis='y', style='sci', scilimits=(0,0)) # Notacja naukowa dla małych wartości\n",
    "\n",
    "# --- WYKRES 3: WYDAJNOŚĆ (SAMPLES PER SECOND) ---\n",
    "if not eval_df.empty:\n",
    "    # Używamy barplotu dla wydajności na epokę\n",
    "    sns.barplot(x=eval_df['epoch'].astype(float).astype(int), y=eval_df['eval_samples_per_second'], ax=axes[2], palette='viridis')\n",
    "    axes[2].set_title('Wydajność procesora/GPU (Próbki na sekundę)', fontsize=16, fontweight='bold')\n",
    "    axes[2].set_xlabel('Epoka', fontsize=12)\n",
    "    axes[2].set_ylabel('Samples per Second', fontsize=12)\n",
    "else:\n",
    "    axes[2].text(0.5, 0.5, 'Brak danych o wydajności ewaluacji', ha='center', fontsize=14)\n",
    "\n",
    "# Automatyczne dopasowanie marginesów, żeby napisy na siebie nie nachodziły\n",
    "plt.tight_layout(pad=3.0)\n",
    "plt.show()\n",
    "\n",
    "# Wyświetlenie tabeli z kluczowymi metrykami pod wykresami\n",
    "if not eval_df.empty:\n",
    "    print(\"\\n\" + \"=\"*50)\n",
    "    print(\"SZCZEGÓŁOWE WYNIKI EWALUACJI\")\n",
    "    print(\"=\"*50)\n",
    "    print(eval_df[['epoch', 'step', 'eval_loss', 'eval_runtime', 'eval_samples_per_second']].to_string(index=False))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python (Projekt AI)",
   "language": "python",
   "name": "projekt_ai"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.13.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}