{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "5f93b7d1",
   "metadata": {},
   "outputs": [],
   "source": [
    "from transformers import AutoModelForSeq2SeqLM\n",
    "from peft import get_peft_config,get_peft_model, get_peft_model_state_dict, LoraConfig, TaskType\n",
    "import torch\n",
    "from datasets import load_dataset\n",
    "import os\n",
    "os.environ[\"TOKENIZERS_PARALLELISM\"] = \"false\"\n",
    "from transformers import AutoTokenizer\n",
    "from torch.utils.data import DataLoader\n",
    "from transformers import default_data_collator,get_linear_schedule_with_warmup\n",
    "from tqdm import tqdm\n",
    "from datasets import load_dataset\n",
    "\n",
    "device = \"cuda\"\n",
    "model_name_or_path = \"bigscience/mt0-large\"\n",
    "tokenizer_name_or_path = \"bigscience/mt0-large\"\n",
    "\n",
    "checkpoint_name = \"financial_sentiment_analysis_lora_v1.pt\"\n",
    "text_column = \"sentence\"\n",
    "label_column = \"text_label\"\n",
    "max_length=128\n",
    "lr = 1e-3\n",
    "num_epochs = 3\n",
    "batch_size=8\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8d0850ac",
   "metadata": {},
   "outputs": [],
   "source": [
    "# creating model\n",
    "peft_config = LoraConfig(\n",
    "    task_type=TaskType.SEQ_2_SEQ_LM, inference_mode=False, r=8, lora_alpha=32, lora_dropout=0.1\n",
    ")\n",
    "\n",
    "model = AutoModelForSeq2SeqLM.from_pretrained(model_name_or_path)\n",
    "model = get_peft_model(model, peft_config)\n",
    "model.print_trainable_parameters()\n",
    "model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "4ee2babf",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/sourab/miniconda3/envs/ml/lib/python3.10/site-packages/huggingface_hub/utils/_deprecation.py:97: FutureWarning: Deprecated argument(s) used in 'dataset_info': token. Will not be supported from version '0.12'.\n",
      "  warnings.warn(message, FutureWarning)\n",
      "Found cached dataset financial_phrasebank (/home/sourab/.cache/huggingface/datasets/financial_phrasebank/sentences_allagree/1.0.0/550bde12e6c30e2674da973a55f57edde5181d53f5a5a34c1531c53f93b7e141)\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6de075f8208349108291ac5ab7f5c980",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4b0e67b6d93f43e4b0f6a2f8978e4b0c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?ba/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a9551029c9884529bda7421a99170b51",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1 [00:00<?, ?ba/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "{'sentence': 'The order was valued at USD12 .2 m.',\n",
       " 'label': 1,\n",
       " 'text_label': 'neutral'}"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# loading dataset\n",
    "dataset = load_dataset(\"financial_phrasebank\", 'sentences_allagree')\n",
    "dataset = dataset[\"train\"].train_test_split(test_size=0.1)\n",
    "dataset[\"validation\"] = dataset[\"test\"]\n",
    "del(dataset[\"test\"])\n",
    "\n",
    "classes = dataset[\"train\"].features[\"label\"].names\n",
    "dataset = dataset.map(\n",
    "    lambda x: {\"text_label\": [classes[label] for label in x[\"label\"]]},\n",
    "    batched=True,\n",
    "    num_proc=1,\n",
    "    \n",
    ")\n",
    "\n",
    "dataset[\"train\"][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "adf9608c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4421971232434db1b6141e91fda2f6d7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Running tokenizer on dataset:   0%|          | 0/3 [00:00<?, ?ba/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9b2ef793d93443949f4a5d5874d4bc05",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Running tokenizer on dataset:   0%|          | 0/1 [00:00<?, ?ba/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# data preprocessing\n",
    "tokenizer = AutoTokenizer.from_pretrained(model_name_or_path)\n",
    "def preprocess_function(examples):\n",
    "    inputs = examples[text_column]\n",
    "    targets = examples[label_column]\n",
    "    model_inputs = tokenizer(inputs, max_length=max_length, padding=\"max_length\", truncation=True, return_tensors=\"pt\")\n",
    "    labels = tokenizer(targets, max_length=3, padding=\"max_length\", truncation=True, return_tensors=\"pt\")\n",
    "    labels = labels[\"input_ids\"]\n",
    "    labels[labels==tokenizer.pad_token_id] = -100\n",
    "    model_inputs[\"labels\"] = labels\n",
    "    return model_inputs\n",
    "\n",
    "processed_datasets = dataset.map(\n",
    "            preprocess_function,\n",
    "            batched=True,\n",
    "            num_proc=1,\n",
    "            remove_columns=dataset[\"train\"].column_names,\n",
    "            load_from_cache_file=False,\n",
    "            desc=\"Running tokenizer on dataset\",\n",
    "        )\n",
    "\n",
    "train_dataset = processed_datasets[\"train\"]\n",
    "eval_dataset = processed_datasets[\"validation\"]\n",
    "\n",
    "train_dataloader = DataLoader(\n",
    "        train_dataset, shuffle=True, collate_fn=default_data_collator, batch_size=batch_size, pin_memory=True\n",
    "    )\n",
    "eval_dataloader = DataLoader(eval_dataset, collate_fn=default_data_collator, batch_size=batch_size, pin_memory=True)\n",
    "\n",
    "\n",
    "\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "f733a3c6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# optimizer and lr scheduler\n",
    "optimizer = torch.optim.AdamW(model.parameters(), lr=lr)\n",
    "lr_scheduler = get_linear_schedule_with_warmup(\n",
    "    optimizer=optimizer,\n",
    "    num_warmup_steps=0,\n",
    "    num_training_steps=(len(train_dataloader) * num_epochs),\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "6b3a4090",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████| 255/255 [00:53<00:00,  4.80it/s]\n",
      "100%|███████████████████████████████████████████████████████████████| 29/29 [00:02<00:00, 14.16it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch=0: train_ppl=tensor(13.6966, device='cuda:0') train_epoch_loss=tensor(2.6171, device='cuda:0') eval_ppl=tensor(1.0046, device='cuda:0') eval_epoch_loss=tensor(0.0046, device='cuda:0')\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████| 255/255 [00:52<00:00,  4.88it/s]\n",
      "100%|███████████████████████████████████████████████████████████████| 29/29 [00:02<00:00, 14.20it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch=1: train_ppl=tensor(1.5893, device='cuda:0') train_epoch_loss=tensor(0.4633, device='cuda:0') eval_ppl=tensor(1.0020, device='cuda:0') eval_epoch_loss=tensor(0.0020, device='cuda:0')\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████| 255/255 [00:52<00:00,  4.87it/s]\n",
      "100%|███████████████████████████████████████████████████████████████| 29/29 [00:02<00:00, 14.18it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch=2: train_ppl=tensor(1.3210, device='cuda:0') train_epoch_loss=tensor(0.2784, device='cuda:0') eval_ppl=tensor(1.0026, device='cuda:0') eval_epoch_loss=tensor(0.0026, device='cuda:0')\n"
     ]
    }
   ],
   "source": [
    "# training and evaluation\n",
    "model = model.to(device)\n",
    "\n",
    "for epoch in range(num_epochs):\n",
    "    model.train()\n",
    "    total_loss = 0\n",
    "    for step, batch in enumerate(tqdm(train_dataloader)):\n",
    "        batch = {k: v.to(device) for k, v in batch.items()}\n",
    "        outputs = model(**batch)\n",
    "        loss = outputs.loss\n",
    "        total_loss += loss.detach().float()\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        lr_scheduler.step()\n",
    "        optimizer.zero_grad()\n",
    "\n",
    "    model.eval()\n",
    "    eval_loss = 0\n",
    "    eval_preds = []\n",
    "    for step, batch in enumerate(tqdm(eval_dataloader)):\n",
    "        batch = {k: v.to(device) for k, v in batch.items()}\n",
    "        with torch.no_grad():\n",
    "            outputs = model(**batch)\n",
    "        loss = outputs.loss\n",
    "        eval_loss += loss.detach().float()\n",
    "        eval_preds.extend(tokenizer.batch_decode(torch.argmax(outputs.logits, -1).detach().cpu().numpy(), skip_special_tokens=True))\n",
    "\n",
    "    eval_epoch_loss = eval_loss/len(train_dataloader)\n",
    "    eval_ppl = torch.exp(eval_epoch_loss)\n",
    "    train_epoch_loss = total_loss/len(eval_dataloader)\n",
    "    train_ppl = torch.exp(train_epoch_loss)\n",
    "    print(f\"{epoch=}: {train_ppl=} {train_epoch_loss=} {eval_ppl=} {eval_epoch_loss=}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "6cafa67b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy=98.23788546255507 % on the evaluation dataset\n",
      "eval_preds[:10]=['neutral', 'neutral', 'positive', 'positive', 'neutral', 'neutral', 'neutral', 'neutral', 'neutral', 'neutral']\n",
      "dataset['validation']['text_label'][:10]=['neutral', 'neutral', 'positive', 'positive', 'neutral', 'neutral', 'neutral', 'neutral', 'neutral', 'neutral']\n"
     ]
    }
   ],
   "source": [
    "# print accuracy\n",
    "correct =0\n",
    "total = 0\n",
    "for pred,true in zip(eval_preds, dataset[\"validation\"][\"text_label\"]):\n",
    "    if pred.strip()==true.strip():\n",
    "        correct+=1\n",
    "    total+=1  \n",
    "accuracy = correct/total*100\n",
    "print(f\"{accuracy=} % on the evaluation dataset\")\n",
    "print(f\"{eval_preds[:10]=}\")\n",
    "print(f\"{dataset['validation']['text_label'][:10]=}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a8de6005",
   "metadata": {},
   "outputs": [],
   "source": [
    "# saving model\n",
    "state_dict = get_peft_model_state_dict(model)\n",
    "torch.save(state_dict, checkpoint_name)\n",
    "print(state_dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "bd20cd4c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19M\tfinancial_sentiment_analysis_lora_v1.pt\r\n"
     ]
    }
   ],
   "source": [
    "!du -h $checkpoint_name"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "76c2fc29",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.10.5 64-bit",
   "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": "3.10.5 (v3.10.5:f377153967, Jun  6 2022, 12:36:10) [Clang 13.0.0 (clang-1300.0.29.30)]"
  },
  "vscode": {
   "interpreter": {
    "hash": "aee8b7b246df8f9039afb4144a1f6fd8d2ca17a180786b69acc140d282b71a49"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}