examples/multi_adapter_examples/Lora_Merging.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "db4208b9-5da4-46df-b77a-0f1836c9e4ec",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/raid/sourab/transformers/src/transformers/utils/hub.py:122: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.\n",
      "  warnings.warn(\n",
      "Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "\n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"1\"\n",
    "from peft import PeftConfig, PeftModel\n",
    "from peft import PeftModel, PeftConfig\n",
    "from transformers import AutoModelForCausalLM, AutoTokenizer\n",
    "from datasets import load_dataset\n",
    "import torch\n",
    "import random\n",
    "\n",
    "peft_model_id = \"smangrul/tinyllama_lora_norobots\"\n",
    "device = \"cuda\"\n",
    "config = PeftConfig.from_pretrained(peft_model_id)\n",
    "model = AutoModelForCausalLM.from_pretrained(config.base_model_name_or_path, load_in_4bit=True, device_map=\"auto\")\n",
    "tokenizer = AutoTokenizer.from_pretrained(peft_model_id)\n",
    "model.resize_token_embeddings(len(tokenizer))\n",
    "model = PeftModel.from_pretrained(model, peft_model_id, adapter_name=\"norobots\")\n",
    "_ = model.load_adapter(\"smangrul/tinyllama_lora_sql\", adapter_name=\"sql\")\n",
    "_ = model.load_adapter(\"smangrul/tinyllama_lora_adcopy\", adapter_name=\"adcopy\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 684,
   "id": "541dab43-9675-42a2-8d90-7437df9f0fa0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 23.3 s, sys: 535 ms, total: 23.8 s\n",
      "Wall time: 796 ms\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "# [0.8, 0.1, 0.1] linear #[1.0, 0.2] 0.7 density dare_linear #[1.5, 0.3] 0.5 density ties #[0.8, 0.5] cat\n",
    "adapters = [\"norobots\", \"adcopy\", \"sql\"]\n",
    "weights = [2.0, 0.3, 0.7]\n",
    "adapter_name = \"merge\"\n",
    "density = 0.2\n",
    "combination_type = \"ties\"\n",
    "if adapter_name in model.peft_config:\n",
    "    model.delete_adapter(adapter_name)\n",
    "model.add_weighted_adapter(adapters, weights, adapter_name, combination_type=combination_type, density=density)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 685,
   "id": "76596671-3677-47f0-9d66-81f40bc4d726",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.eval()\n",
    "model.set_adapter(\"merge\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 691,
   "id": "9d59f9f3-6313-43d8-be36-4ca2bbb105b2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<s><|im_start|>user \n",
      "Write an essay about Generative AI.<|im_end|> \n",
      "<|im_start|>assistant \n",
      "Generative Artificial Intelligence (GAI) is a type of artificial intelligence that uses machine learning to create art, music and other creations. It's like having a human artist who creates something new without the need for inspiration or motivation.<|im_end|>\n"
     ]
    }
   ],
   "source": [
    "messages = [\n",
    "    {\"role\": \"user\", \"content\": \"Write an essay about Generative AI.\"},\n",
    "]\n",
    "text = tokenizer.apply_chat_template(messages, add_generation_prompt=True, tokenize=False)\n",
    "inputs = tokenizer(text, return_tensors=\"pt\")  # , add_special_tokens=False)\n",
    "inputs = {k: v.to(\"cuda\") for k, v in inputs.items()}\n",
    "outputs = model.generate(\n",
    "    **inputs,\n",
    "    max_new_tokens=256,\n",
    "    do_sample=True,\n",
    "    top_p=0.95,\n",
    "    temperature=0.2,\n",
    "    repetition_penalty=1.2,\n",
    "    eos_token_id=tokenizer.eos_token_id,\n",
    ")\n",
    "print(tokenizer.decode(outputs[0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 689,
   "id": "e5c1daeb-59c8-41d7-bebb-7abd052ab917",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<s><|im_start|>system \n",
      "Create a text ad given the following product and description.<|im_end|> \n",
      "<|im_start|>user \n",
      "Product: Sony PS5 PlayStation Console\n",
      "Description: The PS5™ console unleashes new gaming possibilities that you never anticipated.<|im_end|> \n",
      "<|im_start|>assistant \n",
      "Ad Text: Experience the next-gen power of the all-new Sony PS5 with its stunning visuals, innovative gameplay features, and more! Get ready to play in style as you experience the future of gaming on your own terms.<|im_end|>\n"
     ]
    }
   ],
   "source": [
    "messages = [\n",
    "    {\"role\": \"system\", \"content\": \"Create a text ad given the following product and description.\"},\n",
    "    {\n",
    "        \"role\": \"user\",\n",
    "        \"content\": \"Product: Sony PS5 PlayStation Console\\nDescription: The PS5™ console unleashes new gaming possibilities that you never anticipated.\",\n",
    "    },\n",
    "]\n",
    "text = tokenizer.apply_chat_template(messages, add_generation_prompt=True, tokenize=False)\n",
    "inputs = tokenizer(text, return_tensors=\"pt\")  # , add_special_tokens=False)\n",
    "inputs = {k: v.to(\"cuda\") for k, v in inputs.items()}\n",
    "outputs = model.generate(\n",
    "    **inputs,\n",
    "    max_new_tokens=128,\n",
    "    do_sample=True,\n",
    "    top_p=0.95,\n",
    "    temperature=0.2,\n",
    "    repetition_penalty=1.2,\n",
    "    eos_token_id=tokenizer.eos_token_id,\n",
    ")\n",
    "print(tokenizer.decode(outputs[0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 690,
   "id": "5bb08b46-90ae-48a8-8783-ca74b3e26e42",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<s> Table: 2-11365528-2\n",
      "Columns: ['Team', 'Head Coach', 'President', 'Home Ground', 'Location']\n",
      "Natural Query: Who is the Head Coach of the team whose President is Mario Volarevic?\n",
      "SQL Query: SELECT Head Coach FROM 2-11365528-2 WHERE President = Mario Volarevic</s>\n"
     ]
    }
   ],
   "source": [
    "text = \"\"\"Table: 2-11365528-2\n",
    "Columns: ['Team', 'Head Coach', 'President', 'Home Ground', 'Location']\n",
    "Natural Query: Who is the Head Coach of the team whose President is Mario Volarevic?\n",
    "SQL Query:\"\"\"\n",
    "\n",
    "inputs = tokenizer(text, return_tensors=\"pt\")  # , add_special_tokens=False)\n",
    "inputs = {k: v.to(\"cuda\") for k, v in inputs.items()}\n",
    "outputs = model.generate(\n",
    "    **inputs, max_new_tokens=64, repetition_penalty=1.1, eos_token_id=tokenizer(\"</s>\").input_ids[-1]\n",
    ")\n",
    "print(tokenizer.decode(outputs[0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cc927536-bb58-4270-876f-10ff1a94802d",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": "3.10.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}