hf_transformers.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Greedy search"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/ambujpawar/Desktop/GithubProjects/lets_learn_gpt/venv/lib/python3.10/site-packages/tqdm/auto.py:22: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n",
      "Downloading (…)lve/main/config.json: 100%|██████████| 666/666 [00:00<00:00, 219kB/s]\n",
      "Downloading (…)olve/main/vocab.json: 100%|██████████| 1.04M/1.04M [00:00<00:00, 1.80MB/s]\n",
      "Downloading (…)olve/main/merges.txt: 100%|██████████| 456k/456k [00:00<00:00, 1.22MB/s]\n",
      "Downloading (…)/main/tokenizer.json: 100%|██████████| 1.36M/1.36M [00:00<00:00, 2.27MB/s]\n",
      "Downloading (…)\"pytorch_model.bin\";: 100%|██████████| 3.25G/3.25G [04:51<00:00, 11.2MB/s]\n",
      "Downloading (…)neration_config.json: 100%|██████████| 124/124 [00:00<00:00, 52.2kB/s]\n",
      "Setting `pad_token_id` to `eos_token_id`:50256 for open-end generation.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "['Hugging Face Company is a family owned and operated business. We pride ourselves on being the best in the business and our customer service is second to none.\\n\\nIf you have any questions about our products or services, feel free to contact us at any time. We look forward to hearing from you!']"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from transformers import AutoTokenizer, AutoModelForCausalLM\n",
    "\n",
    "checkpoint = \"gpt2-large\"\n",
    "tokenizer = AutoTokenizer.from_pretrained(checkpoint)\n",
    "model = AutoModelForCausalLM.from_pretrained(checkpoint)\n",
    "\n",
    "prompt = \"Hugging Face Company is\"\n",
    "inputs = tokenizer(prompt, return_tensors=\"pt\")\n",
    "\n",
    "outputs = model.generate(**inputs, penalty_alpha=0.6, top_k=4, max_new_tokens=100)\n",
    "tokenizer.batch_decode(outputs, skip_special_tokens=True)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Multinomial sampling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Setting `pad_token_id` to `eos_token_id`:50256 for open-end generation.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[\"Today was an amazing day because my parents had finally met their son, a kid at school with me and he brought me here. He helped me to find my parents and they helped me get here so that my dad could meet him. It was wonderful. My parents were there with me to welcome us to Mumbai and to meet my brother. His brother came with me.\\n\\nQ. And the first thing you saw when you arrived here after completing your Master's was that a lot of new people were coming in with lots of\"]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from transformers import AutoTokenizer, AutoModelForCausalLM\n",
    "\n",
    "#checkpoint = \"gpt2-large\"\n",
    "#tokenizer = AutoTokenizer.from_pretrained(checkpoint)\n",
    "#model = AutoModelForCausalLM.from_pretrained(checkpoint)\n",
    "\n",
    "prompt = \"Today was an amazing day because\"\n",
    "inputs = tokenizer(prompt, return_tensors=\"pt\")\n",
    "\n",
    "outputs = model.generate(**inputs, do_sample=True, max_new_tokens=100)\n",
    "tokenizer.batch_decode(outputs, skip_special_tokens=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading (…)lve/main/config.json: 100%|██████████| 1.21k/1.21k [00:00<00:00, 284kB/s]\n",
      "Downloading (…)ve/main/spiece.model: 100%|██████████| 792k/792k [00:00<00:00, 1.65MB/s]\n",
      "Downloading (…)/main/tokenizer.json: 100%|██████████| 1.39M/1.39M [00:00<00:00, 2.11MB/s]\n",
      "/Users/ambujpawar/Desktop/GithubProjects/lets_learn_gpt/venv/lib/python3.10/site-packages/transformers/models/t5/tokenization_t5_fast.py:155: FutureWarning: This tokenizer was incorrectly instantiated with a model max length of 512 which will be corrected in Transformers v5.\n",
      "For now, this behavior is kept to avoid breaking backwards compatibility when padding/encoding with `truncation is True`.\n",
      "- Be aware that you SHOULD NOT rely on t5-small automatically truncating your input to 512 when padding/encoding.\n",
      "- If you want to encode/pad to sequences longer than 512 you can either instantiate this tokenizer with `model_max_length` or pass `max_length` when encoding/padding.\n",
      "- To avoid this warning, please instantiate this tokenizer with `model_max_length` set to your preferred value.\n",
      "  warnings.warn(\n",
      "Downloading (…)\"pytorch_model.bin\";: 100%|██████████| 242M/242M [00:20<00:00, 11.6MB/s] \n",
      "Downloading (…)neration_config.json: 100%|██████████| 147/147 [00:00<00:00, 47.4kB/s]\n",
      "/Users/ambujpawar/Desktop/GithubProjects/lets_learn_gpt/venv/lib/python3.10/site-packages/transformers/generation/utils.py:1273: UserWarning: Neither `max_length` nor `max_new_tokens` has been set, `max_length` will default to 20 (`generation_config.max_length`). Controlling `max_length` via the config is deprecated and `max_length` will be removed from the config in v5 of Transformers -- we recommend using `max_new_tokens` to control the maximum length of the generation.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'Das Haus ist wunderbar.'"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from transformers import AutoTokenizer, AutoModelForSeq2SeqLM\n",
    "\n",
    "prompt = \"translate English to German: The house is wonderful.\"\n",
    "checkpoint = \"t5-small\"\n",
    "\n",
    "tokenizer = AutoTokenizer.from_pretrained(checkpoint)\n",
    "inputs = tokenizer(prompt, return_tensors=\"pt\")\n",
    "\n",
    "model = AutoModelForSeq2SeqLM.from_pretrained(checkpoint)\n",
    "\n",
    "outputs = model.generate(**inputs, num_beams=5, do_sample=True)\n",
    "tokenizer.decode(outputs[0], skip_special_tokens=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "venv",
   "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": "3.10.9"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "ab35b32f68116c0216e57a17ae4d2a651f25e42e7896b86e875bcfc913c2896e"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}