BarinkDev/LargeLanguageModels

1{2 "cells": [3  {4   "cell_type": "markdown",5   "id": "6eee0106-2fee-4263-855f-2c362ce159d1",6   "metadata": {},7   "source": [8    "# New model from google called Gemma "9   ]10  },11  {12   "cell_type": "markdown",13   "id": "37731f28-7b4b-4229-9270-c0a547c13030",14   "metadata": {},15   "source": [16    "Source: [https://huggingface.co/google/gemma-7b](https://huggingface.co/google/gemma-7b)"17   ]18  },19  {20   "cell_type": "code",21   "execution_count": null,22   "id": "4a9c70eb-6f88-4c82-ba14-eb358a62a198",23   "metadata": {},24   "outputs": [],25   "source": [26    "# pip install bitsandbytes accelerate\n",27    "%pip install bitsandbytes accelerate \n"28   ]29  },30  {31   "cell_type": "code",32   "execution_count": null,33   "id": "3f0c0b9f-c457-4d05-b349-d76b0b65dc10",34   "metadata": {},35   "outputs": [36    {37     "data": {38      "application/vnd.jupyter.widget-view+json": {39       "model_id": "d1559a58ffff43609eaa5074b0178cd9",40       "version_major": 2,41       "version_minor": 042      },43      "text/plain": [44       "Loading checkpoint shards:   0%|          | 0/4 [00:00<?, ?it/s]"45      ]46     },47     "metadata": {},48     "output_type": "display_data"49    },50    {51     "name": "stderr",52     "output_type": "stream",53     "text": [54      "WARNING:root:Some parameters are on the meta device device because they were offloaded to the disk and cpu.\n",55      "2024-02-21 23:34:48.779292: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",56      "2024-02-21 23:34:48.781146: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",57      "2024-02-21 23:34:49.052470: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",58      "2024-02-21 23:34:58.007900: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n",59      "\n",60      "KeyboardInterrupt\n",61      "\n"62     ]63    }64   ],65   "source": [66    "import os \n",67    "from transformers import AutoTokenizer, AutoModelForCausalLM, AutoConfig\n",68    "os.environ['TRANSFORMERS_CACHE'] = '/mnt/f/wkspc-linux/.cache'\n",69    "os.environ['HUGGINGFACE_HUB_CACHE'] = '/mnt/f/wkspc-linux/.cache'\n",70    "os.environ['HF_HOME'] = '/mnt/f/wkspc-linux/.cache'\n",71    "\n",72    "tokenizer = AutoTokenizer.from_pretrained(\"google/gemma-7b\", cache_dir=\"/mnt/f/wkspc-linux/.cache\")\n",73    "model = AutoModelForCausalLM.from_pretrained(\"google/gemma-7b\",device_map=\"auto\", cache_dir=\"/mnt/f/wkspc-linux/.cache\" )\n",74    "\n",75    "\n",76    "input_text = \"Write me a poem about Machine Learning.\"\n",77    "input_ids = tokenizer(input_text, return_tensors=\"pt\").to(\"cuda\")\n",78    "\n",79    "outputs = model.generate(**input_ids, max_new_tokens=200)\n",80    "print(tokenizer.decode(outputs[0]))\n"81   ]82  },83  {84   "cell_type": "code",85   "execution_count": 1,86   "id": "c5eeae00-6f0e-4686-a3b8-3ee3538abb87",87   "metadata": {},88   "outputs": [89    {90     "data": {91      "application/vnd.jupyter.widget-view+json": {92       "model_id": "8dfc17e70a904d88b7fa87d23ec44cfc",93       "version_major": 2,94       "version_minor": 095      },96      "text/plain": [97       "tokenizer.model:   0%|          | 0.00/4.24M [00:00<?, ?B/s]"98      ]99     },100     "metadata": {},101     "output_type": "display_data"102    },103    {104     "name": "stderr",105     "output_type": "stream",106     "text": [107      "`low_cpu_mem_usage` was None, now set to True since model is quantized.\n"108     ]109    },110    {111     "data": {112      "application/vnd.jupyter.widget-view+json": {113       "model_id": "af8ce3d6f3444aaf8bb0d8968ccd9fec",114       "version_major": 2,115       "version_minor": 0116      },117      "text/plain": [118       "Downloading shards:   0%|          | 0/4 [00:00<?, ?it/s]"119      ]120     },121     "metadata": {},122     "output_type": "display_data"123    },124    {125     "data": {126      "application/vnd.jupyter.widget-view+json": {127       "model_id": "ab7543ed101e45aeb18df74325a073e2",128       "version_major": 2,129       "version_minor": 0130      },131      "text/plain": [132       "Loading checkpoint shards:   0%|          | 0/4 [00:00<?, ?it/s]"133      ]134     },135     "metadata": {},136     "output_type": "display_data"137    },138    {139     "name": "stderr",140     "output_type": "stream",141     "text": [142      "2024-02-23 15:55:39.248214: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",143      "2024-02-23 15:55:39.251426: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",144      "2024-02-23 15:55:39.992353: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",145      "2024-02-23 15:56:03.288614: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n"146     ]147    },148    {149     "name": "stdout",150     "output_type": "stream",151     "text": [152      "<bos>Write me a poem about Machine Learning.\n",153      "\n",154      "I’m not a poet, but I’m a Machine Learning Engineer.\n",155      "\n",156      "I’m not a poet, but I’m a\n"157     ]158    }159   ],160   "source": [161    "\n",162    "from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig\n",163    "\n",164    "quantization_config = BitsAndBytesConfig(load_in_8bit=True)\n",165    "\n",166    "tokenizer = AutoTokenizer.from_pretrained(\"google/gemma-7b\", cache_dir=\"/mnt/f/wkspc-linux/.cache\")\n",167    "model = AutoModelForCausalLM.from_pretrained(\"google/gemma-7b\", quantization_config=quantization_config, cache_dir=\"/mnt/f/wkspc-linux/.cache\")\n",168    "\n",169    "input_text = \"Write me a poem about Machine Learning.\"\n",170    "input_ids = tokenizer(input_text, return_tensors=\"pt\").to(\"cuda\")\n",171    "\n",172    "outputs = model.generate(**input_ids, max_new_tokens=30)\n",173    "print(tokenizer.decode(outputs[0]))\n"174   ]175  },176  {177   "cell_type": "code",178   "execution_count": 2,179   "id": "1b0c0f09-a1d4-4b9f-baa1-914ead919b06",180   "metadata": {},181   "outputs": [],182   "source": [183    "def prompt(input_text):\n",184    "    input_ids = tokenizer(input_text, return_tensors=\"pt\").to(\"cuda\")\n",185    "    \n",186    "    outputs = model.generate(**input_ids, max_new_tokens=30)\n",187    "    return tokenizer.decode(outputs[0])"188   ]189  },190  {191   "cell_type": "code",192   "execution_count": null,193   "id": "3d2e9452-72e3-4467-bc75-cdcaa1cd5693",194   "metadata": {},195   "outputs": [],196   "source": [197    "print(prompt(\"What is a good framework for python webscraping?\"))"198   ]199  },200  {201   "cell_type": "code",202   "execution_count": null,203   "id": "42093c90-7c93-4f3d-a109-9b298688ddea",204   "metadata": {},205   "outputs": [],206   "source": []207  }208 ],209 "metadata": {210  "kernelspec": {211   "display_name": "Python 3 (ipykernel)",212   "language": "python",213   "name": "python3"214  },215  "language_info": {216   "codemirror_mode": {217    "name": "ipython",218    "version": 3219   },220   "file_extension": ".py",221   "mimetype": "text/x-python",222   "name": "python",223   "nbconvert_exporter": "python",224   "pygments_lexer": "ipython3",225   "version": "3.11.2"226  }227 },228 "nbformat": 4,229 "nbformat_minor": 5230}231