{ "cells": [ { "cell_type": "markdown", "id": "b3907ee8-2ed6-4997-b85e-078f18ade788", "metadata": {}, "source": [ "# 3. Korean TrOCR Training on Amazon SageMaker\n", "---\n", "\n", "## Introduction\n", "---\n", "\n", "바로 ì´ì „ 모듈ì—ì„œ ê¸°ì¡´ì— ì˜¨í”„ë ˆë¯¸ìŠ¤ì—ì„œ ê°œë°œí–ˆë˜ í™˜ê²½ê³¼ ë™ì¼í•œ 환경으로 모ë¸ì„ ë¹Œë“œí•˜ê³ í›ˆë ¨í–ˆìŠµë‹ˆë‹¤. 하지만 아래와 ê°™ì€ ìƒí™©ë“¤ì—ì„œë„ ê¸°ì¡´ í™˜ê²½ì„ ì‚¬ìš©í•˜ëŠ” ê²ƒì´ ë°”ëžŒì§í• 까요?\n", "\n", "- ì˜¨í”„ë ˆë¯¸ìŠ¤ì˜ GPU는 ì´ 1장으로 í›ˆë ¨ ì‹œê°„ì´ ë„ˆë¬´ 오래 소요ë¨\n", "- 가용 서버 대수가 2대ì¸ë° 10ê°œì˜ ë”¥ëŸ¬ë‹ ëª¨ë¸ì„ ë™ì‹œì— í›ˆë ¨í•´ì•¼ 함\n", "- 필요한 ìƒí™©ì—만 GPU를 활용\n", "- 기타 등등\n", "\n", "Amazon SageMaker는 ë°ì´í„° 과학ìžë“¤ ë° ë¨¸ì‹ ëŸ¬ë‹ ì—”ì§€ë‹ˆì–´ë“¤ì„ ìœ„í•œ ì™„ì „ 관리형 ë¨¸ì‹ ëŸ¬ë‹ ì„œë¹„ìŠ¤ë¡œ í›ˆë ¨ ë° ì¶”ë¡ ìˆ˜í–‰ ì‹œ ì¸í”„ë¼ ì„¤ì •ì— ëŒ€í•œ 추가 ìž‘ì—…ì´ í•„ìš”í•˜ì§€ 않기ì—, ë‹¨ì¼ GPU ê¸°ë°˜ì˜ ë”¥ëŸ¬ë‹ í›ˆë ¨ì„ í¬í•¨í•œ 멀티 GPU ë° ë©€í‹° ì¸ìŠ¤í„´ìŠ¤ 분산 í›ˆë ¨ì„ ë³´ë‹¤ ì‰½ê³ ë¹ ë¥´ê²Œ ìˆ˜í–‰í• ìˆ˜ 있습니다. SageMaker는 다양한 ìœ ì¦ˆì¼€ì´ìŠ¤ë“¤ì— ì í•©í•œ ì˜ˆì œë“¤ì„ ì§€ì†ì 으로 ì—…ë°ì´íŠ¸í•˜ê³ 있으며, í•œêµì–´ 세션 ë° ìžë£Œë“¤ë„ ì œê³µë˜ê³ 있습니다.\n", "\n", "### Notes\n", "\n", "ì´ë¯¸ 기본ì ì¸ Hugging Face 용법 ë° ìžì—°ì–´ ì²˜ë¦¬ì— ìµìˆ™í•˜ì‹ ë¶„ë“¤ì€ ì•ž ëª¨ë“ˆì„ ìƒëžµí•˜ê³ ì´ ëª¨ë“ˆë¶€í„° í•¸ì¦ˆì˜¨ì„ ì‹œìž‘í•˜ì…”ë„ ë©ë‹ˆë‹¤.\n", "ì´ ë…¸íŠ¸ë¶ì€ SageMaker 기본 API를 참조하므로, SageMaker Studio, SageMaker ë…¸íŠ¸ë¶ ì¸ìŠ¤í„´ìŠ¤ ë˜ëŠ” AWS CLIê°€ ì„¤ì •ëœ ë¡œì»¬ 시스템ì—ì„œ 실행해야 합니다. SageMaker Studio ë˜ëŠ” SageMaker ë…¸íŠ¸ë¶ ì¸ìŠ¤í„´ìŠ¤ë¥¼ 사용하는 경우 PyTorch 기반 커ë„ì„ ì„ íƒí•˜ì„¸ìš”.\n", "í›ˆë ¨ job 수행 ì‹œ 최소 `ml.g4dn.xlarge` ì´ìƒì˜ í›ˆë ¨ ì¸ìŠ¤í„´ìŠ¤ê°€ 필요하며, `ml.g5.12xlarge` ì¸ìŠ¤í„´ìŠ¤ë¥¼ 권장합니다. 만약 ì¸ìŠ¤í„´ìŠ¤ ì‚¬ìš©ì— ì œí•œì´ ê±¸ë ¤ 있다면 Request a service quota increase for SageMaker resources를 참조하여 ì¸ìŠ¤í„´ìŠ¤ ì œí•œì„ í•´ì œí•´ 주세요.\n", "\n", "\n", "### References\n", "- TrOCR GirHub: https://github.com/microsoft/unilm/tree/master/trocr\n", "- Hugging Face TrOCR Tutorial: https://huggingface.co/docs/transformers/model_doc/trocr\n", "- Hugging Face Stage-1 pre-trained model: https://huggingface.co/microsoft/trocr-base-stage1" ] }, { "cell_type": "markdown", "id": "d254ea46-b7af-4bc9-91c9-668a0191f47d", "metadata": {}, "source": [ "\n", "## 1. Setup Environments\n", "---\n", "\n", "### Import modules" ] }, { "cell_type": "code", "execution_count": 1, "id": "57cceb15-6722-4349-ad38-e7da3437fdd3", "metadata": {}, "outputs": [], "source": [ "import boto3\n", "import sagemaker\n", "import sagemaker.huggingface\n", "\n", "sess = sagemaker.Session()\n", "# sagemaker session bucket -> used for uploading data, models and logs\n", "# sagemaker will automatically create this bucket if it not exists\n", "sagemaker_session_bucket=None\n", "if sagemaker_session_bucket is None and sess is not None:\n", " # set to default bucket if a bucket name is not given\n", " sagemaker_session_bucket = sess.default_bucket()\n", "\n", "role = sagemaker.get_execution_role()\n", "region = boto3.Session().region_name\n", "sess = sagemaker.Session(default_bucket=sagemaker_session_bucket)\n", "\n", "print(f\"sagemaker role arn: {role}\")\n", "print(f\"sagemaker bucket: {sess.default_bucket()}\")\n", "print(f\"sagemaker session region: {sess.boto_region_name}\")" ] }, { "cell_type": "code", "execution_count": 2, "id": "d24aaf1e-891b-4502-bb44-67a1e117a802", "metadata": {}, "outputs": [], "source": [ "import os\n", "import sys\n", "import json\n", "import logging\n", "import argparse\n", "import torch\n", "from torch import nn\n", "import numpy as np\n", "import pandas as pd\n", "from tqdm import tqdm\n", "from IPython.display import display, HTML\n", "\n", "from transformers import (\n", " AutoTokenizer, VisionEncoderDecoderModel,\n", " Trainer, TrainingArguments, set_seed\n", ")\n", "from transformers import Seq2SeqTrainingArguments, Seq2SeqTrainer\n", "\n", "from transformers.trainer_utils import get_last_checkpoint\n", "from datasets import load_dataset, load_metric, ClassLabel, Sequence\n", "\n", "logging.basicConfig(\n", " level=logging.INFO, \n", " format='[{%(filename)s:%(lineno)d} %(levelname)s - %(message)s',\n", " handlers=[\n", " logging.StreamHandler(sys.stdout)\n", " ]\n", ")\n", "logger = logging.getLogger(__name__)" ] }, { "cell_type": "markdown", "id": "c699c4dc-30a5-45a2-95d0-ab39f9e06820", "metadata": {}, "source": [ "### Uploading data to Amazon S3 Bucket\n", "SageMaker í›ˆë ¨ì„ ìœ„í•´ ë°ì´í„°ì…‹ì„ S3ë¡œ 업로드합니다." ] }, { "cell_type": "code", "execution_count": 6, "id": "c02f239d-0b26-4ccb-b61e-36a02ac02946", "metadata": {}, "outputs": [], "source": [ "bucket = sess.default_bucket()\n", "\n", "# s3 key prefix for the data\n", "s3_prefix = 'samples/datasets/trocr'\n", "\n", "train_local_path = 'train'\n", "\n", "# save train_dataset to s3\n", "train_input_path = f's3://{bucket}/{s3_prefix}/train'" ] }, { "cell_type": "code", "execution_count": 7, "id": "2f0496f0-5405-4f56-a362-b53368665a02", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "CPU times: user 34.5 ms, sys: 41.8 ms, total: 76.4 ms\n", "Wall time: 5.68 s\n" ] } ], "source": [ "%%time\n", "!aws s3 cp {train_local_path} {train_input_path} --recursive --quiet" ] }, { "cell_type": "markdown", "id": "af0bae50-8d80-4b4e-a215-ce0a69f741b4", "metadata": {}, "source": [ "<br>\n", "\n", "## 2. Training with Native Hugging Face (PyTorch Framework)\n", "\n", "---\n", "\n", "### Overview and Training Script\n", "\n", "SageMakerì— ëŒ€í•œ 대표ì ì¸ ì˜¤í•´ê°€ ì—¬ì „ížˆ ë§Žì€ ë¶„ë“¤ì´ SageMaker í›ˆë ¨ì„ ìœ„í•´ 소스 코드를 ì „ë©´ì 으로 ìˆ˜ì •í•´ì•¼ í•œë‹¤ê³ ìƒê°í•©ë‹ˆë‹¤. 하지만, ì‹¤ì œë¡œëŠ” 별ë„ì˜ ì†ŒìŠ¤ 코드 ìˆ˜ì • ì—†ì´ ê¸°ì¡´ ì—¬ëŸ¬ë¶„ì´ ì‚¬ìš©í–ˆë˜ íŒŒì´ì¬ 스í¬ë¦½íŠ¸ì— SageMaker í›ˆë ¨ì— í•„ìš”í•œ SageMaker ì „ìš© 환경 변수들만 추가하면 ë©ë‹ˆë‹¤. \n", "\n", "SageMaker í›ˆë ¨ì€ í›ˆë ¨ ìž‘ì—…ì„ í˜¸ì¶œí• ë•Œ, 1) í›ˆë ¨ EC2 ì¸ìŠ¤í„´ìŠ¤ í”„ë¡œë¹„ì €ë‹ - 2) 컨테ì´ë„ˆ 구ë™ì„ 위한 ë„커 ì´ë¯¸ì§€ ë° í›ˆë ¨ ë°ì´í„° 다운로드 - 3) 컨테ì´ë„ˆ êµ¬ë™ - 4) 컨테ì´ë„ˆ 환경ì—ì„œ í›ˆë ¨ 수행 - 5) 컨테ì´ë„ˆ 환경ì—ì„œ S3ì˜ íŠ¹ì • ë²„í‚·ì— ì €ìž¥ - 6) í›ˆë ¨ ì¸ìŠ¤í„´ìŠ¤ 종료로 구성ë©ë‹ˆë‹¤. ë”°ë¼ì„œ, í›ˆë ¨ 수행 ë¡œì§ì€ 아래 예시와 ê°™ì´ ê¸°ì¡´ 개발 환경과 ë™ì¼í•©ë‹ˆë‹¤.\n", "\n", "```python\n", "/opt/conda/bin/python train.py --epochs 5 --train_batch_size 32 ...\n", "```\n", "\n", "ì´ ê³¼ì •ì—ì„œ 컨테ì´ë„ˆ í™˜ê²½ì— í•„ìš”í•œ 환경 변수(예: ëª¨ë¸ ê²½ë¡œ, í›ˆë ¨ ë°ì´í„° 경로) ë“¤ì€ ì‚¬ì „ì— ì§€ì •ë˜ì–´ 있으며, ì´ í™˜ê²½ ë³€ìˆ˜ë“¤ì´ ì„¤ì •ë˜ì–´ 있어야 í›ˆë ¨ì— í•„ìš”í•œ 파ì¼ë“¤ì˜ 경로를 ì¸ì‹í• 수 있습니다. 대표ì ì¸ í™˜ê²½ ë³€ìˆ˜ë“¤ì— ëŒ€í•œ ìžì„¸í•œ ë‚´ìš©ì€ https://github.com/aws/sagemaker-containers#important-environment-variables ì„ ì°¸ì¡°í•˜ì„¸ìš”." ] }, { "cell_type": "code", "execution_count": 15, "id": "ee1abbf3-91e3-44be-a732-ee529739c5bb", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[{4206700043.py:8} INFO - learning_rate: 4e-05\n" ] } ], "source": [ "from sagemaker.huggingface import HuggingFace\n", "import time\n", "instance_type = 'ml.g5.xlarge'\n", "num_gpus = 1\n", "instance_count = 1\n", "batch_size = 16\n", "learning_rate = 4e-5\n", "logger.info(f'learning_rate: {learning_rate}')\n", "\n", "# hyperparameters, which are passed into the training job\n", "hyperparameters = {\n", " 'n_gpus': num_gpus, # number of GPUs per instance\n", " 'epochs': 1, # number of training epochs\n", " 'seed': 42, # random seed\n", " 'debug': True, # debug mode\n", " 'fp16': True, # use FP16\n", " 'disable_tqdm': True, # disable tqdm?\n", " 'train_batch_size': batch_size, # batch size for training\n", " 'eval_batch_size': batch_size, # batch size for evaluation\n", " 'learning_rate': learning_rate, # learning rate used during training\n", "}" ] }, { "cell_type": "code", "execution_count": 16, "id": "f95220b0-8dae-41d0-9b60-177a45a3d144", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[{3316609382.py:24} INFO - training job name: kornlp-trocr-training-2022-11-23-23-57-27\n" ] } ], "source": [ "# define Training Job Name \n", "job_name = f'kornlp-trocr-training-{time.strftime(\"%Y-%m-%d-%H-%M-%S\", time.localtime())}'\n", "chkpt_s3_path = f's3://{bucket}/{s3_prefix}/native/checkpoints'\n", "\n", "# create the Estimator\n", "sm_estimator = HuggingFace(\n", " entry_point = 'train.py', # fine-tuning script used in training jon\n", " source_dir = './scripts', # directory where fine-tuning script is stored\n", " #instance_type = 'local_gpu',\n", " instance_type = instance_type, # instances type used for the training job\n", " instance_count = instance_count, # the number of instances used for training\n", " base_job_name = job_name, # the name of the training job\n", " role = role, # IAM role used in training job to access AWS ressources, e.g. S3\n", " transformers_version = '4.17.0', # the transformers version used in the training job\n", " pytorch_version = '1.10.2', # the pytorch_version version used in the training job\n", " py_version = 'py38', # the python version used in the training job\n", " hyperparameters = hyperparameters, # the hyperparameter used for running the training job\n", " disable_profiler = True,\n", " debugger_hook_config = False, \n", " #checkpoint_s3_uri = chkpt_s3_path,\n", " #checkpoint_local_path ='/opt/ml/checkpoints', \n", ")\n", "\n", "logger.info(f'training job name: {job_name}')" ] }, { "cell_type": "markdown", "id": "60e9439c-cabe-40c4-9676-df020ac96eab", "metadata": {}, "source": [ "`fit()` 메소드를 호출하여 í›ˆë ¨ jobì„ ì‹œìž‘í•©ë‹ˆë‹¤. `fit()` ë©”ì†Œë“œì˜ ì¸ìžê°’ 중 `wait=True`ë¡œ ì„¤ì •í• ê²½ìš°ì—는 ë™ê¸°(synchronous) ë°©ì‹ìœ¼ë¡œ ë™ì§í•˜ê²Œ ë˜ë©°, `wait=False`ì¼ ê²½ìš°ì—는 비ë™ê¸°(aynchronous) ë°©ì‹ìœ¼ë¡œ ë™ìž‘하여 여러 ê°œì˜ í›ˆë ¨ jobì„ ë™ì‹œì— ì‹¤í–‰í• ìˆ˜ 있습니다." ] }, { "cell_type": "code", "execution_count": 17, "id": "39c95681-5df7-4910-8d83-2d5c32b49176", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[{image_uris.py:599} INFO - image_uri is not presented, retrieving image_uri based on instance_type, framework etc.\n", "[{session.py:609} INFO - Creating training-job with name: kornlp-trocr-training-2022-11-23-23-57--2022-11-23-23-57-28-096\n" ] } ], "source": [ "# define a data input dictonary with our uploaded s3 uris\n", "data = {\n", " 'train': train_input_path,\n", "}\n", "\n", "# starting the train job with our uploaded datasets as input\n", "sm_estimator.fit(data, wait=False)\n", "train_job_name = sm_estimator.latest_training_job.job_name" ] }, { "cell_type": "markdown", "id": "0a3c6f36-f8f0-4aaa-86cb-357deed8ae47", "metadata": {}, "source": [ "### View Training Job\n", "SageMaker 콘솔 ì°½ì—ì„œ í›ˆë ¨ ë‚´ì—ì„ ì§ì ‘ 확ì¸í• ìˆ˜ë„ ìžˆì§€ë§Œ, 아래 코드 ì…€ì—ì„œ ìƒì„±ë˜ëŠ” ë§í¬ë¥¼ í´ë¦í•˜ë©´ ë” íŽ¸ë¦¬í•˜ê²Œ í›ˆë ¨ ë‚´ì—ì„ í™•ì¸í• 수 있습니다." ] }, { "cell_type": "code", "execution_count": 18, "id": "af0de9c7-d357-4009-8105-bc99ffa317de", "metadata": {}, "outputs": [ { "data": { "text/html": [ " [Hugging Face Training - Native] Review us-east-1#/jobs/kornlp-trocr-training-2022-11-23-23-57--2022-11-23-23-57-28-096\">Training Job" ], "text/plain": [ "<IPython.core.display.HTML object>" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ " [Hugging Face Training - Native] Review us-east-1#logStream:group=/aws/sagemaker/TrainingJobs;prefix=kornlp-trocr-training-2022-11-23-23-57--2022-11-23-23-57-28-096;streamFilter=typeLogStreamPrefix\">CloudWatch Logs" ], "text/plain": [ "<IPython.core.display.HTML object>" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from IPython.core.display import display, HTML\n", "\n", "def make_console_link(region, train_job_name, train_task='[Training]'):\n", " train_job_link = f' {train_task} Review {region}#/jobs/{train_job_name}\">Training Job' \n", " cloudwatch_link = f' {train_task} Review {region}#logStream:group=/aws/sagemaker/TrainingJobs;prefix={train_job_name};streamFilter=typeLogStreamPrefix\">CloudWatch Logs'\n", " return train_job_link, cloudwatch_link \n", " \n", "train_job_link, cloudwatch_link = make_console_link(region, train_job_name, '[Hugging Face Training - Native]')\n", "\n", "display(HTML(train_job_link))\n", "display(HTML(cloudwatch_link))" ] }, { "cell_type": "markdown", "id": "46a39bec-1add-4b6d-ad38-004845cf5d48", "metadata": {}, "source": [ "### Wait for the training jobs to complete\n", "í›ˆë ¨ì´ ì™„ë£Œë 때까지 기다립니다. `estimator.fit(...)`ì—ì„œ `wait=False`ë¡œ ì„¤ì •í•œ 경우, 아래 코드 ì…€ì˜ ì£¼ì„ì„ í•´ì œ 후 실행하여 ë™ê¸° ë°©ì‹ìœ¼ë¡œ ë³€ê²½í• ìˆ˜ë„ ìžˆìŠµë‹ˆë‹¤. í›ˆë ¨ 완료까지는 약 15-20ë¶„ì˜ ì‹œê°„ì´ ì†Œìš”ë©ë‹ˆë‹¤." ] }, { "cell_type": "code", "execution_count": 19, "id": "e40addb2-097b-4159-9bfc-19e7b7dc4860", "metadata": {}, "outputs": [], "source": [ "#sess.logs_for_job(job_name=train_job_name, wait=True)" ] }, { "cell_type": "markdown", "id": "62be83ba-e279-4033-8c64-e59d499d3eb8", "metadata": {}, "source": [ "### Copy model artifacts from S3 to local path\n", "\n", "í›ˆë ¨ëœ ëª¨ë¸ íŒŒë¼ë©”터는 `model.tar.gz`ë¡œ 압축ë˜ì–´ S3ì— ì €ìž¥ë©ë‹ˆë‹¤. 만약 SageMaker ìƒì—ì„œ í›ˆë ¨í•œ 모ë¸ì„ 곧바로 ë°°í¬í•œë‹¤ë©´, 아래 코드 ì…€ì„ ì‹¤í–‰í• í•„ìš”ëŠ” 없지만, 로컬/개발 환경ì—ì„œ í›ˆë ¨ëœ ëª¨ë¸ì„ 간단히 테스트하거나 다른 환경ì—ì„œ 모ë¸ì„ ì„œë¹™í• ë•ŒëŠ” S3ì— ì €ìž¥ëœ ëª¨ë¸ì„ 다운로드하셔야 합니다." ] }, { "cell_type": "code", "execution_count": 25, "id": "d748649b-e6c0-45c5-ba58-a71c0c2f2e32", "metadata": {}, "outputs": [], "source": [ "local_model_dir = './model'\n", "!rm -rf {local_model_dir}\n", "s3_model_path = sm_estimator.model_data\n", "os.makedirs(local_model_dir, exist_ok=True)" ] }, { "cell_type": "code", "execution_count": 26, "id": "17ddeb16-51f8-49a0-bbf3-3f74e8ec8536", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "download: s3://sagemaker-us-east-1-143656149352/kornlp-trocr-training-2022-11-23-23-57--2022-11-23-23-57-28-096/output/model.tar.gz to model/model.tar.gz\n", "tokenizer_config.json\n", "config.json\n", "tokenizer.json\n", "special_tokens_map.json\n", "training_args.bin\n", "pytorch_model.bin\n", "vocab.txt\n" ] } ], "source": [ "%%bash -s \"$local_model_dir\" \"$s3_model_path\"\n", "aws s3 cp $2 $1\n", "cd $1\n", "tar -xzvf model.tar.gz" ] }, { "cell_type": "code", "execution_count": 23, "id": "3f8e32fe-7a4c-45d1-a8d8-5df1ca9b5a65", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Stored 's3_model_path' (str)\n", "Stored 'local_model_dir' (str)\n" ] } ], "source": [ "%store s3_model_path local_model_dir" ] }, { "cell_type": "markdown", "id": "5a44725e-847d-4c39-a8d3-50be3bf931e7", "metadata": {}, "source": [ "<br>\n", "\n", "## Inference\n", "---\n", "\n", "SageMaker í›ˆë ¨ ì¸ìŠ¤í„´ìŠ¤ì—ì„œ í›ˆë ¨í•œ 모ë¸ì€ ì—¬ëŸ¬ë¶„ì˜ ìœ ì¦ˆì¼€ì´ìŠ¤ì— 맞게 ìžìœ ë¡ê²Œ ì‚¬ìš©í• ìˆ˜ 있습니다. " ] }, { "cell_type": "code", "execution_count": 27, "id": "b1b77654-9aa6-4931-ae8b-2085b04f2071", "metadata": {}, "outputs": [], "source": [ "from transformers import TrOCRProcessor, VisionEncoderDecoderModel, AutoTokenizer\n", "import requests \n", "from PIL import Image\n", "\n", "processor = TrOCRProcessor.from_pretrained(\"microsoft/trocr-base-handwritten\") \n", "model = VisionEncoderDecoderModel.from_pretrained(local_model_dir)\n", "tokenizer = AutoTokenizer.from_pretrained(local_model_dir)" ] }, { "cell_type": "markdown", "id": "3af4fec1-6e66-49ad-9a4f-c855ef384a1d", "metadata": {}, "source": [ "샘플 ë°ì´í„°ë¡œ ìžìœ ë¡ê²Œ ì¶”ë¡ ì„ ìˆ˜í–‰í•´ 보세요. ì°¸ê³ ë¡œ, 샘플 ë°ì´í„° 중 ëžœë¤í•œ ë¬¸ìž¥ì€ ê²€ì¦ì…‹ì—ë„ ì—†ëŠ” ì‹ ê·œ ë°ì´í„°ìž…니다." ] }, { "cell_type": "code", "execution_count": 33, "id": "51974002-72d7-4632-a1d2-49cb5c091974", "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "<PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=1018x64 at 0x7F056FA01D00>" ] }, "execution_count": 33, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import glob\n", "sample_img_paths = glob.glob('sample_imgs/*.jpg')\n", "img_idx = np.random.randint(len(sample_img_paths))\n", "image = Image.open(sample_img_paths[img_idx])\n", "image" ] }, { "cell_type": "code", "execution_count": 34, "id": "04b67c32-6089-45f2-bdb3-bd44a0f0a4cd", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "ë…ì¼ì–´ 세미나 ë˜ ì¸ì‚¬ë§ 비난 다양하다 여주다 ë¶€ì¸ ê°ˆì¦ ìœ ë°œ\n" ] } ], "source": [ "pixel_values = processor(image, return_tensors=\"pt\").pixel_values \n", "generated_ids = model.generate(pixel_values, max_length=64)\n", "generated_text = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0] \n", "print(generated_text)" ] } ], "metadata": { "kernelspec": { "display_name": "conda_pytorch_p38", "language": "python", "name": "conda_pytorch_p38" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.8.12" } }, "nbformat": 4, "nbformat_minor": 5 }