内存泄漏导致进程被杀死 #1070
Labels
bug
Something isn't working
Comments
|
内存泄露，占用到一定量后，进程被杀死： |
|
装paddlegpu了吗？ |
装了 |
|
出问题的文件可以上传一份给我们复测吗 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Description of the bug | 错误描述
在处理一个PDF页数为379页的文件时，模型会不断占用内存，达到一定量后被oom杀死（本地机子上也会不断占用内存，最后出现MemoryError）。
magic-pdf --version 0.10.0
运行时内存占用如下：
显存占用如下：
求大佬赐教，非常着急，感谢感谢！！！
How to reproduce the bug | 如何复现
以下是日志文件内容：
WARNING: OMP_NUM_THREADS set to 12, not 1. The computation speed will not be optimized if you use data parallel. It will fail if this PaddlePaddle binary is compiled with OpenBlas since OpenBlas does not support multi-threads.
PLEASE USE OMP_NUM_THREADS WISELY.
import tensorrt_llm failed, if do not use tensorrt, ignore this message
import lmdeploy failed, if do not use lmdeploy, ignore this message
2024-11-23 20:56:07.879 | INFO | magic_pdf.libs.pdf_check:detect_invalid_chars:57 - cid_count: 0, text_len: 10, cid_chars_radio: 0.0
2024-11-23 20:56:07.880 | WARNING | magic_pdf.filter.pdf_classify_by_type:classify:334 - pdf is not classified by area and text_len, by_image_area: False, by_text: False, by_avg_words: False, by_img_num: True, by_text_layout: False, by_img_narrow_strips: True, by_invalid_chars: True
2024-11-23 20:56:07.882 | INFO | magic_pdf.model.pdf_extract_kit:init:78 - DocAnalysis init, this may take some times, layout_model: layoutlmv3, apply_formula: True, apply_ocr: True, apply_table: False, table_model: rapid_table, lang: None
2024-11-23 20:56:07.882 | INFO | magic_pdf.model.pdf_extract_kit:init:91 - using device: cuda
2024-11-23 20:56:07.882 | INFO | magic_pdf.model.pdf_extract_kit:init:95 - using models_dir: /root/.cache/modelscope/hub/opendatalab/PDF-Extract-Kit-1___0/models
CustomVisionEncoderDecoderModel init
VariableUnimerNetModel init
VariableUnimerNetPatchEmbeddings init
VariableUnimerNetModel init
VariableUnimerNetPatchEmbeddings init
CustomMBartForCausalLM init
CustomMBartDecoder init
[11/23 20:56:18 detectron2]: Rank of current process: 0. World size: 1
[11/23 20:56:18 detectron2]: Environment info:
sys.platform linux
Python 3.10.15 (main, Oct 3 2024, 07:27:34) [GCC 11.2.0]
numpy 1.26.4
detectron2 0.6 @/root/miniconda3/envs/MinerU/lib/python3.10/site-packages/detectron2
Compiler GCC 11.4
CUDA compiler not available
DETECTRON2_ENV_MODULE
PyTorch 2.3.1+cu121 @/root/miniconda3/envs/MinerU/lib/python3.10/site-packages/torch
PyTorch debug build False
torch._C._GLIBCXX_USE_CXX11_ABI False
GPU available Yes
GPU 0 NVIDIA GeForce RTX 3080 (arch=8.6)
Driver version 550.120
CUDA_HOME /usr/local/cuda
Pillow 11.0.0
torchvision 0.18.1+cu121 @/root/miniconda3/envs/MinerU/lib/python3.10/site-packages/torchvision
torchvision arch flags 5.0, 6.0, 7.0, 7.5, 8.0, 8.6, 9.0
fvcore 0.1.5.post20221221
iopath 0.1.9
cv2 4.6.0
PyTorch built with:
[11/23 20:56:18 detectron2]: Command line arguments: {'config_file': '/root/miniconda3/envs/MinerU/lib/python3.10/site-packages/magic_pdf/resources/model_config/layoutlmv3/layoutlmv3_base_inference.yaml', 'resume': False, 'eval_only': False, 'num_gpus': 1, 'num_machines': 1, 'machine_rank': 0, 'dist_url': 'tcp://127.0.0.1:57823', 'opts': ['MODEL.WEIGHTS', '/root/.cache/modelscope/hub/opendatalab/PDF-Extract-Kit-1___0/models/Layout/LayoutLMv3/model_final.pth']}
[11/23 20:56:18 detectron2]: Contents of args.config_file=/root/miniconda3/envs/MinerU/lib/python3.10/site-packages/magic_pdf/resources/model_config/layoutlmv3/layoutlmv3_base_inference.yaml:
AUG:
DETR: true
CACHE_DIR: ~/cache/huggingface
CUDNN_BENCHMARK: false
DATALOADER:
ASPECT_RATIO_GROUPING: true
FILTER_EMPTY_ANNOTATIONS: false
NUM_WORKERS: 4
REPEAT_THRESHOLD: 0.0
SAMPLER_TRAIN: TrainingSampler
DATASETS:
PRECOMPUTED_PROPOSAL_TOPK_TEST: 1000
PRECOMPUTED_PROPOSAL_TOPK_TRAIN: 2000
PROPOSAL_FILES_TEST: []
PROPOSAL_FILES_TRAIN: []
TEST:
TRAIN:
GLOBAL:
HACK: 1.0
ICDAR_DATA_DIR_TEST: ''
ICDAR_DATA_DIR_TRAIN: ''
INPUT:
CROP:
ENABLED: true
SIZE:
TYPE: absolute_range
FORMAT: RGB
MASK_FORMAT: polygon
MAX_SIZE_TEST: 1333
MAX_SIZE_TRAIN: 1333
MIN_SIZE_TEST: 800
MIN_SIZE_TRAIN:
MIN_SIZE_TRAIN_SAMPLING: choice
RANDOM_FLIP: horizontal
MODEL:
ANCHOR_GENERATOR:
ANGLES:
ASPECT_RATIOS:
NAME: DefaultAnchorGenerator
OFFSET: 0.0
SIZES:
BACKBONE:
FREEZE_AT: 2
NAME: build_vit_fpn_backbone
CONFIG_PATH: ''
DEVICE: cuda
FPN:
FUSE_TYPE: sum
IN_FEATURES:
NORM: ''
OUT_CHANNELS: 256
IMAGE_ONLY: true
KEYPOINT_ON: false
LOAD_PROPOSALS: false
MASK_ON: true
META_ARCHITECTURE: VLGeneralizedRCNN
PANOPTIC_FPN:
COMBINE:
ENABLED: true
INSTANCES_CONFIDENCE_THRESH: 0.5
OVERLAP_THRESH: 0.5
STUFF_AREA_LIMIT: 4096
INSTANCE_LOSS_WEIGHT: 1.0
PIXEL_MEAN:
PIXEL_STD:
PROPOSAL_GENERATOR:
MIN_SIZE: 0
NAME: RPN
RESNETS:
DEFORM_MODULATED: false
DEFORM_NUM_GROUPS: 1
DEFORM_ON_PER_STAGE:
DEPTH: 50
NORM: FrozenBN
NUM_GROUPS: 1
OUT_FEATURES:
RES2_OUT_CHANNELS: 256
RES5_DILATION: 1
STEM_OUT_CHANNELS: 64
STRIDE_IN_1X1: true
WIDTH_PER_GROUP: 64
RETINANET:
BBOX_REG_LOSS_TYPE: smooth_l1
BBOX_REG_WEIGHTS:
FOCAL_LOSS_ALPHA: 0.25
FOCAL_LOSS_GAMMA: 2.0
IN_FEATURES:
IOU_LABELS:
IOU_THRESHOLDS:
NMS_THRESH_TEST: 0.5
NORM: ''
NUM_CLASSES: 10
NUM_CONVS: 4
PRIOR_PROB: 0.01
SCORE_THRESH_TEST: 0.05
SMOOTH_L1_LOSS_BETA: 0.1
TOPK_CANDIDATES_TEST: 1000
ROI_BOX_CASCADE_HEAD:
BBOX_REG_WEIGHTS:
IOUS:
ROI_BOX_HEAD:
BBOX_REG_LOSS_TYPE: smooth_l1
BBOX_REG_LOSS_WEIGHT: 1.0
BBOX_REG_WEIGHTS:
CLS_AGNOSTIC_BBOX_REG: true
CONV_DIM: 256
FC_DIM: 1024
NAME: FastRCNNConvFCHead
NORM: ''
NUM_CONV: 0
NUM_FC: 2
POOLER_RESOLUTION: 7
POOLER_SAMPLING_RATIO: 0
POOLER_TYPE: ROIAlignV2
SMOOTH_L1_BETA: 0.0
TRAIN_ON_PRED_BOXES: false
ROI_HEADS:
BATCH_SIZE_PER_IMAGE: 512
IN_FEATURES:
IOU_LABELS:
IOU_THRESHOLDS:
NAME: CascadeROIHeads
NMS_THRESH_TEST: 0.5
NUM_CLASSES: 10
POSITIVE_FRACTION: 0.25
PROPOSAL_APPEND_GT: true
SCORE_THRESH_TEST: 0.05
ROI_KEYPOINT_HEAD:
CONV_DIMS:
LOSS_WEIGHT: 1.0
MIN_KEYPOINTS_PER_IMAGE: 1
NAME: KRCNNConvDeconvUpsampleHead
NORMALIZE_LOSS_BY_VISIBLE_KEYPOINTS: true
NUM_KEYPOINTS: 17
POOLER_RESOLUTION: 14
POOLER_SAMPLING_RATIO: 0
POOLER_TYPE: ROIAlignV2
ROI_MASK_HEAD:
CLS_AGNOSTIC_MASK: false
CONV_DIM: 256
NAME: MaskRCNNConvUpsampleHead
NORM: ''
NUM_CONV: 4
POOLER_RESOLUTION: 14
POOLER_SAMPLING_RATIO: 0
POOLER_TYPE: ROIAlignV2
RPN:
BATCH_SIZE_PER_IMAGE: 256
BBOX_REG_LOSS_TYPE: smooth_l1
BBOX_REG_LOSS_WEIGHT: 1.0
BBOX_REG_WEIGHTS:
BOUNDARY_THRESH: -1
CONV_DIMS:
HEAD_NAME: StandardRPNHead
IN_FEATURES:
IOU_LABELS:
IOU_THRESHOLDS:
LOSS_WEIGHT: 1.0
NMS_THRESH: 0.7
POSITIVE_FRACTION: 0.5
POST_NMS_TOPK_TEST: 1000
POST_NMS_TOPK_TRAIN: 2000
PRE_NMS_TOPK_TEST: 1000
PRE_NMS_TOPK_TRAIN: 2000
SMOOTH_L1_BETA: 0.0
SEM_SEG_HEAD:
COMMON_STRIDE: 4
CONVS_DIM: 128
IGNORE_VALUE: 255
IN_FEATURES:
LOSS_WEIGHT: 1.0
NAME: SemSegFPNHead
NORM: GN
NUM_CLASSES: 10
VIT:
DROP_PATH: 0.1
IMG_SIZE:
NAME: layoutlmv3_base
OUT_FEATURES:
POS_TYPE: abs
WEIGHTS:
OUTPUT_DIR:
SCIHUB_DATA_DIR_TRAIN: ~/publaynet/layout_scihub/train
SEED: 42
SOLVER:
AMP:
ENABLED: true
BACKBONE_MULTIPLIER: 1.0
BASE_LR: 0.0002
BIAS_LR_FACTOR: 1.0
CHECKPOINT_PERIOD: 2000
CLIP_GRADIENTS:
CLIP_TYPE: full_model
CLIP_VALUE: 1.0
ENABLED: true
NORM_TYPE: 2.0
GAMMA: 0.1
GRADIENT_ACCUMULATION_STEPS: 1
IMS_PER_BATCH: 32
LR_SCHEDULER_NAME: WarmupCosineLR
MAX_ITER: 20000
MOMENTUM: 0.9
NESTEROV: false
OPTIMIZER: ADAMW
REFERENCE_WORLD_SIZE: 0
STEPS:
WARMUP_FACTOR: 0.01
WARMUP_ITERS: 333
WARMUP_METHOD: linear
WEIGHT_DECAY: 0.05
WEIGHT_DECAY_BIAS: null
WEIGHT_DECAY_NORM: 0.0
TEST:
AUG:
ENABLED: false
FLIP: true
MAX_SIZE: 4000
MIN_SIZES:
DETECTIONS_PER_IMAGE: 100
EVAL_PERIOD: 1000
EXPECTED_RESULTS: []
KEYPOINT_OKS_SIGMAS: []
PRECISE_BN:
ENABLED: false
NUM_ITER: 200
VERSION: 2
VIS_PERIOD: 0
[11/23 20:56:21 d2.checkpoint.detection_checkpoint]: [DetectionCheckpointer] Loading from /root/.cache/modelscope/hub/opendatalab/PDF-Extract-Kit-1___0/models/Layout/LayoutLMv3/model_final.pth ...
[11/23 20:56:21 fvcore.common.checkpoint]: [Checkpointer] Loading from /root/.cache/modelscope/hub/opendatalab/PDF-Extract-Kit-1___0/models/Layout/LayoutLMv3/model_final.pth ...
2024-11-23 20:56:23.160 | INFO | magic_pdf.model.pdf_extract_kit:init:170 - DocAnalysis init done!
2024-11-23 20:56:23.161 | INFO | magic_pdf.model.doc_analyze_by_custom_model:custom_model_init:131 - model init cost: 15.280268430709839
Killed
Operating system | 操作系统
Linux
Python version | Python 版本
3.10
Software version | 软件版本 (magic-pdf --version)
0.9.x
Device mode | 设备模式
cuda
The text was updated successfully, but these errors were encountered: