{"cells":[{"metadata":{},"cell_type":"markdown","source":"# Let's use Different Models!\n\n## Object\n\nThis notebook is for kagglers who wants to use more models after simple Faste RCNN codes or EfficientDet.  \nI will show the another models in this notebook (Cascade Models, Different Backbones, and etc ...) \n\nI hope you enjoy and don't forget to upvote :D\n\n# Code History\n* Ver1, 2 - Start (3/1)  \n\n\n# Results\n\nCascade RCNN - Using valdiation dataset\nCOCO Eval mAP(IoU 0.5)\n\n| Model | 5 epoch  | 10 epoch |\n|---:|---:|---:|\n|ResNet50| 0.182 | 0.239 |\n"},{"metadata":{},"cell_type":"markdown","source":"\n# Simple Kernel Survey\nI tried to catch up previous kernels and summarized like this.  \nThe whole pipeline for this competetion can be divided into 3 parts.   \nIf you interested below topics, check the number and references\n![image.png](attachment:image.png)\n\n\n1. Dataset\n    * Some kernels tried to use VinbigData to COCO Format - 2)\n    * One useful kernel did comparision between Box Fusion techniques and made Dataset - 2)\n    * Some kernels tried to use Image Processing techniques and get better results - 3)\n    * Some kernels tried to find proper Image Resolution - 12)\n2. Train\n    * Some kernels use TF EfficientDet modules - 6)\n    * YoloV5 - 8)\n    * Faster RCNN - 1)\n    * Detectron2 - 7)\n3. Inference\n    * After Inference to reduce the fp, one useful kernel use 2 class filter - 5)\n    * Good EDA using YoloV5 - 10, 11)","attachments":{"image.png":{"image/png":"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"}}},{"metadata":{},"cell_type":"markdown","source":"# Reference\nI heavily borrowed the installation and config codes from 1) and used the dataset from 2)  \n\nplease upvote below references.\n\n1) [MMDET(pytorch) Framework Training: FasterRCNN Base](https://www.kaggle.com/gauravsingh1/mmdet-pytorch-framework-training-fasterrcnn-base)\n\n2) [VinBigData - Fusing Bboxes + Coco Dataset](https://www.kaggle.com/sreevishnudamodaran/vinbigdata-fusing-bboxes-coco-dataset/#data)\n\n3) [Which X-ray preprocessing performs well?](https://www.kaggle.com/kuuuuub/which-x-ray-preprocessing-performs-well)\n\n4) [x-ray image Enhancement test](https://www.kaggle.com/kuuuuub/x-ray-image-enhancement-test/data)\n\n5) [VinBigData-CXR-AD YOLOv5 14 Class [train]](https://www.kaggle.com/awsaf49/vinbigdata-cxr-ad-yolov5-14-class-train)\n\n6) [⚡VBD EfficientDET TF2 Object Detection API⚡📈](https://www.kaggle.com/sreevishnudamodaran/vbd-efficientdet-tf2-object-detection-api)\n\n7) [📸VinBigData detectron2 train](https://www.kaggle.com/corochann/vinbigdata-detectron2-train)\n\n8) [VinBigData-CXR-AD YOLOv5 14 Class [train]](https://www.kaggle.com/awsaf49/vinbigdata-cxr-ad-yolov5-14-class-train)\n\n9) [VinBigData 🌟2 Class Filter🌟](https://www.kaggle.com/awsaf49/vinbigdata-2-class-filter)\n\n10) [Quick data analysis with YOLOv5 at a glance](https://www.kaggle.com/jamsilkaggle/quick-data-analysis-with-yolov5-at-a-glance)\n\n11) [VinBigData EDA+Infer Analysis with YOLOv5](https://www.kaggle.com/kimse0ha/vinbigdata-eda-infer-analysis-with-yolov5)\n\n12) [256 vs 512 vs 1024? Which dataset is useful?](https://www.kaggle.com/seokhyunseo/256-vs-512-vs-1024-which-dataset-is-useful)"},{"metadata":{},"cell_type":"markdown","source":"# Install MMDetection"},{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"!pip -qq install mmcv-full","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"!git clone https://github.com/open-mmlab/mmdetection.git\n%cd mmdetection\n\n!pip -qq install -e .","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Download Pretrain Model"},{"metadata":{"trusted":true},"cell_type":"code","source":"# # Pretrain Model Download\n# # Cascade\n# !mkdir checkpoints\n# !wget -c http://download.openmmlab.com/mmdetection/v2.0/cascade_rcnn/cascade_rcnn_r50_caffe_fpn_1x_coco/cascade_rcnn_r50_caffe_fpn_1x_coco_bbox_mAP-0.404_20200504_174853-b857be87.pth -O checkpoints/cascade_rcnn_r50_caffe_fpn_1x_coco_bbox_mAP-0.404_20200504_174853-b857be87.pth\n\n# # RegNet\n# !mkdir checkpoints\n# !wget -c http://download.openmmlab.com/mmdetection/v2.0/regnet/faster_rcnn_regnetx-3.2GF_fpn_mstrain_3x_coco/faster_rcnn_regnetx-3.2GF_fpn_mstrain_3x_coco_20200520_224253-bf85ae3e.pth\n\n# # VFNet\n!mkdir checkpoints\n!wget -c https://openmmlab.oss-cn-hangzhou.aliyuncs.com/mmdetection/v2.0/vfnet/vfnet_r50_fpn_mdconv_c3-c5_mstrain_2x_coco/vfnet_r50_fpn_mdconv_c3-c5_mstrain_2x_coco_20201027pth-6879c318.pth\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"from mmcv import Config\nfrom mmdet.apis import set_random_seed\nfrom mmdet.datasets import build_dataset\nfrom mmdet.models import build_detector\nfrom mmdet.apis import train_detector, init_detector, inference_detector\n\nfrom IPython.display import clear_output","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Cascade RCNN Setting"},{"metadata":{"trusted":true},"cell_type":"code","source":"# ## Configuration Setting for Cascade RCNN\n# cfg = Config.fromfile('./configs/cascade_rcnn/cascade_rcnn_r50_fpn_1x_coco.py')\n# DATASET_TYPE = 'CocoDataset'\n# PREFIX = '../../input/vinbigdata-coco-dataset-with-wbf-3x-downscaled/vinbigdata-coco-dataset-with-wbf-3x-downscaled/'\n# cfg.dataset_type = DATASET_TYPE\n# cfg.classes = (\"Aortic_enlargement\", \"Atelectasis\", \n#                \"Calcification\", \"Cardiomegaly\", \n#                \"Consolidation\", \"ILD\", \"Infiltration\", \n#                \"Lung_Opacity\", \"Nodule/Mass\", \"Other_lesion\", \n#                \"Pleural_effusion\", \"Pleural_thickening\", \n#                \"Pneumothorax\", \"Pulmonary_fibrosis\")\n\n# for i in cfg.model.roi_head.bbox_head:\n#     i.num_classes = 14","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# RegNet (Best) Setting"},{"metadata":{"trusted":true},"cell_type":"code","source":"# ## Configuration Setting\n# cfg = Config.fromfile('./configs/regnet/faster_rcnn_regnetx-3.2GF_fpn_mstrain_3x_coco.py')\n# DATASET_TYPE = 'CocoDataset'\n# PREFIX = '../../input/vinbigdata-coco-dataset-with-wbf-3x-downscaled/vinbigdata-coco-dataset-with-wbf-3x-downscaled/'\n# cfg.dataset_type = DATASET_TYPE\n# cfg.classes = (\"Aortic_enlargement\", \"Atelectasis\", \n#                \"Calcification\", \"Cardiomegaly\", \n#                \"Consolidation\", \"ILD\", \"Infiltration\", \n#                \"Lung_Opacity\", \"Nodule/Mass\", \"Other_lesion\", \n#                \"Pleural_effusion\", \"Pleural_thickening\", \n#                \"Pneumothorax\", \"Pulmonary_fibrosis\")\n\n# cfg.model.roi_head.bbox_head.num_classes = 14","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# VFNet"},{"metadata":{"trusted":true},"cell_type":"code","source":"## Configuration Setting\ncfg = Config.fromfile('./configs/vfnet/vfnet_r50_fpn_mstrain_2x_coco.py')\nDATASET_TYPE = 'CocoDataset'\nPREFIX = '../../input/vinbigdata-coco-dataset-with-wbf-3x-downscaled/vinbigdata-coco-dataset-with-wbf-3x-downscaled/'\ncfg.dataset_type = DATASET_TYPE\ncfg.classes = (\"Aortic_enlargement\", \"Atelectasis\", \n               \"Calcification\", \"Cardiomegaly\", \n               \"Consolidation\", \"ILD\", \"Infiltration\", \n               \"Lung_Opacity\", \"Nodule/Mass\", \"Other_lesion\", \n               \"Pleural_effusion\", \"Pleural_thickening\", \n               \"Pneumothorax\", \"Pulmonary_fibrosis\")\n\ncfg.model.bbox_head.num_classes = 14","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"cfg.data.train.img_prefix = PREFIX\ncfg.data.train.classes = cfg.classes\ncfg.data.train.ann_file = PREFIX + 'train_annotations.json'\ncfg.data.train.type = DATASET_TYPE\n\n\ncfg.data.val.img_prefix = PREFIX\ncfg.data.val.classes = cfg.classes\ncfg.data.val.ann_file = PREFIX + 'val_annotations.json'\ncfg.data.val.type = DATASET_TYPE\n\n\n\ncfg.data.test.img_prefix = PREFIX\ncfg.data.test.classes = cfg.classes\ncfg.data.test.ann_file = PREFIX + 'val_annotations.json'\ncfg.data.test.type = DATASET_TYPE","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"cfg.optimizer.lr = 0.02 / 8\ncfg.lr_config.warmup = None\ncfg.log_config.interval = 100\n\n# Change the evaluation metric since we use customized dataset.\ncfg.evaluation.metric = 'bbox'\n# We can set the evaluation interval to reduce the evaluation times\ncfg.evaluation.interval = 5\n# We can set the checkpoint saving interval to reduce the storage cost\ncfg.checkpoint_config.interval = 5\n\n# Set seed thus the results are more reproducible\ncfg.seed = 0\nset_random_seed(0, deterministic=False)\ncfg.gpu_ids = range(1)\n\n# we can use here mask_rcnn.\n# cfg.load_from = './checkpoints/cascade_rcnn_r50_caffe_fpn_1x_coco_bbox_mAP-0.404_20200504_174853-b857be87.pth'\n# cfg.load_from ='faster_rcnn_regnetx-3.2GF_fpn_mstrain_3x_coco_20200520_224253-bf85ae3e.pth'\ncfg.load_from = 'vfnet_r50_fpn_mdconv_c3-c5_mstrain_2x_coco_20201027pth-6879c318.pth'\ncfg.work_dir = \"../vinbig_output\"\n\ncfg.runner.max_epochs = 12\ncfg.total_epochs = 12","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"clear_output()\nmodel = build_detector(cfg.model)\ndatasets = [build_dataset(cfg.data.train)]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import torch \n","execution_count":null,"outputs":[]},{"metadata":{"_kg_hide-output":true,"trusted":true},"cell_type":"code","source":"train_detector(model, datasets[0], cfg, distributed=False, validate=True)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Inference Validation "},{"metadata":{"trusted":true},"cell_type":"code","source":"import os\nos.chdir('../')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"!python ./mmdetection/tools/analysis_tools/analyze_logs.py plot_curve ./vinbig_output/None.log.json --keys s2.loss_cls --legend s2.loss_cls --out \"loss_cls.jpg\"\n!rm -rf \"./mmdetection\"","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat":4,"nbformat_minor":4}