first try

app.py

@@ -0,0 +1,75 @@
+#try:
+#    import detectron2
+#except:
+import os 
+os.system('pip install git+https://github.com/SysCV/transfiner.git')
+
+from matplotlib.pyplot import axis
+import gradio as gr
+import requests
+import numpy as np
+from torch import nn
+import requests
+
+import torch
+
+from detectron2 import model_zoo
+from detectron2.engine import DefaultPredictor
+from detectron2.config import get_cfg
+from detectron2.utils.visualizer import Visualizer
+from detectron2.data import MetadataCatalog
+
+'''
+url1 = 'https://cdn.pixabay.com/photo/2014/09/07/21/52/city-438393_1280.jpg'
+r = requests.get(url1, allow_redirects=True)
+open("city1.jpg", 'wb').write(r.content)
+url2 = 'https://cdn.pixabay.com/photo/2016/02/19/11/36/canal-1209808_1280.jpg'
+r = requests.get(url2, allow_redirects=True)
+open("city2.jpg", 'wb').write(r.content)
+'''
+
+model_name='./configs/COCO-InstanceSegmentation/mask_rcnn_X_101_32x8d_FPN_3x.yaml'
+
+# model = model_zoo.get(model_name, trained=True)
+
+cfg = get_cfg()
+# add project-specific config (e.g., TensorMask) here if you're not running a model in detectron2's core library
+cfg.merge_from_file(model_zoo.get_config_file(model_name))
+cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.5  # set threshold for this model
+# Find a model from detectron2's model zoo. You can use the https://dl.fbaipublicfiles... url as w ell
+cfg.MODEL.WEIGHTS = model_zoo.get_checkpoint_url(model_name)
+
+if not torch.cuda.is_available():
+    cfg.MODEL.DEVICE='cpu'
+
+predictor = DefaultPredictor(cfg)
+
+
+def inference(image):
+    img = np.array(image.resize((1024,1024)))
+    outputs = predictor(img)
+
+    v = Visualizer(img, MetadataCatalog.get(cfg.DATASETS.TRAIN[0]), scale=1.2)
+    out = v.draw_instance_predictions(outputs["instances"].to("cpu"))
+    
+    return out.get_image()
+
+
+
+title = "Detectron2-MaskRCNN X101"
+description = "demo for Detectron2. To use it, simply upload your image, or click one of the examples to load them. Read more at the links below.\
+    </br><b>Model: COCO-InstanceSegmentation/mask_rcnn_X_101_32x8d_FPN_3x.yaml</b>"
+article = "<p style='text-align: center'><a href='https://arxiv.org/abs/2012.07177'>Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentation</a> | <a href='https://github.com/facebookresearch/detectron2/blob/main/MODEL_ZOO.md'>Detectron model ZOO</a></p>"
+
+gr.Interface(
+    inference, 
+    [gr.inputs.Image(type="pil", label="Input")], 
+    gr.outputs.Image(type="numpy", label="Output"),
+    title=title,
+    description=description,
+    article=article,
+    examples=[
+            ["demo/sample_imgs/000000224200.jpg"],
+            ["demo/sample_imgs/000000344909.jpg"]
+        ]).launch()
+

configs/Base-RCNN-C4.yaml

@@ -0,0 +1,18 @@
+MODEL:
+  META_ARCHITECTURE: "GeneralizedRCNN"
+  RPN:
+    PRE_NMS_TOPK_TEST: 6000
+    POST_NMS_TOPK_TEST: 1000
+  ROI_HEADS:
+    NAME: "Res5ROIHeads"
+DATASETS:
+  TRAIN: ("coco_2017_train",)
+  TEST: ("coco_2017_val",)
+SOLVER:
+  IMS_PER_BATCH: 16
+  BASE_LR: 0.02
+  STEPS: (60000, 80000)
+  MAX_ITER: 90000
+INPUT:
+  MIN_SIZE_TRAIN: (640, 672, 704, 736, 768, 800)
+VERSION: 2

configs/Base-RCNN-DilatedC5.yaml

@@ -0,0 +1,31 @@
+MODEL:
+  META_ARCHITECTURE: "GeneralizedRCNN"
+  RESNETS:
+    OUT_FEATURES: ["res5"]
+    RES5_DILATION: 2
+  RPN:
+    IN_FEATURES: ["res5"]
+    PRE_NMS_TOPK_TEST: 6000
+    POST_NMS_TOPK_TEST: 1000
+  ROI_HEADS:
+    NAME: "StandardROIHeads"
+    IN_FEATURES: ["res5"]
+  ROI_BOX_HEAD:
+    NAME: "FastRCNNConvFCHead"
+    NUM_FC: 2
+    POOLER_RESOLUTION: 7
+  ROI_MASK_HEAD:
+    NAME: "MaskRCNNConvUpsampleHead"
+    NUM_CONV: 4
+    POOLER_RESOLUTION: 14
+DATASETS:
+  TRAIN: ("coco_2017_train",)
+  TEST: ("coco_2017_val",)
+SOLVER:
+  IMS_PER_BATCH: 16
+  BASE_LR: 0.02
+  STEPS: (60000, 80000)
+  MAX_ITER: 90000
+INPUT:
+  MIN_SIZE_TRAIN: (640, 672, 704, 736, 768, 800)
+VERSION: 2

configs/Base-RCNN-FPN-4gpu.yaml

@@ -0,0 +1,44 @@
+MODEL:
+  META_ARCHITECTURE: "GeneralizedRCNN"
+  BACKBONE:
+    NAME: "build_resnet_fpn_backbone"
+  RESNETS:
+    OUT_FEATURES: ["res2", "res3", "res4", "res5"]
+  FPN:
+    IN_FEATURES: ["res2", "res3", "res4", "res5"]
+  ANCHOR_GENERATOR:
+    SIZES: [[32], [64], [128], [256], [512]]  # One size for each in feature map
+    ASPECT_RATIOS: [[0.5, 1.0, 2.0]]  # Three aspect ratios (same for all in feature maps)
+  RPN:
+    IN_FEATURES: ["p2", "p3", "p4", "p5", "p6"]
+    PRE_NMS_TOPK_TRAIN: 2000  # Per FPN level
+    PRE_NMS_TOPK_TEST: 1000  # Per FPN level
+    # Detectron1 uses 2000 proposals per-batch,
+    # (See "modeling/rpn/rpn_outputs.py" for details of this legacy issue)
+    # which is approximately 1000 proposals per-image since the default batch size for FPN is 2.
+    POST_NMS_TOPK_TRAIN: 1000
+    POST_NMS_TOPK_TEST: 1000
+  ROI_HEADS:
+    NAME: "StandardROIHeads"
+    IN_FEATURES: ["p2", "p3", "p4", "p5"]
+  ROI_BOX_HEAD:
+    NAME: "FastRCNNConvFCHead"
+    NUM_FC: 2
+    POOLER_RESOLUTION: 7
+  ROI_MASK_HEAD:
+    NAME: "MaskRCNNConvUpsampleHead"
+    NUM_CONV: 4
+    POOLER_RESOLUTION: 14
+DATASETS:
+  TRAIN: ("coco_2017_train",)
+  #TEST: ("coco_2017_val",)
+  #TEST: ("lvis_v0.5_val_cocofied",)
+  TEST: ("coco_2017_test-dev",)
+SOLVER:
+  IMS_PER_BATCH: 16 #8 #16
+  BASE_LR: 0.02 # 0.02
+  STEPS: (60000, 80000)
+  MAX_ITER: 90000
+INPUT:
+  MIN_SIZE_TRAIN: (640, 672, 704, 736, 768, 800)
+VERSION: 2

configs/Base-RCNN-FPN.yaml

@@ -0,0 +1,42 @@
+MODEL:
+  META_ARCHITECTURE: "GeneralizedRCNN"
+  BACKBONE:
+    NAME: "build_resnet_fpn_backbone"
+  RESNETS:
+    OUT_FEATURES: ["res2", "res3", "res4", "res5"]
+  FPN:
+    IN_FEATURES: ["res2", "res3", "res4", "res5"]
+  ANCHOR_GENERATOR:
+    SIZES: [[32], [64], [128], [256], [512]]  # One size for each in feature map
+    ASPECT_RATIOS: [[0.5, 1.0, 2.0]]  # Three aspect ratios (same for all in feature maps)
+  RPN:
+    IN_FEATURES: ["p2", "p3", "p4", "p5", "p6"]
+    PRE_NMS_TOPK_TRAIN: 2000  # Per FPN level
+    PRE_NMS_TOPK_TEST: 1000  # Per FPN level
+    # Detectron1 uses 2000 proposals per-batch,
+    # (See "modeling/rpn/rpn_outputs.py" for details of this legacy issue)
+    # which is approximately 1000 proposals per-image since the default batch size for FPN is 2.
+    POST_NMS_TOPK_TRAIN: 1000
+    POST_NMS_TOPK_TEST: 1000
+  ROI_HEADS:
+    NAME: "StandardROIHeads"
+    IN_FEATURES: ["p2", "p3", "p4", "p5"]
+  ROI_BOX_HEAD:
+    NAME: "FastRCNNConvFCHead"
+    NUM_FC: 2
+    POOLER_RESOLUTION: 7
+  ROI_MASK_HEAD:
+    NAME: "MaskRCNNConvUpsampleHead"
+    NUM_CONV: 4
+    POOLER_RESOLUTION: 14
+DATASETS:
+  TRAIN: ("coco_2017_train",)
+  TEST: ("coco_2017_val",)
+SOLVER:
+  IMS_PER_BATCH: 16 #16
+  BASE_LR: 0.02
+  STEPS: (60000, 80000)
+  MAX_ITER: 90000
+INPUT:
+  MIN_SIZE_TRAIN: (640, 672, 704, 736, 768, 800)
+VERSION: 2

configs/Base-RetinaNet.yaml

@@ -0,0 +1,25 @@
+MODEL:
+  META_ARCHITECTURE: "RetinaNet"
+  BACKBONE:
+    NAME: "build_retinanet_resnet_fpn_backbone"
+  RESNETS:
+    OUT_FEATURES: ["res3", "res4", "res5"]
+  ANCHOR_GENERATOR:
+    SIZES: !!python/object/apply:eval ["[[x, x * 2**(1.0/3), x * 2**(2.0/3) ] for x in [32, 64, 128, 256, 512 ]]"]
+  FPN:
+    IN_FEATURES: ["res3", "res4", "res5"]
+  RETINANET:
+    IOU_THRESHOLDS: [0.4, 0.5]
+    IOU_LABELS: [0, -1, 1]
+    SMOOTH_L1_LOSS_BETA: 0.0
+DATASETS:
+  TRAIN: ("coco_2017_train",)
+  TEST: ("coco_2017_val",)
+SOLVER:
+  IMS_PER_BATCH: 16
+  BASE_LR: 0.01  # Note that RetinaNet uses a different default learning rate
+  STEPS: (60000, 80000)
+  MAX_ITER: 90000
+INPUT:
+  MIN_SIZE_TRAIN: (640, 672, 704, 736, 768, 800)
+VERSION: 2

configs/COCO-Detection/fast_rcnn_R_50_FPN_1x.yaml

@@ -0,0 +1,17 @@
+_BASE_: "../Base-RCNN-FPN.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-50.pkl"
+  MASK_ON: False
+  LOAD_PROPOSALS: True
+  RESNETS:
+    DEPTH: 50
+  PROPOSAL_GENERATOR:
+    NAME: "PrecomputedProposals"
+DATASETS:
+  TRAIN: ("coco_2017_train",)
+  PROPOSAL_FILES_TRAIN: ("detectron2://COCO-Detection/rpn_R_50_FPN_1x/137258492/coco_2017_train_box_proposals_21bc3a.pkl", )
+  TEST: ("coco_2017_val",)
+  PROPOSAL_FILES_TEST: ("detectron2://COCO-Detection/rpn_R_50_FPN_1x/137258492/coco_2017_val_box_proposals_ee0dad.pkl", )
+DATALOADER:
+  # proposals are part of the dataset_dicts, and take a lot of RAM
+  NUM_WORKERS: 2

configs/COCO-Detection/faster_rcnn_R_101_C4_3x.yaml

@@ -0,0 +1,9 @@
+_BASE_: "../Base-RCNN-C4.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-101.pkl"
+  MASK_ON: False
+  RESNETS:
+    DEPTH: 101
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-Detection/faster_rcnn_R_101_DC5_3x.yaml

@@ -0,0 +1,9 @@
+_BASE_: "../Base-RCNN-DilatedC5.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-101.pkl"
+  MASK_ON: False
+  RESNETS:
+    DEPTH: 101
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-Detection/faster_rcnn_R_101_FPN_3x.yaml

@@ -0,0 +1,9 @@
+_BASE_: "../Base-RCNN-FPN.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-101.pkl"
+  MASK_ON: False
+  RESNETS:
+    DEPTH: 101
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-Detection/faster_rcnn_R_50_C4_1x.yaml

@@ -0,0 +1,6 @@
+_BASE_: "../Base-RCNN-C4.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-50.pkl"
+  MASK_ON: False
+  RESNETS:
+    DEPTH: 50

configs/COCO-Detection/faster_rcnn_R_50_C4_3x.yaml

@@ -0,0 +1,9 @@
+_BASE_: "../Base-RCNN-C4.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-50.pkl"
+  MASK_ON: False
+  RESNETS:
+    DEPTH: 50
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-Detection/faster_rcnn_R_50_DC5_1x.yaml

@@ -0,0 +1,6 @@
+_BASE_: "../Base-RCNN-DilatedC5.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-50.pkl"
+  MASK_ON: False
+  RESNETS:
+    DEPTH: 50

configs/COCO-Detection/faster_rcnn_R_50_DC5_3x.yaml

@@ -0,0 +1,9 @@
+_BASE_: "../Base-RCNN-DilatedC5.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-50.pkl"
+  MASK_ON: False
+  RESNETS:
+    DEPTH: 50
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-Detection/faster_rcnn_R_50_FPN_1x.yaml

@@ -0,0 +1,6 @@
+_BASE_: "../Base-RCNN-FPN.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-50.pkl"
+  MASK_ON: False
+  RESNETS:
+    DEPTH: 50

configs/COCO-Detection/faster_rcnn_R_50_FPN_3x.yaml

@@ -0,0 +1,9 @@
+_BASE_: "../Base-RCNN-FPN.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-50.pkl"
+  MASK_ON: False
+  RESNETS:
+    DEPTH: 50
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-Detection/faster_rcnn_X_101_32x8d_FPN_3x.yaml

@@ -0,0 +1,13 @@
+_BASE_: "../Base-RCNN-FPN.yaml"
+MODEL:
+  MASK_ON: False
+  WEIGHTS: "detectron2://ImageNetPretrained/FAIR/X-101-32x8d.pkl"
+  PIXEL_STD: [57.375, 57.120, 58.395]
+  RESNETS:
+    STRIDE_IN_1X1: False  # this is a C2 model
+    NUM_GROUPS: 32
+    WIDTH_PER_GROUP: 8
+    DEPTH: 101
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-Detection/retinanet_R_101_FPN_3x.yaml

@@ -0,0 +1,8 @@
+_BASE_: "../Base-RetinaNet.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-101.pkl"
+  RESNETS:
+    DEPTH: 101
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-Detection/retinanet_R_50_FPN_1x.py

@@ -0,0 +1,9 @@
+from ..common.optim import SGD as optimizer
+from ..common.coco_schedule import lr_multiplier_1x as lr_multiplier
+from ..common.data.coco import dataloader
+from ..common.models.retinanet import model
+from ..common.train import train
+
+dataloader.train.mapper.use_instance_mask = False
+model.backbone.bottom_up.freeze_at = 2
+optimizer.lr = 0.01

configs/COCO-Detection/retinanet_R_50_FPN_1x.yaml

@@ -0,0 +1,5 @@
+_BASE_: "../Base-RetinaNet.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-50.pkl"
+  RESNETS:
+    DEPTH: 50

configs/COCO-Detection/retinanet_R_50_FPN_3x.yaml

@@ -0,0 +1,8 @@
+_BASE_: "../Base-RetinaNet.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-50.pkl"
+  RESNETS:
+    DEPTH: 50
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-Detection/rpn_R_50_C4_1x.yaml

@@ -0,0 +1,10 @@
+_BASE_: "../Base-RCNN-C4.yaml"
+MODEL:
+  META_ARCHITECTURE: "ProposalNetwork"
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-50.pkl"
+  MASK_ON: False
+  RESNETS:
+    DEPTH: 50
+  RPN:
+    PRE_NMS_TOPK_TEST: 12000
+    POST_NMS_TOPK_TEST: 2000

configs/COCO-Detection/rpn_R_50_FPN_1x.yaml

@@ -0,0 +1,9 @@
+_BASE_: "../Base-RCNN-FPN.yaml"
+MODEL:
+  META_ARCHITECTURE: "ProposalNetwork"
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-50.pkl"
+  MASK_ON: False
+  RESNETS:
+    DEPTH: 50
+  RPN:
+    POST_NMS_TOPK_TEST: 2000

configs/COCO-InstanceSegmentation/mask_rcnn_R_101_C4_3x.yaml

@@ -0,0 +1,9 @@
+_BASE_: "../Base-RCNN-C4.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-101.pkl"
+  MASK_ON: True
+  RESNETS:
+    DEPTH: 101
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-InstanceSegmentation/mask_rcnn_R_101_DC5_3x.yaml

@@ -0,0 +1,9 @@
+_BASE_: "../Base-RCNN-DilatedC5.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-101.pkl"
+  MASK_ON: True
+  RESNETS:
+    DEPTH: 101
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-InstanceSegmentation/mask_rcnn_R_101_FPN_3x.yaml

@@ -0,0 +1,9 @@
+_BASE_: "../Base-RCNN-FPN.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-101.pkl"
+  MASK_ON: True
+  RESNETS:
+    DEPTH: 101
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-InstanceSegmentation/mask_rcnn_R_101_FPN_3x_4gpu_transfiner.yaml

@@ -0,0 +1,9 @@
+_BASE_: "../Base-RCNN-FPN-4gpu.yaml"
+MODEL:
+  WEIGHTS: "./init_weights/model_final_a3ec72.pkl"
+  MASK_ON: True
+  RESNETS:
+    DEPTH: 101
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-InstanceSegmentation/mask_rcnn_R_101_FPN_3x_4gpu_transfiner_deform.yaml

@@ -0,0 +1,11 @@
+_BASE_: "../Base-RCNN-FPN-4gpu.yaml"
+MODEL:
+  WEIGHTS: "./init_weights/model_final_a3ec72.pkl"
+  MASK_ON: True
+  RESNETS:
+    DEPTH: 101
+    DEFORM_ON_PER_STAGE: [False, True, True, True] # on Res3,Res4,Res5
+    DEFORM_MODULATED: False
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-InstanceSegmentation/mask_rcnn_R_101_FPN_3x_4gpu_transfiner_lvis.yaml

@@ -0,0 +1,12 @@
+_BASE_: "../Base-RCNN-FPN-4gpu.yaml"
+MODEL:
+  WEIGHTS: "./init_weights/model_final_a3ec72.pkl"
+  MASK_ON: True
+  RESNETS:
+    DEPTH: 101
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000
+DATASETS:
+  TEST: ("lvis_v0.5_val_cocofied",)
+

configs/COCO-InstanceSegmentation/mask_rcnn_R_50_C4_1x.py

@@ -0,0 +1,7 @@
+from ..common.train import train
+from ..common.optim import SGD as optimizer
+from ..common.coco_schedule import lr_multiplier_1x as lr_multiplier
+from ..common.data.coco import dataloader
+from ..common.models.mask_rcnn_c4 import model
+
+model.backbone.freeze_at = 2

configs/COCO-InstanceSegmentation/mask_rcnn_R_50_C4_1x.yaml

@@ -0,0 +1,6 @@
+_BASE_: "../Base-RCNN-C4.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-50.pkl"
+  MASK_ON: True
+  RESNETS:
+    DEPTH: 50

configs/COCO-InstanceSegmentation/mask_rcnn_R_50_C4_3x.yaml

@@ -0,0 +1,9 @@
+_BASE_: "../Base-RCNN-C4.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-50.pkl"
+  MASK_ON: True
+  RESNETS:
+    DEPTH: 50
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-InstanceSegmentation/mask_rcnn_R_50_DC5_1x.yaml

@@ -0,0 +1,6 @@
+_BASE_: "../Base-RCNN-DilatedC5.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-50.pkl"
+  MASK_ON: True
+  RESNETS:
+    DEPTH: 50

configs/COCO-InstanceSegmentation/mask_rcnn_R_50_DC5_3x.yaml

@@ -0,0 +1,9 @@
+_BASE_: "../Base-RCNN-DilatedC5.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-50.pkl"
+  MASK_ON: True
+  RESNETS:
+    DEPTH: 50
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_1x.py

@@ -0,0 +1,7 @@
+from ..common.optim import SGD as optimizer
+from ..common.coco_schedule import lr_multiplier_1x as lr_multiplier
+from ..common.data.coco import dataloader
+from ..common.models.mask_rcnn_fpn import model
+from ..common.train import train
+
+model.backbone.bottom_up.freeze_at = 2

configs/COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_1x.yaml

@@ -0,0 +1,6 @@
+_BASE_: "../Base-RCNN-FPN.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-50.pkl"
+  MASK_ON: True
+  RESNETS:
+    DEPTH: 50

configs/COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_1x_4gpu.yaml

@@ -0,0 +1,6 @@
+_BASE_: "../Base-RCNN-FPN-4gpu.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-50.pkl"
+  MASK_ON: True
+  RESNETS:
+    DEPTH: 50

configs/COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_1x_4gpu_transfiner.yaml

@@ -0,0 +1,6 @@
+_BASE_: "../Base-RCNN-FPN-4gpu.yaml"
+MODEL:
+  WEIGHTS: "./init_weights/model_final_a54504.pkl"
+  MASK_ON: True
+  RESNETS:
+    DEPTH: 50

configs/COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_1x_giou.yaml

@@ -0,0 +1,12 @@
+_BASE_: "../Base-RCNN-FPN.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-50.pkl"
+  MASK_ON: True
+  RESNETS:
+    DEPTH: 50
+  RPN:
+    BBOX_REG_LOSS_TYPE: "giou"
+    BBOX_REG_LOSS_WEIGHT: 2.0
+  ROI_BOX_HEAD:
+    BBOX_REG_LOSS_TYPE: "giou"
+    BBOX_REG_LOSS_WEIGHT: 10.0

configs/COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml

@@ -0,0 +1,9 @@
+_BASE_: "../Base-RCNN-FPN.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-50.pkl"
+  MASK_ON: True
+  RESNETS:
+    DEPTH: 50
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x_4gpu_transfiner.yaml

@@ -0,0 +1,9 @@
+_BASE_: "../Base-RCNN-FPN-4gpu.yaml"
+MODEL:
+  WEIGHTS: "./init_weights/model_final_f10217.pkl"
+  MASK_ON: True
+  RESNETS:
+    DEPTH: 50
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x_4gpu_transfiner_deform.yaml

@@ -0,0 +1,11 @@
+_BASE_: "../Base-RCNN-FPN-4gpu.yaml"
+MODEL:
+  WEIGHTS: "./init_weights/model_final_f10217.pkl"
+  MASK_ON: True
+  RESNETS:
+    DEPTH: 50
+    DEFORM_ON_PER_STAGE: [False, True, True, True] # on Res3,Res4,Res5
+    DEFORM_MODULATED: False
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x_4gpu_transfiner_lvis.yaml

@@ -0,0 +1,12 @@
+_BASE_: "../Base-RCNN-FPN-4gpu.yaml"
+MODEL:
+  WEIGHTS: "./init_weights/model_final_f10217.pkl"
+  MASK_ON: True
+  RESNETS:
+    DEPTH: 50
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000
+DATASETS:
+  TEST: ("lvis_v0.5_val_cocofied",)
+

configs/COCO-InstanceSegmentation/mask_rcnn_X_101_32x8d_FPN_3x.yaml

@@ -0,0 +1,13 @@
+_BASE_: "../Base-RCNN-FPN.yaml"
+MODEL:
+  MASK_ON: True
+  WEIGHTS: "detectron2://ImageNetPretrained/FAIR/X-101-32x8d.pkl"
+  PIXEL_STD: [57.375, 57.120, 58.395]
+  RESNETS:
+    STRIDE_IN_1X1: False  # this is a C2 model
+    NUM_GROUPS: 32
+    WIDTH_PER_GROUP: 8
+    DEPTH: 101
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-InstanceSegmentation/mask_rcnn_X_101_32x8d_FPN_3x_transfiner.yaml

@@ -0,0 +1,13 @@
+_BASE_: "../Base-RCNN-FPN-4gpu.yaml"
+MODEL:
+  MASK_ON: True
+  WEIGHTS: "./init_weights/model_final_x101.pkl"
+  PIXEL_STD: [57.375, 57.120, 58.395]
+  RESNETS:
+    STRIDE_IN_1X1: False  # this is a C2 model
+    NUM_GROUPS: 32
+    WIDTH_PER_GROUP: 8
+    DEPTH: 101
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000

configs/COCO-InstanceSegmentation/mask_rcnn_regnetx_4gf_dds_fpn_1x.py

@@ -0,0 +1,34 @@
+from ..common.optim import SGD as optimizer
+from ..common.coco_schedule import lr_multiplier_1x as lr_multiplier
+from ..common.data.coco import dataloader
+from ..common.models.mask_rcnn_fpn import model
+from ..common.train import train
+
+from detectron2.config import LazyCall as L
+from detectron2.modeling.backbone import RegNet
+from detectron2.modeling.backbone.regnet import SimpleStem, ResBottleneckBlock
+
+
+# Replace default ResNet with RegNetX-4GF from the DDS paper. Config source:
+# https://github.com/facebookresearch/pycls/blob/2c152a6e5d913e898cca4f0a758f41e6b976714d/configs/dds_baselines/regnetx/RegNetX-4.0GF_dds_8gpu.yaml#L4-L9  # noqa
+model.backbone.bottom_up = L(RegNet)(
+    stem_class=SimpleStem,
+    stem_width=32,
+    block_class=ResBottleneckBlock,
+    depth=23,
+    w_a=38.65,
+    w_0=96,
+    w_m=2.43,
+    group_width=40,
+    freeze_at=2,
+    norm="FrozenBN",
+    out_features=["s1", "s2", "s3", "s4"],
+)
+model.pixel_std = [57.375, 57.120, 58.395]
+
+optimizer.weight_decay = 5e-5
+train.init_checkpoint = (
+    "https://dl.fbaipublicfiles.com/pycls/dds_baselines/160906383/RegNetX-4.0GF_dds_8gpu.pyth"
+)
+# RegNets benefit from enabling cudnn benchmark mode
+train.cudnn_benchmark = True

configs/COCO-InstanceSegmentation/mask_rcnn_regnety_4gf_dds_fpn_1x.py

@@ -0,0 +1,35 @@
+from ..common.optim import SGD as optimizer
+from ..common.coco_schedule import lr_multiplier_1x as lr_multiplier
+from ..common.data.coco import dataloader
+from ..common.models.mask_rcnn_fpn import model
+from ..common.train import train
+
+from detectron2.config import LazyCall as L
+from detectron2.modeling.backbone import RegNet
+from detectron2.modeling.backbone.regnet import SimpleStem, ResBottleneckBlock
+
+
+# Replace default ResNet with RegNetY-4GF from the DDS paper. Config source:
+# https://github.com/facebookresearch/pycls/blob/2c152a6e5d913e898cca4f0a758f41e6b976714d/configs/dds_baselines/regnety/RegNetY-4.0GF_dds_8gpu.yaml#L4-L10  # noqa
+model.backbone.bottom_up = L(RegNet)(
+    stem_class=SimpleStem,
+    stem_width=32,
+    block_class=ResBottleneckBlock,
+    depth=22,
+    w_a=31.41,
+    w_0=96,
+    w_m=2.24,
+    group_width=64,
+    se_ratio=0.25,
+    freeze_at=2,
+    norm="FrozenBN",
+    out_features=["s1", "s2", "s3", "s4"],
+)
+model.pixel_std = [57.375, 57.120, 58.395]
+
+optimizer.weight_decay = 5e-5
+train.init_checkpoint = (
+    "https://dl.fbaipublicfiles.com/pycls/dds_baselines/160906838/RegNetY-4.0GF_dds_8gpu.pyth"
+)
+# RegNets benefit from enabling cudnn benchmark mode
+train.cudnn_benchmark = True

configs/COCO-Keypoints/Base-Keypoint-RCNN-FPN.yaml

@@ -0,0 +1,15 @@
+_BASE_: "../Base-RCNN-FPN.yaml"
+MODEL:
+  KEYPOINT_ON: True
+  ROI_HEADS:
+    NUM_CLASSES: 1
+  ROI_BOX_HEAD:
+    SMOOTH_L1_BETA: 0.5  # Keypoint AP degrades (though box AP improves) when using plain L1 loss
+  RPN:
+    # Detectron1 uses 2000 proposals per-batch, but this option is per-image in detectron2.
+    # 1000 proposals per-image is found to hurt box AP.
+    # Therefore we increase it to 1500 per-image.
+    POST_NMS_TOPK_TRAIN: 1500
+DATASETS:
+  TRAIN: ("keypoints_coco_2017_train",)
+  TEST: ("keypoints_coco_2017_val",)

configs/COCO-Keypoints/keypoint_rcnn_R_101_FPN_3x.yaml

@@ -0,0 +1,8 @@
+_BASE_: "Base-Keypoint-RCNN-FPN.yaml"
+MODEL:
+  WEIGHTS: "detectron2://ImageNetPretrained/MSRA/R-101.pkl"
+  RESNETS:
+    DEPTH: 101
+SOLVER:
+  STEPS: (210000, 250000)
+  MAX_ITER: 270000