first commit

rtdetrv2_pytorch/configs/rtdetr/include/dataloader.yml

@@ -0,0 +1,31 @@
+
+train_dataloader: 
+  dataset: 
+    return_masks: False
+    transforms:
+      ops:
+        - {type: RandomPhotometricDistort, p: 0.5}
+        - {type: RandomZoomOut, fill: 0}
+        - {type: RandomIoUCrop, p: 0.8}
+        - {type: SanitizeBoundingBoxes, min_size: 1}
+        - {type: RandomHorizontalFlip}
+        - {type: Resize, size: [640, 640], }
+        - {type: SanitizeBoundingBoxes, min_size: 1}
+        - {type: ConvertPILImage, dtype: 'float32', scale: True}   
+        - {type: ConvertBoxes, fmt: 'cxcywh', normalize: True}  
+  collate_fn:
+    type: BatchImageCollateFunction
+    scales: [480, 512, 544, 576, 608, 640, 640, 640, 672, 704, 736, 768, 800]
+  shuffle: True
+  num_workers: 4
+  total_batch_size: 16
+
+val_dataloader:
+  dataset: 
+    transforms:
+      ops: 
+        - {type: Resize, size: [640, 640]}
+        - {type: ConvertPILImage, dtype: 'float32', scale: True}   
+  shuffle: False
+  total_batch_size: 16
+  num_workers: 8

rtdetrv2_pytorch/configs/rtdetr/include/optimizer.yml

@@ -0,0 +1,40 @@
+
+use_ema: True 
+ema:
+  type: ModelEMA
+  decay: 0.9999
+  warmups: 2000
+
+
+epoches: 72
+clip_max_norm: 0.1
+
+
+optimizer:
+  type: AdamW
+  params: 
+    - 
+      params: '^(?=.*backbone)(?!.*(?:norm|bn)).*$'
+      lr: 0.00001
+    -
+      params: '^(?=.*backbone)(?=.*(?:norm|bn)).*$'
+      weight_decay: 0.
+      lr: 0.00001
+    - 
+      params: '^(?=.*(?:encoder|decoder))(?=.*(?:norm|bn|bias)).*$'
+      weight_decay: 0.
+
+  lr: 0.0001
+  betas: [0.9, 0.999]
+  weight_decay: 0.0001
+
+
+lr_scheduler:
+  type: MultiStepLR
+  milestones: [1000]
+  gamma: 0.1
+
+
+lr_warmup_scheduler:
+  type: LinearWarmup
+  warmup_duration: 2000

rtdetrv2_pytorch/configs/rtdetr/include/rtdetr_r50vd.yml

@@ -0,0 +1,79 @@
+task: detection
+
+model: RTDETR
+criterion: RTDETRCriterion
+postprocessor: RTDETRPostProcessor
+
+
+use_focal_loss: True
+eval_spatial_size: [640, 640] # h w
+
+
+RTDETR: 
+  backbone: PResNet
+  encoder: HybridEncoder
+  decoder: RTDETRTransformer
+  
+
+PResNet:
+  depth: 50
+  variant: d
+  freeze_at: 0
+  return_idx: [1, 2, 3]
+  num_stages: 4
+  freeze_norm: True
+  pretrained: True 
+
+
+HybridEncoder:
+  in_channels: [512, 1024, 2048]
+  feat_strides: [8, 16, 32]
+
+  # intra
+  hidden_dim: 256
+  use_encoder_idx: [2]
+  num_encoder_layers: 1
+  nhead: 8
+  dim_feedforward: 1024
+  dropout: 0.
+  enc_act: 'gelu'
+  
+  # cross
+  expansion: 1.0
+  depth_mult: 1
+  act: 'silu'
+
+  version: v1
+
+RTDETRTransformer:
+  feat_channels: [256, 256, 256]
+  feat_strides: [8, 16, 32]
+  hidden_dim: 256
+  num_levels: 3
+
+  num_layers: 6
+  num_queries: 300
+
+  num_denoising: 100
+  label_noise_ratio: 0.5
+  box_noise_scale: 1.0 # 1.0 0.4
+
+  eval_idx: -1
+
+
+RTDETRPostProcessor:
+  num_top_queries: 300
+
+
+RTDETRCriterion:
+  weight_dict: {loss_vfl: 1, loss_bbox: 5, loss_giou: 2,}
+  losses: ['vfl', 'boxes', ]
+  alpha: 0.75
+  gamma: 2.0
+
+  matcher:
+    type: HungarianMatcher
+    weight_dict: {cost_class: 2, cost_bbox: 5, cost_giou: 2}
+    alpha: 0.25
+    gamma: 2.0
+