Add model

config.json

@@ -1,19 +1,12 @@
 {
   "act_decoder": "relu",
   "act_encoder": "silu",
-  "act_presnet": "relu",
   "architectures": [
-    "RTDetrForObjectDetection"
+    "RTDetrModel"
   ],
   "aux_loss": true,
-  "block_nums": [
-    3,
-    4,
-    6,
-    3
-  ],
+  "backbone": "resnet50d",
   "box_noise_scale": 1.0,
-  "depth": 50,
   "depth_mult": 1.0,
   "dim_feedforward": 1024,
   "dropout": 0.0,
@@ -40,7 +33,7 @@
   "focal_loss_alpha": 0.75,
   "focal_loss_gamma": 2.0,
   "freeze_at": 0,
-  "freeze_norm": true,
+  "freeze_batch_norm_2d": true,
   "hidden_dim": 256,
   "in_channels": [
     512,
@@ -48,14 +41,13 @@
     2048
   ],
   "initializer_range": 0.02,
-  "is_encoder_decoder": true,
   "label_noise_ratio": 0.5,
   "learnt_init_query": false,
   "matcher_alpha": 0.25,
-  "matcher_bbox_cost": 5,
-  "matcher_class_cost": 2,
+  "matcher_bbox_cost": 5.0,
+  "matcher_class_cost": 2.0,
   "matcher_gamma": 2.0,
-  "matcher_giou_cost": 2,
+  "matcher_giou_cost": 2.0,
   "model_type": "rt_detr",
   "num_classes": 80,
   "num_decoder_layers": 6,
@@ -65,10 +57,13 @@
   "num_head": 8,
   "num_levels": 3,
   "num_queries": 300,
-  "num_stages": 4,
+  "out_indices": [
+    2,
+    3,
+    4
+  ],
   "pe_temperature": 10000,
   "position_embed_type": "sine",
-  "pretrained": true,
   "return_idx": [
     1,
     2,
@@ -80,8 +75,7 @@
     2
   ],
   "use_focal_loss": true,
-  "variant": "d",
-  "weight_loss_bbox": 5,
-  "weight_loss_giou": 2,
-  "weight_loss_vfl": 1
+  "weight_loss_bbox": 5.0,
+  "weight_loss_giou": 2.0,
+  "weight_loss_vfl": 1.0
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bb36532d579033ddb231c9d884a86477ce1574005c35d2976a48de3ddb36641b
-size 172334605
+oid sha256:9cb35e2d81385cfdfa7f25a6d393c2da0cc67f44d158829943f3dc8a128ce2cf
+size 172345001