Layout model based on RT-DETR-v2

RT-DETRv2 trained on document datasets

model_artifacts/layout/config.json

@@ -3,13 +3,31 @@
   "activation_function": "silu",
   "anchor_image_size": null,
   "architectures": [
-    "RTDetrForObjectDetection"
+    "RTDetrV2ForObjectDetection"
   ],
   "attention_dropout": 0.0,
   "auxiliary_loss": true,
   "backbone": null,
   "backbone_config": {
+    "depths": [
+      3,
+      4,
+      6,
+      3
+    ],
+    "downsample_in_bottleneck": false,
+    "downsample_in_first_stage": false,
+    "embedding_size": 64,
+    "hidden_act": "relu",
+    "hidden_sizes": [
+      256,
+      512,
+      1024,
+      2048
+    ],
+    "layer_type": "bottleneck",
     "model_type": "rt_detr_resnet",
+    "num_channels": 3,
     "out_features": [
       "stage2",
       "stage3",
@@ -19,6 +37,13 @@
       2,
       3,
       4
+    ],
+    "stage_names": [
+      "stem",
+      "stage1",
+      "stage2",
+      "stage3",
+      "stage4"
     ]
   },
   "backbone_kwargs": null,
@@ -34,8 +59,10 @@
     256
   ],
   "decoder_layers": 6,
+  "decoder_method": "default",
+  "decoder_n_levels": 3,
   "decoder_n_points": 4,
-  "disable_custom_kernels": true,
+  "decoder_offset_scale": 0.5,
   "dropout": 0.0,
   "encode_proj_layers": [
     2
@@ -62,47 +89,45 @@
   "freeze_backbone_batch_norms": true,
   "hidden_expansion": 1.0,
   "id2label": {
-    "0": "background",
-    "1": "Caption",
-    "10": "Text",
-    "11": "Title",
-    "12": "Document Index",
-    "13": "Code",
-    "14": "Checkbox-Selected",
-    "15": "Checkbox-Unselected",
-    "16": "Form",
-    "17": "Key-Value Region",
-    "2": "Footnote",
-    "3": "Formula",
-    "4": "List-item",
-    "5": "Page-footer",
-    "6": "Page-header",
-    "7": "Picture",
-    "8": "Section-header",
-    "9": "Table"
+    "0": "Caption",
+    "1": "Footnote",
+    "2": "Formula",
+    "3": "List-item",
+    "4": "Page-footer",
+    "5": "Page-header",
+    "6": "Picture",
+    "7": "Section-header",
+    "8": "Table",
+    "9": "Text",
+    "10": "Title",
+    "11": "Document Index",
+    "12": "Code",
+    "13": "Checkbox-Selected",
+    "14": "Checkbox-Unselected",
+    "15": "Form",
+    "16": "Key-Value Region"
   },
   "initializer_bias_prior_prob": null,
   "initializer_range": 0.01,
   "is_encoder_decoder": true,
   "label2id": {
-    "Caption": "1",
-    "Checkbox-Selected": "14",
-    "Checkbox-Unselected": "15",
-    "Code": "13",
-    "Document Index": "12",
-    "Footnote": "2",
-    "Form": "16",
-    "Formula": "3",
-    "Key-Value Region": "17",
-    "List-item": "4",
-    "Page-footer": "5",
-    "Page-header": "6",
-    "Picture": "7",
-    "Section-header": "8",
-    "Table": "9",
-    "Text": "10",
-    "Title": "11",
-    "background": "0"
+    "Caption": 0,
+    "Checkbox-Selected": 13,
+    "Checkbox-Unselected": 14,
+    "Code": 12,
+    "Document Index": 11,
+    "Footnote": 1,
+    "Form": 15,
+    "Formula": 2,
+    "Key-Value Region": 16,
+    "List-item": 3,
+    "Page-footer": 4,
+    "Page-header": 5,
+    "Picture": 6,
+    "Section-header": 7,
+    "Table": 8,
+    "Text": 9,
+    "Title": 10
   },
   "label_noise_ratio": 0.5,
   "layer_norm_eps": 1e-05,
@@ -112,14 +137,14 @@
   "matcher_class_cost": 2.0,
   "matcher_gamma": 2.0,
   "matcher_giou_cost": 2.0,
-  "model_type": "rt_detr",
+  "model_type": "rt_detr_v2",
   "normalize_before": false,
   "num_denoising": 100,
   "num_feature_levels": 3,
   "num_queries": 300,
   "positional_encoding_temperature": 10000,
   "torch_dtype": "float32",
-  "transformers_version": "4.46.2",
+  "transformers_version": "4.51.3",
   "use_focal_loss": true,
   "use_pretrained_backbone": false,
   "use_timm_backbone": false,

model_artifacts/layout/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:31e60b4709571b613bc8736a9c982fb550d8d7a1809160a68a8282af60c8910b
-size 171666216
+oid sha256:b37f85ce9e044df9069471a4b0b764e1be93a44cb7adadfe37e496e928e676f5
+size 171658996