Training in progress, epoch 1

config.json

@@ -2,11 +2,11 @@
   "activation_dropout": 0.0,
   "activation_function": "relu",
   "architectures": [
-    "ConditionalDetrForObjectDetection"
+    "TableTransformerForObjectDetection"
   ],
   "attention_dropout": 0.0,
   "auxiliary_loss": false,
-  "backbone": "resnet50",
+  "backbone": "resnet18",
   "backbone_config": null,
   "backbone_kwargs": {
     "in_chans": 3,
@@ -19,8 +19,8 @@
   },
   "bbox_cost": 5,
   "bbox_loss_coefficient": 5,
-  "class_cost": 2,
-  "cls_loss_coefficient": 2,
+  "ce_loss_coefficient": 1,
+  "class_cost": 1,
   "d_model": 256,
   "decoder_attention_heads": 8,
   "decoder_ffn_dim": 2048,
@@ -33,32 +33,34 @@
   "encoder_ffn_dim": 2048,
   "encoder_layerdrop": 0.0,
   "encoder_layers": 6,
-  "focal_alpha": 0.25,
+  "eos_coefficient": 0.4,
   "giou_cost": 2,
   "giou_loss_coefficient": 2,
   "id2label": {
-    "0": "Coverall",
-    "1": "Face_Shield",
-    "2": "Gloves",
-    "3": "Goggles",
-    "4": "Mask"
+    "0": "table",
+    "1": "table column",
+    "2": "table row",
+    "3": "table column header",
+    "4": "table projected row header",
+    "5": "table spanning cell"
   },
   "init_std": 0.02,
   "init_xavier_std": 1.0,
   "is_encoder_decoder": true,
   "label2id": {
-    "Coverall": 0,
-    "Face_Shield": 1,
-    "Gloves": 2,
-    "Goggles": 3,
-    "Mask": 4
+    "table": 0,
+    "table column": 1,
+    "table column header": 3,
+    "table projected row header": 4,
+    "table row": 2,
+    "table spanning cell": 5
   },
   "mask_loss_coefficient": 1,
   "max_position_embeddings": 1024,
-  "model_type": "conditional_detr",
+  "model_type": "table-transformer",
   "num_channels": 3,
   "num_hidden_layers": 6,
-  "num_queries": 300,
+  "num_queries": 125,
   "position_embedding_type": "sine",
   "scale_embedding": false,
   "torch_dtype": "float32",

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8b37b819df34d0e1e7f15f605fe71a12be79d1094d7541bf6696c052fe22d713
-size 174079796
+oid sha256:515b8b3c8999f7118544191d90ded04526e4fd1a9b629e65b447ee8d81b3bcc4
+size 115433860

preprocessor_config.json

@@ -10,7 +10,7 @@
     0.456,
     0.406
   ],
-  "image_processor_type": "ConditionalDetrImageProcessor",
+  "image_processor_type": "DetrImageProcessor",
   "image_std": [
     0.229,
     0.224,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:33543a85381009b17d2c5fa7c9ea44253a42c19653ac5c509acd1d9a9e13e43f
+oid sha256:744302a7d7102b36784fd0e51062b8ec0c932dc9f1710cb43d72843d0946fffd
 size 5304