Training in progress, epoch 1

config.json

@@ -1,61 +1,40 @@
 {
-  "_name_or_path": "facebook/detr-resnet-50",
-  "activation_dropout": 0.0,
-  "activation_function": "relu",
+  "_name_or_path": "hustvl/yolos-tiny",
   "architectures": [
-    "DetrForObjectDetection"
+    "YolosForObjectDetection"
   ],
-  "attention_dropout": 0.0,
+  "attention_probs_dropout_prob": 0.0,
   "auxiliary_loss": false,
-  "backbone": "resnet50",
-  "backbone_config": null,
-  "backbone_kwargs": {
-    "in_chans": 3,
-    "out_indices": [
-      1,
-      2,
-      3,
-      4
-    ]
-  },
   "bbox_cost": 5,
   "bbox_loss_coefficient": 5,
   "class_cost": 1,
-  "classifier_dropout": 0.0,
-  "d_model": 256,
-  "decoder_attention_heads": 8,
-  "decoder_ffn_dim": 2048,
-  "decoder_layerdrop": 0.0,
-  "decoder_layers": 6,
-  "dice_loss_coefficient": 1,
-  "dilation": false,
-  "dropout": 0.1,
-  "encoder_attention_heads": 8,
-  "encoder_ffn_dim": 2048,
-  "encoder_layerdrop": 0.0,
-  "encoder_layers": 6,
   "eos_coefficient": 0.1,
   "giou_cost": 2,
   "giou_loss_coefficient": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 192,
   "id2label": {
     "1": "kangaroo"
   },
-  "init_std": 0.02,
-  "init_xavier_std": 1.0,
-  "is_encoder_decoder": true,
+  "image_size": [
+    800,
+    1333
+  ],
+  "initializer_range": 0.02,
+  "intermediate_size": 768,
   "label2id": {
     "kangaroo": 1
   },
-  "mask_loss_coefficient": 1,
-  "max_position_embeddings": 1024,
-  "model_type": "detr",
+  "layer_norm_eps": 1e-12,
+  "model_type": "yolos",
+  "num_attention_heads": 3,
   "num_channels": 3,
-  "num_hidden_layers": 6,
-  "num_queries": 100,
-  "position_embedding_type": "sine",
-  "scale_embedding": false,
+  "num_detection_tokens": 100,
+  "num_hidden_layers": 12,
+  "patch_size": 16,
+  "qkv_bias": true,
   "torch_dtype": "float32",
   "transformers_version": "4.48.3",
-  "use_pretrained_backbone": true,
-  "use_timm_backbone": true
+  "use_mid_position_embeddings": false
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b50d8f05a432e324fcfb85ac4736e2934b4771bb3214632ed8dcd039d0bd1b0f
-size 166494824
+oid sha256:f8c30109ba7287fc098cd79b6b989934bd40a8598f7c87dcd2c4c59442baf7ad
+size 25909400

preprocessor_config.json

@@ -10,7 +10,7 @@
     0.456,
     0.406
   ],
-  "image_processor_type": "DetrImageProcessor",
+  "image_processor_type": "YolosImageProcessor",
   "image_std": [
     0.229,
     0.224,
@@ -21,6 +21,6 @@
   "rescale_factor": 0.00392156862745098,
   "size": {
     "longest_edge": 1333,
-    "shortest_edge": 800
+    "shortest_edge": 512
   }
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0322d204d0eea8aebac847506b770e36217e5836e413806dfb3dca51d178e3cd
+oid sha256:300f04ae519018652b6ecdcee53dc7b88d11a86dc4573b7276915550237d8fda
 size 5304