Training in progress, step 200

README.md

@@ -1,6 +1,6 @@
 ---
 license: apache-2.0
-base_model: SenseTime/deformable-detr
+base_model: facebook/detr-resnet-50
 tags:
 - generated_from_trainer
 model-index:
@@ -13,7 +13,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # detr
 
-This model is a fine-tuned version of [SenseTime/deformable-detr](https://huggingface.co/SenseTime/deformable-detr) on the None dataset.
+This model is a fine-tuned version of [facebook/detr-resnet-50](https://huggingface.co/facebook/detr-resnet-50) on the None dataset.
 
 ## Model description
 
@@ -38,7 +38,7 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 15
+- num_epochs: 20
 - mixed_precision_training: Native AMP
 
 ### Training results
@@ -47,7 +47,7 @@ The following hyperparameters were used during training:
 
 ### Framework versions
 
-- Transformers 4.38.2
+- Transformers 4.38.1
 - Pytorch 2.2.1+cu121
-- Datasets 2.18.0
+- Datasets 2.17.1
 - Tokenizers 0.15.2

config.json

@@ -1,9 +1,9 @@
 {
-  "_name_or_path": "SenseTime/deformable-detr",
+  "_name_or_path": "facebook/detr-resnet-50",
   "activation_dropout": 0.0,
   "activation_function": "relu",
   "architectures": [
-    "DeformableDetrForObjectDetection"
+    "DetrForObjectDetection"
   ],
   "attention_dropout": 0.0,
   "auxiliary_loss": false,
@@ -13,23 +13,20 @@
   "bbox_cost": 5,
   "bbox_loss_coefficient": 5,
   "class_cost": 1,
+  "classifier_dropout": 0.0,
   "d_model": 256,
   "decoder_attention_heads": 8,
-  "decoder_ffn_dim": 1024,
+  "decoder_ffn_dim": 2048,
   "decoder_layerdrop": 0.0,
   "decoder_layers": 6,
-  "decoder_n_points": 4,
   "dice_loss_coefficient": 1,
   "dilation": false,
-  "disable_custom_kernels": false,
   "dropout": 0.1,
   "encoder_attention_heads": 8,
-  "encoder_ffn_dim": 1024,
+  "encoder_ffn_dim": 2048,
   "encoder_layerdrop": 0.0,
   "encoder_layers": 6,
-  "encoder_n_points": 4,
   "eos_coefficient": 0.1,
-  "focal_alpha": 0.25,
   "giou_cost": 2,
   "giou_loss_coefficient": 2,
   "id2label": {
@@ -43,16 +40,14 @@
   },
   "mask_loss_coefficient": 1,
   "max_position_embeddings": 1024,
-  "model_type": "deformable_detr",
+  "model_type": "detr",
   "num_channels": 3,
-  "num_feature_levels": 4,
-  "num_queries": 300,
+  "num_hidden_layers": 6,
+  "num_queries": 150,
   "position_embedding_type": "sine",
+  "scale_embedding": false,
   "torch_dtype": "float32",
-  "transformers_version": "4.38.2",
-  "two_stage": false,
-  "two_stage_num_proposals": 300,
+  "transformers_version": "4.38.1",
   "use_pretrained_backbone": true,
-  "use_timm_backbone": true,
-  "with_box_refine": false
+  "use_timm_backbone": true
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:511243e3fc69b8a971865ffd71b042f08fcc0bcfc65714942640983896e8724b
-size 160676460
+oid sha256:b2c3f006aa8d99951bc725eeac437f5c25d93a93aab07ad4b43f5939cd1655c9
+size 166546024

preprocessor_config.json

@@ -10,7 +10,7 @@
     0.456,
     0.406
   ],
-  "image_processor_type": "DeformableDetrImageProcessor",
+  "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:b93bbc9d246d3cab23e8ec1e4fd3b7d7076041f54d586188dd75f21fdb9289bf
+oid sha256:e6eb9b88d2bec2c45b07ccd100f7ffc91323aefe2611c9395f2361133829b4d9
 size 4856