Model save

README.md

@@ -0,0 +1,105 @@
+---
+library_name: transformers
+license: apache-2.0
+base_model: PekingU/rtdetr_r50vd
+tags:
+- generated_from_trainer
+model-index:
+- name: suas-2025-rtdetr-finetuned
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# suas-2025-rtdetr-finetuned
+
+This model is a fine-tuned version of [PekingU/rtdetr_r50vd](https://huggingface.co/PekingU/rtdetr_r50vd) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 3.8868
+- Map: 0.8924
+- Map 50: 0.9568
+- Map 75: 0.9554
+- Map Small: 0.8115
+- Map Medium: 0.9048
+- Map Large: 0.9511
+- Mar 1: 0.8296
+- Mar 10: 0.938
+- Mar 100: 0.9407
+- Mar Small: 0.8752
+- Mar Medium: 0.9535
+- Mar Large: 0.9911
+- Map Baseball-bat: 0.8962
+- Mar 100 Baseball-bat: 0.9382
+- Map Basketball: 0.8344
+- Mar 100 Basketball: 0.9077
+- Map Car: -1.0
+- Mar 100 Car: -1.0
+- Map Football: 0.814
+- Mar 100 Football: 0.8709
+- Map Human: 0.9172
+- Mar 100 Human: 0.9686
+- Map Luggage: 0.8623
+- Mar 100 Luggage: 0.9241
+- Map Mattress: 0.9809
+- Mar 100 Mattress: 0.9971
+- Map Motorcycle: 0.9393
+- Mar 100 Motorcycle: 0.9751
+- Map Skis: 0.8496
+- Mar 100 Skis: 0.9692
+- Map Snowboard: 0.9857
+- Mar 100 Snowboard: 0.9947
+- Map Soccer-ball: 0.8382
+- Mar 100 Soccer-ball: 0.8755
+- Map Stop-sign: 0.9735
+- Mar 100 Stop-sign: 0.9957
+- Map Tennis-racket: 0.9007
+- Mar 100 Tennis-racket: 0.9245
+- Map Umbrella: 0.8994
+- Mar 100 Umbrella: 0.9775
+- Map Volleyball: 0.8024
+- Mar 100 Volleyball: 0.8507
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 3e-05
+- train_batch_size: 8
+- eval_batch_size: 8
+- seed: 1337
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- num_epochs: 5.0
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Map    | Map 50 | Map 75 | Map Small | Map Medium | Map Large | Mar 1  | Mar 10 | Mar 100 | Mar Small | Mar Medium | Mar Large | Map Baseball-bat | Mar 100 Baseball-bat | Map Basketball | Mar 100 Basketball | Map Car | Mar 100 Car | Map Football | Mar 100 Football | Map Human | Mar 100 Human | Map Luggage | Mar 100 Luggage | Map Mattress | Mar 100 Mattress | Map Motorcycle | Mar 100 Motorcycle | Map Skis | Mar 100 Skis | Map Snowboard | Mar 100 Snowboard | Map Soccer-ball | Mar 100 Soccer-ball | Map Stop-sign | Mar 100 Stop-sign | Map Tennis-racket | Mar 100 Tennis-racket | Map Umbrella | Mar 100 Umbrella | Map Volleyball | Mar 100 Volleyball |
+|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:----------:|:---------:|:------:|:------:|:-------:|:---------:|:----------:|:---------:|:----------------:|:--------------------:|:--------------:|:------------------:|:-------:|:-----------:|:------------:|:----------------:|:---------:|:-------------:|:-----------:|:---------------:|:------------:|:----------------:|:--------------:|:------------------:|:--------:|:------------:|:-------------:|:-----------------:|:---------------:|:-------------------:|:-------------:|:-----------------:|:-----------------:|:---------------------:|:------------:|:----------------:|:--------------:|:------------------:|
+| 12.0474       | 1.0   | 438  | 5.5119          | 0.8067 | 0.8868 | 0.8817 | 0.7597    | 0.7886     | 0.8834    | 0.7817 | 0.9168 | 0.924   | 0.8348    | 0.9511     | 0.9848    | 0.7992           | 0.9114               | 0.819          | 0.901              | -1.0    | -1.0        | 0.7627       | 0.8381           | 0.6589    | 0.9657        | 0.8118      | 0.9173          | 0.9643       | 0.9961           | 0.8513         | 0.9658             | 0.5388   | 0.9767       | 0.9801        | 0.9925            | 0.7934          | 0.85                | 0.9652        | 0.988             | 0.8307            | 0.9057                | 0.8198       | 0.9603           | 0.6983         | 0.7671             |
+| 6.1205        | 2.0   | 876  | 5.0519          | 0.8186 | 0.8885 | 0.8836 | 0.746     | 0.7754     | 0.902     | 0.7791 | 0.9033 | 0.9094  | 0.8072    | 0.9373     | 0.9872    | 0.8735           | 0.9189               | 0.8175         | 0.9172             | -1.0    | -1.0        | 0.7607       | 0.8174           | 0.798     | 0.9653        | 0.8087      | 0.8941          | 0.9531       | 0.9942           | 0.7115         | 0.9665             | 0.7689   | 0.9508       | 0.9767        | 0.9957            | 0.7882          | 0.846               | 0.92          | 0.9799            | 0.8979            | 0.9208                | 0.7901       | 0.9309           | 0.5955         | 0.6338             |
+| 5.3217        | 3.0   | 1314 | 4.0739          | 0.8811 | 0.9513 | 0.9499 | 0.7797    | 0.8841     | 0.9549    | 0.8259 | 0.9337 | 0.9374  | 0.8626    | 0.9518     | 0.9937    | 0.8664           | 0.9269               | 0.8426         | 0.9086             | -1.0    | -1.0        | 0.7995       | 0.8595           | 0.891     | 0.9688        | 0.8574      | 0.9292          | 0.9881       | 0.9998           | 0.9232         | 0.9719             | 0.8199   | 0.9958       | 0.9816        | 0.9932            | 0.82            | 0.862               | 0.979         | 0.994             | 0.8971            | 0.9198                | 0.8925       | 0.9689           | 0.777          | 0.8258             |
+| 4.7931        | 4.0   | 1752 | 3.8984          | 0.8943 | 0.9605 | 0.9589 | 0.8046    | 0.9139     | 0.9535    | 0.8289 | 0.9363 | 0.9404  | 0.8726    | 0.9571     | 0.9849    | 0.8889           | 0.9341               | 0.8321         | 0.9053             | -1.0    | -1.0        | 0.809        | 0.8622           | 0.9072    | 0.9685        | 0.8876      | 0.9389          | 0.9796       | 0.9993           | 0.9379         | 0.9757             | 0.8801   | 0.96         | 0.9855        | 0.9972            | 0.8379          | 0.8745              | 0.9811        | 0.994             | 0.9046            | 0.9302                | 0.8907       | 0.9722           | 0.7979         | 0.8531             |
+| 4.634         | 5.0   | 2190 | 3.8868          | 0.8924 | 0.9568 | 0.9554 | 0.8115    | 0.9048     | 0.9511    | 0.8296 | 0.938  | 0.9407  | 0.8752    | 0.9535     | 0.9911    | 0.8962           | 0.9382               | 0.8344         | 0.9077             | -1.0    | -1.0        | 0.814        | 0.8709           | 0.9172    | 0.9686        | 0.8623      | 0.9241          | 0.9809       | 0.9971           | 0.9393         | 0.9751             | 0.8496   | 0.9692       | 0.9857        | 0.9947            | 0.8382          | 0.8755              | 0.9735        | 0.9957            | 0.9007            | 0.9245                | 0.8994       | 0.9775           | 0.8024         | 0.8507             |
+
+
+### Framework versions
+
+- Transformers 4.47.0
+- Pytorch 2.5.1+cu124
+- Datasets 3.1.0
+- Tokenizers 0.21.0

config.json

@@ -0,0 +1,125 @@
+{
+  "_name_or_path": "PekingU/rtdetr_r50vd",
+  "activation_dropout": 0.0,
+  "activation_function": "silu",
+  "anchor_image_size": null,
+  "architectures": [
+    "RTDetrForObjectDetection"
+  ],
+  "attention_dropout": 0.0,
+  "auxiliary_loss": true,
+  "backbone": null,
+  "backbone_config": {
+    "model_type": "rt_detr_resnet",
+    "out_features": [
+      "stage2",
+      "stage3",
+      "stage4"
+    ],
+    "out_indices": [
+      2,
+      3,
+      4
+    ]
+  },
+  "backbone_kwargs": null,
+  "batch_norm_eps": 1e-05,
+  "box_noise_scale": 1.0,
+  "d_model": 256,
+  "decoder_activation_function": "relu",
+  "decoder_attention_heads": 8,
+  "decoder_ffn_dim": 1024,
+  "decoder_in_channels": [
+    256,
+    256,
+    256
+  ],
+  "decoder_layers": 6,
+  "decoder_n_points": 4,
+  "disable_custom_kernels": true,
+  "dropout": 0.0,
+  "encode_proj_layers": [
+    2
+  ],
+  "encoder_activation_function": "gelu",
+  "encoder_attention_heads": 8,
+  "encoder_ffn_dim": 1024,
+  "encoder_hidden_dim": 256,
+  "encoder_in_channels": [
+    512,
+    1024,
+    2048
+  ],
+  "encoder_layers": 1,
+  "eos_coefficient": 0.0001,
+  "eval_size": null,
+  "feat_strides": [
+    8,
+    16,
+    32
+  ],
+  "focal_loss_alpha": 0.75,
+  "focal_loss_gamma": 2.0,
+  "freeze_backbone_batch_norms": true,
+  "hidden_expansion": 1.0,
+  "id2label": {
+    "0": "baseball-bat",
+    "1": "basketball",
+    "2": "car",
+    "3": "football",
+    "4": "human",
+    "5": "luggage",
+    "6": "mattress",
+    "7": "motorcycle",
+    "8": "skis",
+    "9": "snowboard",
+    "10": "soccer-ball",
+    "11": "stop-sign",
+    "12": "tennis-racket",
+    "13": "umbrella",
+    "14": "volleyball"
+  },
+  "initializer_bias_prior_prob": null,
+  "initializer_range": 0.01,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "baseball-bat": 0,
+    "basketball": 1,
+    "car": 2,
+    "football": 3,
+    "human": 4,
+    "luggage": 5,
+    "mattress": 6,
+    "motorcycle": 7,
+    "skis": 8,
+    "snowboard": 9,
+    "soccer-ball": 10,
+    "stop-sign": 11,
+    "tennis-racket": 12,
+    "umbrella": 13,
+    "volleyball": 14
+  },
+  "label_noise_ratio": 0.5,
+  "layer_norm_eps": 1e-05,
+  "learn_initial_query": false,
+  "matcher_alpha": 0.25,
+  "matcher_bbox_cost": 5.0,
+  "matcher_class_cost": 2.0,
+  "matcher_gamma": 2.0,
+  "matcher_giou_cost": 2.0,
+  "model_type": "rt_detr",
+  "normalize_before": false,
+  "num_denoising": 100,
+  "num_feature_levels": 3,
+  "num_queries": 300,
+  "positional_encoding_temperature": 10000,
+  "torch_dtype": "float32",
+  "transformers_version": "4.47.0",
+  "use_focal_loss": true,
+  "use_pretrained_backbone": false,
+  "use_timm_backbone": false,
+  "weight_loss_bbox": 5.0,
+  "weight_loss_giou": 2.0,
+  "weight_loss_vfl": 1.0,
+  "with_box_refine": true
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ba40fe530999755a8e5eba10c156d7b9a1edf2f63f74fafcac9c906781511f3
+size 171641556

preprocessor_config.json

@@ -0,0 +1,29 @@
+{
+  "do_convert_annotations": true,
+  "do_normalize": false,
+  "do_pad": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "format": "coco_detection",
+  "image_mean": [
+    0.485,
+    0.456,
+    0.406
+  ],
+  "image_processor_type": "RTDetrImageProcessor",
+  "image_std": [
+    0.229,
+    0.224,
+    0.225
+  ],
+  "pad_size": {
+    "height": 640,
+    "width": 640
+  },
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "max_height": 640,
+    "max_width": 640
+  }
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc16fe384f0a7520dc25ca672a63e3b0a5d6b5bfadf3a40f197a05d61b09a540
+size 5368