Training in progress, epoch 3, checkpoint

checkpoint-4908/config.json

@@ -0,0 +1,125 @@
+{
+  "activation_dropout": 0.0,
+  "activation_function": "silu",
+  "anchor_image_size": null,
+  "architectures": [
+    "RTDetrV2ForObjectDetection"
+  ],
+  "attention_dropout": 0.0,
+  "auxiliary_loss": true,
+  "backbone": null,
+  "backbone_config": {
+    "depths": [
+      2,
+      2,
+      2,
+      2
+    ],
+    "downsample_in_bottleneck": false,
+    "downsample_in_first_stage": false,
+    "dtype": "float32",
+    "embedding_size": 64,
+    "hidden_act": "relu",
+    "hidden_sizes": [
+      64,
+      128,
+      256,
+      512
+    ],
+    "layer_type": "basic",
+    "model_type": "rt_detr_resnet",
+    "num_channels": 3,
+    "out_features": [
+      "stage2",
+      "stage3",
+      "stage4"
+    ],
+    "out_indices": [
+      2,
+      3,
+      4
+    ],
+    "stage_names": [
+      "stem",
+      "stage1",
+      "stage2",
+      "stage3",
+      "stage4"
+    ]
+  },
+  "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": 3,
+  "decoder_method": "default",
+  "decoder_n_levels": 3,
+  "decoder_n_points": 4,
+  "decoder_offset_scale": 0.5,
+  "disable_custom_kernels": true,
+  "dropout": 0.0,
+  "dtype": "float32",
+  "encode_proj_layers": [
+    2
+  ],
+  "encoder_activation_function": "gelu",
+  "encoder_attention_heads": 8,
+  "encoder_ffn_dim": 1024,
+  "encoder_hidden_dim": 256,
+  "encoder_in_channels": [
+    128,
+    256,
+    512
+  ],
+  "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": 0.5,
+  "id2label": {
+    "0": "barcode"
+  },
+  "initializer_bias_prior_prob": null,
+  "initializer_range": 0.01,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "barcode": 0
+  },
+  "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_v2",
+  "normalize_before": false,
+  "num_denoising": 100,
+  "num_feature_levels": 3,
+  "num_queries": 300,
+  "positional_encoding_temperature": 10000,
+  "transformers_version": "4.57.3",
+  "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
+}

checkpoint-4908/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2beaa332e4bde5c11fc19ac5ec1b92b9d010de5f1f796a7511d725b9abb562e1
+size 80498888

checkpoint-4908/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:024da308cff0c6b2577bd6e1d6ae93b7d1b8dc780db784f2691063a3e8409b9e
+size 160838091

checkpoint-4908/preprocessor_config.json

@@ -0,0 +1,36 @@
+{
+  "crop_size": null,
+  "data_format": "channels_first",
+  "default_to_square": false,
+  "device": null,
+  "disable_grouping": null,
+  "do_center_crop": null,
+  "do_convert_annotations": true,
+  "do_convert_rgb": null,
+  "do_normalize": false,
+  "do_pad": false,
+  "do_rescale": true,
+  "do_resize": true,
+  "format": "coco_detection",
+  "image_mean": [
+    0.485,
+    0.456,
+    0.406
+  ],
+  "image_processor_type": "RTDetrImageProcessorFast",
+  "image_std": [
+    0.229,
+    0.224,
+    0.225
+  ],
+  "input_data_format": null,
+  "pad_size": null,
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "return_segmentation_masks": null,
+  "return_tensors": null,
+  "size": {
+    "height": 480,
+    "width": 480
+  }
+}

checkpoint-4908/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:adfb9e0863adec2893c24cf0e82da26541873541722130aac028ab5205a1b8f0
+size 14645

checkpoint-4908/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90eb2a8fd32b9b10b4fb2707118b20eb20ad02e9f615aa363eb72f18bc26e835
+size 1465

checkpoint-4908/trainer_state.json

@@ -0,0 +1,163 @@
+{
+  "best_global_step": 4908,
+  "best_metric": 0.6553,
+  "best_model_checkpoint": "output/checkpoint-4908",
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 4908,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.3056234718826406,
+      "grad_norm": 22.77391815185547,
+      "learning_rate": 4.255239520958084e-05,
+      "loss": 52.1418,
+      "step": 500
+    },
+    {
+      "epoch": 0.6112469437652812,
+      "grad_norm": 15.196788787841797,
+      "learning_rate": 2.383982035928144e-05,
+      "loss": 7.7913,
+      "step": 1000
+    },
+    {
+      "epoch": 0.9168704156479217,
+      "grad_norm": 13.545781135559082,
+      "learning_rate": 5.127245508982037e-06,
+      "loss": 7.43,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 5.054670810699463,
+      "eval_map": 0.623,
+      "eval_map_50": 0.7756,
+      "eval_map_75": 0.6841,
+      "eval_map_barcode": 0.623,
+      "eval_map_large": 0.6594,
+      "eval_map_medium": 0.6674,
+      "eval_map_small": 0.2648,
+      "eval_mar_1": 0.3542,
+      "eval_mar_10": 0.8117,
+      "eval_mar_100": 0.8521,
+      "eval_mar_100_barcode": 0.8521,
+      "eval_mar_large": 0.8749,
+      "eval_mar_medium": 0.8552,
+      "eval_mar_small": 0.5325,
+      "eval_runtime": 71.7672,
+      "eval_samples_per_second": 39.071,
+      "eval_steps_per_second": 4.891,
+      "step": 1636
+    },
+    {
+      "epoch": 1.2224938875305624,
+      "grad_norm": 14.290430068969727,
+      "learning_rate": 3.921954314720812e-05,
+      "loss": 7.2123,
+      "step": 2000
+    },
+    {
+      "epoch": 1.528117359413203,
+      "grad_norm": 15.990227699279785,
+      "learning_rate": 3.6046954314720814e-05,
+      "loss": 7.0995,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8337408312958434,
+      "grad_norm": 11.737943649291992,
+      "learning_rate": 3.2874365482233505e-05,
+      "loss": 6.9538,
+      "step": 3000
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 4.950796604156494,
+      "eval_map": 0.6297,
+      "eval_map_50": 0.7829,
+      "eval_map_75": 0.6878,
+      "eval_map_barcode": 0.6297,
+      "eval_map_large": 0.7109,
+      "eval_map_medium": 0.6449,
+      "eval_map_small": 0.261,
+      "eval_mar_1": 0.3612,
+      "eval_mar_10": 0.8238,
+      "eval_mar_100": 0.8604,
+      "eval_mar_100_barcode": 0.8604,
+      "eval_mar_large": 0.8794,
+      "eval_mar_medium": 0.8624,
+      "eval_mar_small": 0.5972,
+      "eval_runtime": 67.4554,
+      "eval_samples_per_second": 41.568,
+      "eval_steps_per_second": 5.203,
+      "step": 3272
+    },
+    {
+      "epoch": 2.139364303178484,
+      "grad_norm": 14.22537899017334,
+      "learning_rate": 2.9701776649746192e-05,
+      "loss": 6.8515,
+      "step": 3500
+    },
+    {
+      "epoch": 2.444987775061125,
+      "grad_norm": 19.026344299316406,
+      "learning_rate": 2.6529187817258883e-05,
+      "loss": 6.7831,
+      "step": 4000
+    },
+    {
+      "epoch": 2.750611246943765,
+      "grad_norm": 13.106438636779785,
+      "learning_rate": 2.3356598984771574e-05,
+      "loss": 6.6733,
+      "step": 4500
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 5.005562782287598,
+      "eval_map": 0.6553,
+      "eval_map_50": 0.8073,
+      "eval_map_75": 0.7122,
+      "eval_map_barcode": 0.6553,
+      "eval_map_large": 0.7359,
+      "eval_map_medium": 0.6782,
+      "eval_map_small": 0.2319,
+      "eval_mar_1": 0.3481,
+      "eval_mar_10": 0.8201,
+      "eval_mar_100": 0.8542,
+      "eval_mar_100_barcode": 0.8542,
+      "eval_mar_large": 0.8712,
+      "eval_mar_medium": 0.8573,
+      "eval_mar_small": 0.6107,
+      "eval_runtime": 62.2135,
+      "eval_samples_per_second": 45.071,
+      "eval_steps_per_second": 5.642,
+      "step": 4908
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 8180,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.268295119107072e+18,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-4908/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b614e8f3b3598fc33e0b4cc64963b79aec4b9857f0dbc4ca1781747b2531cea
+size 5905