Upload folder using huggingface_hub

config.json

@@ -0,0 +1,52 @@
+{
+  "_name_or_path": "facebook/dinov2-with-registers-small",
+  "apply_layernorm": true,
+  "architectures": [
+    "Dinov2WithRegistersForImageClassification"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "drop_path_rate": 0.0,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 384,
+  "image_size": 518,
+  "initializer_range": 0.02,
+  "interpolate_antialias": true,
+  "interpolate_offset": 0.0,
+  "layer_norm_eps": 1e-06,
+  "layerscale_value": 1.0,
+  "mlp_ratio": 4,
+  "model_type": "dinov2_with_registers",
+  "num_attention_heads": 6,
+  "num_channels": 3,
+  "num_hidden_layers": 12,
+  "num_register_tokens": 4,
+  "out_features": [
+    "stage12"
+  ],
+  "out_indices": [
+    12
+  ],
+  "patch_size": 14,
+  "problem_type": "single_label_classification",
+  "qkv_bias": true,
+  "reshape_hidden_states": true,
+  "stage_names": [
+    "stem",
+    "stage1",
+    "stage2",
+    "stage3",
+    "stage4",
+    "stage5",
+    "stage6",
+    "stage7",
+    "stage8",
+    "stage9",
+    "stage10",
+    "stage11",
+    "stage12"
+  ],
+  "torch_dtype": "float32",
+  "transformers_version": "4.48.0",
+  "use_swiglu_ffn": false
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ae32286831883c3b1833ed0c21764008ee3cb690aa947501478767a461c5a83
+size 88267432

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eef0d344dc7c5e9180eda212cdb90b24bbc7adb1e6748db6d283d9c84790b76a
+size 176666106

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d138cfe3a4adf21f048848ee35837c9a757a0a3616ff7adbb45b69aac247435
+size 14244

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25d8f282f995a3c1a59d83dfe504bd41bcdf8f79e071dbd6f372623251fc02b2
+size 1064

trainer_state.json

@@ -0,0 +1,177 @@
+{
+  "best_metric": 0.9713101001869517,
+  "best_model_checkpoint": "./results/checkpoint-1500",
+  "epoch": 0.2901353965183752,
+  "eval_steps": 500,
+  "global_step": 1500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.019342359767891684,
+      "grad_norm": 1871413.875,
+      "learning_rate": 1e-05,
+      "loss": 0.6564,
+      "step": 100
+    },
+    {
+      "epoch": 0.03868471953578337,
+      "grad_norm": 3112622.75,
+      "learning_rate": 2e-05,
+      "loss": 0.4794,
+      "step": 200
+    },
+    {
+      "epoch": 0.058027079303675046,
+      "grad_norm": 2537678.5,
+      "learning_rate": 3e-05,
+      "loss": 0.4037,
+      "step": 300
+    },
+    {
+      "epoch": 0.07736943907156674,
+      "grad_norm": 2662464.0,
+      "learning_rate": 4e-05,
+      "loss": 0.3265,
+      "step": 400
+    },
+    {
+      "epoch": 0.09671179883945841,
+      "grad_norm": 1437540.625,
+      "learning_rate": 5e-05,
+      "loss": 0.3744,
+      "step": 500
+    },
+    {
+      "epoch": 0.09671179883945841,
+      "eval_accuracy": 0.8466634429400387,
+      "eval_f1": 0.8663012543480553,
+      "eval_loss": 0.33316776156425476,
+      "eval_precision": 0.7684238083662135,
+      "eval_recall": 0.9927523977484116,
+      "eval_roc_auc": 0.8465467895292255,
+      "eval_runtime": 1179.0948,
+      "eval_samples_per_second": 140.311,
+      "eval_steps_per_second": 0.549,
+      "step": 500
+    },
+    {
+      "epoch": 0.11605415860735009,
+      "grad_norm": 2934780.75,
+      "learning_rate": 4.892933618843683e-05,
+      "loss": 0.3006,
+      "step": 600
+    },
+    {
+      "epoch": 0.13539651837524178,
+      "grad_norm": 1091141.375,
+      "learning_rate": 4.785867237687366e-05,
+      "loss": 0.2714,
+      "step": 700
+    },
+    {
+      "epoch": 0.15473887814313347,
+      "grad_norm": 3842419.0,
+      "learning_rate": 4.6788008565310496e-05,
+      "loss": 0.1778,
+      "step": 800
+    },
+    {
+      "epoch": 0.17408123791102514,
+      "grad_norm": 1596435.75,
+      "learning_rate": 4.571734475374733e-05,
+      "loss": 0.1743,
+      "step": 900
+    },
+    {
+      "epoch": 0.19342359767891681,
+      "grad_norm": 1802532.125,
+      "learning_rate": 4.4646680942184155e-05,
+      "loss": 0.1531,
+      "step": 1000
+    },
+    {
+      "epoch": 0.19342359767891681,
+      "eval_accuracy": 0.9334683268858801,
+      "eval_f1": 0.929946602343387,
+      "eval_loss": 0.16633380949497223,
+      "eval_precision": 0.9827945706714019,
+      "eval_recall": 0.8824922088275795,
+      "eval_roc_auc": 0.9335090317978244,
+      "eval_runtime": 970.6452,
+      "eval_samples_per_second": 170.443,
+      "eval_steps_per_second": 0.667,
+      "step": 1000
+    },
+    {
+      "epoch": 0.2127659574468085,
+      "grad_norm": 1385128.5,
+      "learning_rate": 4.357601713062099e-05,
+      "loss": 0.1432,
+      "step": 1100
+    },
+    {
+      "epoch": 0.23210831721470018,
+      "grad_norm": 306230.6875,
+      "learning_rate": 4.2505353319057815e-05,
+      "loss": 0.1386,
+      "step": 1200
+    },
+    {
+      "epoch": 0.2514506769825919,
+      "grad_norm": 1276595.5,
+      "learning_rate": 4.143468950749465e-05,
+      "loss": 0.1273,
+      "step": 1300
+    },
+    {
+      "epoch": 0.27079303675048355,
+      "grad_norm": 417115.5625,
+      "learning_rate": 4.036402569593148e-05,
+      "loss": 0.1075,
+      "step": 1400
+    },
+    {
+      "epoch": 0.2901353965183752,
+      "grad_norm": 1014551.625,
+      "learning_rate": 3.9293361884368315e-05,
+      "loss": 0.1314,
+      "step": 1500
+    },
+    {
+      "epoch": 0.2901353965183752,
+      "eval_accuracy": 0.971058994197292,
+      "eval_f1": 0.9713101001869517,
+      "eval_loss": 0.08320203423500061,
+      "eval_precision": 0.9637107322061307,
+      "eval_recall": 0.9790302708187375,
+      "eval_roc_auc": 0.971052629057589,
+      "eval_runtime": 954.2031,
+      "eval_samples_per_second": 173.38,
+      "eval_steps_per_second": 0.678,
+      "step": 1500
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 5170,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "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": 4.9963434246144e+18,
+  "train_batch_size": 128,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:061af69f446ec4684d6cdb1ee48c5bd4e0e460f08c3bbc6485e12e3606db3bfa
+size 5304