Training in progress, epoch 15

config.json

@@ -45,6 +45,7 @@
     12
   ],
   "patch_size": 14,
+  "problem_type": "single_label_classification",
   "qkv_bias": true,
   "reshape_hidden_states": true,
   "stage_names": [

logs/events.out.tfevents.1738395705.6d188e23525e.4440.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ce66af159782c25d8db9b796b8105aab875514034a78a85bb451472f12705b8d
-size 88
+oid sha256:e5b4775155bffa2402966a6b54e096e5e3675a53dd20de7c677f7147358b299d
+size 6215

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:75382291af57fc53437ca2826c098c00299a87fa4fb3f117fe4ecf2c117eca44
+oid sha256:5481bf2dc65152cb86b77c66aa22f81ae2a3537bfbef87f912f42dbeba3c8d4f
 size 346396816

trainer_state.json

@@ -0,0 +1,257 @@
+{
+  "best_metric": 0.11363224685192108,
+  "best_model_checkpoint": "./drive/Shareddrives/CS198-Drones/training_output/dinov2-base_rice-leaf-disease-augmented_tl_020125/checkpoint-3500",
+  "epoch": 14.0,
+  "eval_steps": 500,
+  "global_step": 3500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 5.269941329956055,
+      "learning_rate": 0.0006666666666666666,
+      "loss": 1.1218,
+      "step": 250
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.825,
+      "eval_loss": 0.5307279229164124,
+      "eval_runtime": 107.9568,
+      "eval_samples_per_second": 18.526,
+      "eval_steps_per_second": 0.296,
+      "step": 250
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 3.551548719406128,
+      "learning_rate": 0.0009629629629629629,
+      "loss": 0.4069,
+      "step": 500
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.885,
+      "eval_loss": 0.3582110106945038,
+      "eval_runtime": 106.2413,
+      "eval_samples_per_second": 18.825,
+      "eval_steps_per_second": 0.301,
+      "step": 500
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 2.2986912727355957,
+      "learning_rate": 0.0008888888888888888,
+      "loss": 0.2746,
+      "step": 750
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.92,
+      "eval_loss": 0.24640271067619324,
+      "eval_runtime": 105.657,
+      "eval_samples_per_second": 18.929,
+      "eval_steps_per_second": 0.303,
+      "step": 750
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 5.520092964172363,
+      "learning_rate": 0.0008148148148148148,
+      "loss": 0.2157,
+      "step": 1000
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9315,
+      "eval_loss": 0.22238598763942719,
+      "eval_runtime": 105.3986,
+      "eval_samples_per_second": 18.976,
+      "eval_steps_per_second": 0.304,
+      "step": 1000
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 2.029740333557129,
+      "learning_rate": 0.0007407407407407407,
+      "loss": 0.1779,
+      "step": 1250
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.949,
+      "eval_loss": 0.1752648651599884,
+      "eval_runtime": 105.061,
+      "eval_samples_per_second": 19.037,
+      "eval_steps_per_second": 0.305,
+      "step": 1250
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 2.1556665897369385,
+      "learning_rate": 0.0006666666666666666,
+      "loss": 0.1539,
+      "step": 1500
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.942,
+      "eval_loss": 0.1718205064535141,
+      "eval_runtime": 105.2523,
+      "eval_samples_per_second": 19.002,
+      "eval_steps_per_second": 0.304,
+      "step": 1500
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 2.0605902671813965,
+      "learning_rate": 0.0005925925925925926,
+      "loss": 0.1361,
+      "step": 1750
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.9515,
+      "eval_loss": 0.16032838821411133,
+      "eval_runtime": 106.2364,
+      "eval_samples_per_second": 18.826,
+      "eval_steps_per_second": 0.301,
+      "step": 1750
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.2079261541366577,
+      "learning_rate": 0.0005185185185185185,
+      "loss": 0.1271,
+      "step": 2000
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.958,
+      "eval_loss": 0.14475062489509583,
+      "eval_runtime": 104.8111,
+      "eval_samples_per_second": 19.082,
+      "eval_steps_per_second": 0.305,
+      "step": 2000
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 1.770965337753296,
+      "learning_rate": 0.0004444444444444444,
+      "loss": 0.1114,
+      "step": 2250
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.951,
+      "eval_loss": 0.144443079829216,
+      "eval_runtime": 106.9925,
+      "eval_samples_per_second": 18.693,
+      "eval_steps_per_second": 0.299,
+      "step": 2250
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.8546315431594849,
+      "learning_rate": 0.00037037037037037035,
+      "loss": 0.1023,
+      "step": 2500
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.959,
+      "eval_loss": 0.13085949420928955,
+      "eval_runtime": 105.0113,
+      "eval_samples_per_second": 19.046,
+      "eval_steps_per_second": 0.305,
+      "step": 2500
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 0.8261191248893738,
+      "learning_rate": 0.0002962962962962963,
+      "loss": 0.0968,
+      "step": 2750
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.9625,
+      "eval_loss": 0.1240282952785492,
+      "eval_runtime": 105.5924,
+      "eval_samples_per_second": 18.941,
+      "eval_steps_per_second": 0.303,
+      "step": 2750
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.7736948728561401,
+      "learning_rate": 0.0002222222222222222,
+      "loss": 0.0911,
+      "step": 3000
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.9645,
+      "eval_loss": 0.12481024116277695,
+      "eval_runtime": 106.6655,
+      "eval_samples_per_second": 18.75,
+      "eval_steps_per_second": 0.3,
+      "step": 3000
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 2.4998667240142822,
+      "learning_rate": 0.00014814814814814815,
+      "loss": 0.0858,
+      "step": 3250
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.962,
+      "eval_loss": 0.11893002688884735,
+      "eval_runtime": 107.3766,
+      "eval_samples_per_second": 18.626,
+      "eval_steps_per_second": 0.298,
+      "step": 3250
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.4618275165557861,
+      "learning_rate": 7.407407407407407e-05,
+      "loss": 0.0818,
+      "step": 3500
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.9645,
+      "eval_loss": 0.11363224685192108,
+      "eval_runtime": 102.5748,
+      "eval_samples_per_second": 19.498,
+      "eval_steps_per_second": 0.312,
+      "step": 3500
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 3750,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 15,
+  "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": 1.7518552646418432e+19,
+  "train_batch_size": 64,
+  "trial_name": null,
+  "trial_params": null
+}