End of training

README.md

@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6486
-- Accuracy: 0.8068
+- Loss: 0.6495
+- Accuracy: 0.8182
 
 ## Model description
 

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 32.0,
+    "eval_accuracy": 0.8181818181818182,
+    "eval_loss": 0.6495000720024109,
+    "eval_runtime": 2.4398,
+    "eval_samples_per_second": 36.069,
+    "eval_steps_per_second": 1.23,
+    "total_flos": 5.861874920136376e+17,
+    "train_loss": 0.4995464000850916,
+    "train_runtime": 456.7219,
+    "train_samples_per_second": 49.308,
+    "train_steps_per_second": 0.35
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 32.0,
+    "eval_accuracy": 0.8181818181818182,
+    "eval_loss": 0.6495000720024109,
+    "eval_runtime": 2.4398,
+    "eval_samples_per_second": 36.069,
+    "eval_steps_per_second": 1.23
+}

runs/Feb09_20-39-23_e92dafee4e15/events.out.tfevents.1739134112.e92dafee4e15.11219.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36549d73c0d09e616b26a31479b09c6e929f991a319608d5dc70eddc02c95484
+size 411

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 32.0,
+    "total_flos": 5.861874920136376e+17,
+    "train_loss": 0.4995464000850916,
+    "train_runtime": 456.7219,
+    "train_samples_per_second": 49.308,
+    "train_steps_per_second": 0.35
+}

trainer_state.json

@@ -0,0 +1,421 @@
+{
+  "best_metric": 0.8181818181818182,
+  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-humeda-DAV51/checkpoint-135",
+  "epoch": 32.0,
+  "eval_steps": 500,
+  "global_step": 160,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.3409090909090909,
+      "eval_loss": 1.5572398900985718,
+      "eval_runtime": 1.8995,
+      "eval_samples_per_second": 46.328,
+      "eval_steps_per_second": 1.579,
+      "step": 5
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.45454545454545453,
+      "eval_loss": 1.3889658451080322,
+      "eval_runtime": 1.5085,
+      "eval_samples_per_second": 58.337,
+      "eval_steps_per_second": 1.989,
+      "step": 10
+    },
+    {
+      "epoch": 2.4444444444444446,
+      "grad_norm": 3.2280192375183105,
+      "learning_rate": 3.7500000000000003e-05,
+      "loss": 1.4186,
+      "step": 12
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5795454545454546,
+      "eval_loss": 1.2637884616851807,
+      "eval_runtime": 1.5778,
+      "eval_samples_per_second": 55.775,
+      "eval_steps_per_second": 1.901,
+      "step": 15
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6022727272727273,
+      "eval_loss": 1.0291478633880615,
+      "eval_runtime": 1.9454,
+      "eval_samples_per_second": 45.235,
+      "eval_steps_per_second": 1.542,
+      "step": 20
+    },
+    {
+      "epoch": 4.888888888888889,
+      "grad_norm": 10.705877304077148,
+      "learning_rate": 4.962019382530521e-05,
+      "loss": 1.0199,
+      "step": 24
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5909090909090909,
+      "eval_loss": 1.0125187635421753,
+      "eval_runtime": 1.5559,
+      "eval_samples_per_second": 56.559,
+      "eval_steps_per_second": 1.928,
+      "step": 25
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6477272727272727,
+      "eval_loss": 0.8327766060829163,
+      "eval_runtime": 2.1281,
+      "eval_samples_per_second": 41.351,
+      "eval_steps_per_second": 1.41,
+      "step": 30
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.625,
+      "eval_loss": 0.8662083148956299,
+      "eval_runtime": 2.2359,
+      "eval_samples_per_second": 39.357,
+      "eval_steps_per_second": 1.342,
+      "step": 35
+    },
+    {
+      "epoch": 7.222222222222222,
+      "grad_norm": 11.93485164642334,
+      "learning_rate": 4.765769467591625e-05,
+      "loss": 0.7093,
+      "step": 36
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.7045454545454546,
+      "eval_loss": 0.7048011422157288,
+      "eval_runtime": 2.2585,
+      "eval_samples_per_second": 38.964,
+      "eval_steps_per_second": 1.328,
+      "step": 40
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6818181818181818,
+      "eval_loss": 0.8032174706459045,
+      "eval_runtime": 1.541,
+      "eval_samples_per_second": 57.107,
+      "eval_steps_per_second": 1.947,
+      "step": 45
+    },
+    {
+      "epoch": 9.666666666666666,
+      "grad_norm": 9.79036808013916,
+      "learning_rate": 4.415111107797445e-05,
+      "loss": 0.576,
+      "step": 48
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.7272727272727273,
+      "eval_loss": 0.6944234371185303,
+      "eval_runtime": 1.5414,
+      "eval_samples_per_second": 57.092,
+      "eval_steps_per_second": 1.946,
+      "step": 50
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.6931818181818182,
+      "eval_loss": 0.7729998230934143,
+      "eval_runtime": 1.5448,
+      "eval_samples_per_second": 56.966,
+      "eval_steps_per_second": 1.942,
+      "step": 55
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 7.894675254821777,
+      "learning_rate": 3.933941090877615e-05,
+      "loss": 0.4817,
+      "step": 60
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.7386363636363636,
+      "eval_loss": 0.6605115532875061,
+      "eval_runtime": 1.5875,
+      "eval_samples_per_second": 55.433,
+      "eval_steps_per_second": 1.89,
+      "step": 60
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.75,
+      "eval_loss": 0.7315654754638672,
+      "eval_runtime": 1.5216,
+      "eval_samples_per_second": 57.834,
+      "eval_steps_per_second": 1.972,
+      "step": 65
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.7727272727272727,
+      "eval_loss": 0.6380431652069092,
+      "eval_runtime": 1.5391,
+      "eval_samples_per_second": 57.175,
+      "eval_steps_per_second": 1.949,
+      "step": 70
+    },
+    {
+      "epoch": 14.444444444444445,
+      "grad_norm": 7.401230812072754,
+      "learning_rate": 3.355050358314172e-05,
+      "loss": 0.413,
+      "step": 72
+    },
+    {
+      "epoch": 15.0,
+      "eval_accuracy": 0.7727272727272727,
+      "eval_loss": 0.657296895980835,
+      "eval_runtime": 1.5157,
+      "eval_samples_per_second": 58.057,
+      "eval_steps_per_second": 1.979,
+      "step": 75
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.75,
+      "eval_loss": 0.6569694876670837,
+      "eval_runtime": 1.5591,
+      "eval_samples_per_second": 56.442,
+      "eval_steps_per_second": 1.924,
+      "step": 80
+    },
+    {
+      "epoch": 16.88888888888889,
+      "grad_norm": 7.306827068328857,
+      "learning_rate": 2.717889356869146e-05,
+      "loss": 0.3959,
+      "step": 84
+    },
+    {
+      "epoch": 17.0,
+      "eval_accuracy": 0.7954545454545454,
+      "eval_loss": 0.6172751784324646,
+      "eval_runtime": 1.5252,
+      "eval_samples_per_second": 57.698,
+      "eval_steps_per_second": 1.967,
+      "step": 85
+    },
+    {
+      "epoch": 18.0,
+      "eval_accuracy": 0.7840909090909091,
+      "eval_loss": 0.6292572617530823,
+      "eval_runtime": 1.5219,
+      "eval_samples_per_second": 57.823,
+      "eval_steps_per_second": 1.971,
+      "step": 90
+    },
+    {
+      "epoch": 19.0,
+      "eval_accuracy": 0.7727272727272727,
+      "eval_loss": 0.6490908265113831,
+      "eval_runtime": 1.554,
+      "eval_samples_per_second": 56.627,
+      "eval_steps_per_second": 1.93,
+      "step": 95
+    },
+    {
+      "epoch": 19.22222222222222,
+      "grad_norm": 8.758996963500977,
+      "learning_rate": 2.0658795558326743e-05,
+      "loss": 0.3043,
+      "step": 96
+    },
+    {
+      "epoch": 20.0,
+      "eval_accuracy": 0.7954545454545454,
+      "eval_loss": 0.6381738781929016,
+      "eval_runtime": 1.5294,
+      "eval_samples_per_second": 57.539,
+      "eval_steps_per_second": 1.962,
+      "step": 100
+    },
+    {
+      "epoch": 21.0,
+      "eval_accuracy": 0.7954545454545454,
+      "eval_loss": 0.6272431015968323,
+      "eval_runtime": 1.5171,
+      "eval_samples_per_second": 58.004,
+      "eval_steps_per_second": 1.977,
+      "step": 105
+    },
+    {
+      "epoch": 21.666666666666668,
+      "grad_norm": 6.60971212387085,
+      "learning_rate": 1.443454345648252e-05,
+      "loss": 0.295,
+      "step": 108
+    },
+    {
+      "epoch": 22.0,
+      "eval_accuracy": 0.8068181818181818,
+      "eval_loss": 0.6422553062438965,
+      "eval_runtime": 1.695,
+      "eval_samples_per_second": 51.916,
+      "eval_steps_per_second": 1.77,
+      "step": 110
+    },
+    {
+      "epoch": 23.0,
+      "eval_accuracy": 0.8068181818181818,
+      "eval_loss": 0.6413140296936035,
+      "eval_runtime": 1.5383,
+      "eval_samples_per_second": 57.205,
+      "eval_steps_per_second": 1.95,
+      "step": 115
+    },
+    {
+      "epoch": 24.0,
+      "grad_norm": 5.603902816772461,
+      "learning_rate": 8.930309757836517e-06,
+      "loss": 0.2365,
+      "step": 120
+    },
+    {
+      "epoch": 24.0,
+      "eval_accuracy": 0.7840909090909091,
+      "eval_loss": 0.6388431191444397,
+      "eval_runtime": 1.5038,
+      "eval_samples_per_second": 58.516,
+      "eval_steps_per_second": 1.995,
+      "step": 120
+    },
+    {
+      "epoch": 25.0,
+      "eval_accuracy": 0.7840909090909091,
+      "eval_loss": 0.6457162499427795,
+      "eval_runtime": 1.9638,
+      "eval_samples_per_second": 44.81,
+      "eval_steps_per_second": 1.528,
+      "step": 125
+    },
+    {
+      "epoch": 26.0,
+      "eval_accuracy": 0.7954545454545454,
+      "eval_loss": 0.6513168215751648,
+      "eval_runtime": 2.1698,
+      "eval_samples_per_second": 40.556,
+      "eval_steps_per_second": 1.383,
+      "step": 130
+    },
+    {
+      "epoch": 26.444444444444443,
+      "grad_norm": 7.909633159637451,
+      "learning_rate": 4.521198892775203e-06,
+      "loss": 0.2507,
+      "step": 132
+    },
+    {
+      "epoch": 27.0,
+      "eval_accuracy": 0.8181818181818182,
+      "eval_loss": 0.6495000720024109,
+      "eval_runtime": 1.5676,
+      "eval_samples_per_second": 56.138,
+      "eval_steps_per_second": 1.914,
+      "step": 135
+    },
+    {
+      "epoch": 28.0,
+      "eval_accuracy": 0.8181818181818182,
+      "eval_loss": 0.6463306546211243,
+      "eval_runtime": 1.5106,
+      "eval_samples_per_second": 58.255,
+      "eval_steps_per_second": 1.986,
+      "step": 140
+    },
+    {
+      "epoch": 28.88888888888889,
+      "grad_norm": 10.235626220703125,
+      "learning_rate": 1.5076844803522922e-06,
+      "loss": 0.2385,
+      "step": 144
+    },
+    {
+      "epoch": 29.0,
+      "eval_accuracy": 0.8068181818181818,
+      "eval_loss": 0.6467605233192444,
+      "eval_runtime": 1.5371,
+      "eval_samples_per_second": 57.25,
+      "eval_steps_per_second": 1.952,
+      "step": 145
+    },
+    {
+      "epoch": 30.0,
+      "eval_accuracy": 0.8068181818181818,
+      "eval_loss": 0.6479575037956238,
+      "eval_runtime": 1.5003,
+      "eval_samples_per_second": 58.654,
+      "eval_steps_per_second": 2.0,
+      "step": 150
+    },
+    {
+      "epoch": 31.0,
+      "eval_accuracy": 0.8068181818181818,
+      "eval_loss": 0.648425817489624,
+      "eval_runtime": 1.681,
+      "eval_samples_per_second": 52.35,
+      "eval_steps_per_second": 1.785,
+      "step": 155
+    },
+    {
+      "epoch": 31.22222222222222,
+      "grad_norm": 6.484171390533447,
+      "learning_rate": 9.513254770636137e-08,
+      "loss": 0.2432,
+      "step": 156
+    },
+    {
+      "epoch": 32.0,
+      "eval_accuracy": 0.8068181818181818,
+      "eval_loss": 0.6486064195632935,
+      "eval_runtime": 2.1912,
+      "eval_samples_per_second": 40.16,
+      "eval_steps_per_second": 1.369,
+      "step": 160
+    },
+    {
+      "epoch": 32.0,
+      "step": 160,
+      "total_flos": 5.861874920136376e+17,
+      "train_loss": 0.4995464000850916,
+      "train_runtime": 456.7219,
+      "train_samples_per_second": 49.308,
+      "train_steps_per_second": 0.35
+    }
+  ],
+  "logging_steps": 12,
+  "max_steps": 160,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 40,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.861874920136376e+17,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}