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: 0.8008
-- Accuracy: 0.7159
+- Loss: 0.8354
+- Accuracy: 0.7273
 
 ## Model description
 

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 32.0,
+    "eval_accuracy": 0.7272727272727273,
+    "eval_loss": 0.835405170917511,
+    "eval_runtime": 1.6358,
+    "eval_samples_per_second": 53.795,
+    "eval_steps_per_second": 1.834,
+    "total_flos": 5.861874920136376e+17,
+    "train_loss": 0.7368841990828514,
+    "train_runtime": 406.4801,
+    "train_samples_per_second": 55.402,
+    "train_steps_per_second": 0.394
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 32.0,
+    "eval_accuracy": 0.7272727272727273,
+    "eval_loss": 0.835405170917511,
+    "eval_runtime": 1.6358,
+    "eval_samples_per_second": 53.795,
+    "eval_steps_per_second": 1.834
+}

runs/Feb09_20-05-06_e92dafee4e15/events.out.tfevents.1739132057.e92dafee4e15.2938.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03869957b53c5e328fcf6718097c872e3b1f0aaf53ed2cc0c62fb8b0cc410a4f
+size 411

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 32.0,
+    "total_flos": 5.861874920136376e+17,
+    "train_loss": 0.7368841990828514,
+    "train_runtime": 406.4801,
+    "train_samples_per_second": 55.402,
+    "train_steps_per_second": 0.394
+}

trainer_state.json

@@ -0,0 +1,421 @@
+{
+  "best_metric": 0.7272727272727273,
+  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-humeda-DAV50/checkpoint-105",
+  "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": 0.38636363636363635,
+      "eval_loss": 1.545121669769287,
+      "eval_runtime": 4.5369,
+      "eval_samples_per_second": 19.396,
+      "eval_steps_per_second": 0.661,
+      "step": 5
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.38636363636363635,
+      "eval_loss": 1.521955966949463,
+      "eval_runtime": 1.746,
+      "eval_samples_per_second": 50.402,
+      "eval_steps_per_second": 1.718,
+      "step": 10
+    },
+    {
+      "epoch": 2.4444444444444446,
+      "grad_norm": 3.2259411811828613,
+      "learning_rate": 1.125e-05,
+      "loss": 1.4177,
+      "step": 12
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.42045454545454547,
+      "eval_loss": 1.49383544921875,
+      "eval_runtime": 2.224,
+      "eval_samples_per_second": 39.568,
+      "eval_steps_per_second": 1.349,
+      "step": 15
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.4431818181818182,
+      "eval_loss": 1.411116123199463,
+      "eval_runtime": 1.5468,
+      "eval_samples_per_second": 56.89,
+      "eval_steps_per_second": 1.939,
+      "step": 20
+    },
+    {
+      "epoch": 4.888888888888889,
+      "grad_norm": 4.5672287940979,
+      "learning_rate": 1.488605814759156e-05,
+      "loss": 1.2671,
+      "step": 24
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.45454545454545453,
+      "eval_loss": 1.2941091060638428,
+      "eval_runtime": 1.5259,
+      "eval_samples_per_second": 57.671,
+      "eval_steps_per_second": 1.966,
+      "step": 25
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.45454545454545453,
+      "eval_loss": 1.2036188840866089,
+      "eval_runtime": 1.4933,
+      "eval_samples_per_second": 58.931,
+      "eval_steps_per_second": 2.009,
+      "step": 30
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5113636363636364,
+      "eval_loss": 1.081556797027588,
+      "eval_runtime": 1.5572,
+      "eval_samples_per_second": 56.513,
+      "eval_steps_per_second": 1.927,
+      "step": 35
+    },
+    {
+      "epoch": 7.222222222222222,
+      "grad_norm": 4.809324741363525,
+      "learning_rate": 1.4297308402774876e-05,
+      "loss": 0.9869,
+      "step": 36
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5795454545454546,
+      "eval_loss": 1.045181155204773,
+      "eval_runtime": 1.7801,
+      "eval_samples_per_second": 49.435,
+      "eval_steps_per_second": 1.685,
+      "step": 40
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.625,
+      "eval_loss": 0.9875959753990173,
+      "eval_runtime": 3.6966,
+      "eval_samples_per_second": 23.805,
+      "eval_steps_per_second": 0.812,
+      "step": 45
+    },
+    {
+      "epoch": 9.666666666666666,
+      "grad_norm": 6.9938645362854,
+      "learning_rate": 1.3349133623196613e-05,
+      "loss": 0.8456,
+      "step": 48
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.5909090909090909,
+      "eval_loss": 0.9790555834770203,
+      "eval_runtime": 2.2466,
+      "eval_samples_per_second": 39.171,
+      "eval_steps_per_second": 1.335,
+      "step": 50
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.6022727272727273,
+      "eval_loss": 0.9661941528320312,
+      "eval_runtime": 2.2611,
+      "eval_samples_per_second": 38.919,
+      "eval_steps_per_second": 1.327,
+      "step": 55
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 5.75779390335083,
+      "learning_rate": 1.1934822362726869e-05,
+      "loss": 0.7126,
+      "step": 60
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.6363636363636364,
+      "eval_loss": 0.9301889538764954,
+      "eval_runtime": 1.8721,
+      "eval_samples_per_second": 47.006,
+      "eval_steps_per_second": 1.602,
+      "step": 60
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.625,
+      "eval_loss": 0.937901496887207,
+      "eval_runtime": 1.5045,
+      "eval_samples_per_second": 58.489,
+      "eval_steps_per_second": 1.994,
+      "step": 65
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.6704545454545454,
+      "eval_loss": 0.9035769104957581,
+      "eval_runtime": 1.5233,
+      "eval_samples_per_second": 57.771,
+      "eval_steps_per_second": 1.969,
+      "step": 70
+    },
+    {
+      "epoch": 14.444444444444445,
+      "grad_norm": 6.649902820587158,
+      "learning_rate": 1.0218285287127763e-05,
+      "loss": 0.6561,
+      "step": 72
+    },
+    {
+      "epoch": 15.0,
+      "eval_accuracy": 0.6590909090909091,
+      "eval_loss": 0.8846054077148438,
+      "eval_runtime": 1.6087,
+      "eval_samples_per_second": 54.702,
+      "eval_steps_per_second": 1.865,
+      "step": 75
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.6590909090909091,
+      "eval_loss": 0.8689384460449219,
+      "eval_runtime": 1.5843,
+      "eval_samples_per_second": 55.545,
+      "eval_steps_per_second": 1.894,
+      "step": 80
+    },
+    {
+      "epoch": 16.88888888888889,
+      "grad_norm": 6.314008712768555,
+      "learning_rate": 8.316501561389734e-06,
+      "loss": 0.6367,
+      "step": 84
+    },
+    {
+      "epoch": 17.0,
+      "eval_accuracy": 0.6590909090909091,
+      "eval_loss": 0.8543261885643005,
+      "eval_runtime": 2.1048,
+      "eval_samples_per_second": 41.809,
+      "eval_steps_per_second": 1.425,
+      "step": 85
+    },
+    {
+      "epoch": 18.0,
+      "eval_accuracy": 0.6931818181818182,
+      "eval_loss": 0.8342323899269104,
+      "eval_runtime": 2.2453,
+      "eval_samples_per_second": 39.193,
+      "eval_steps_per_second": 1.336,
+      "step": 90
+    },
+    {
+      "epoch": 19.0,
+      "eval_accuracy": 0.6704545454545454,
+      "eval_loss": 0.8184678554534912,
+      "eval_runtime": 1.549,
+      "eval_samples_per_second": 56.813,
+      "eval_steps_per_second": 1.937,
+      "step": 95
+    },
+    {
+      "epoch": 19.22222222222222,
+      "grad_norm": 7.302158355712891,
+      "learning_rate": 6.359074603575625e-06,
+      "loss": 0.5463,
+      "step": 96
+    },
+    {
+      "epoch": 20.0,
+      "eval_accuracy": 0.7159090909090909,
+      "eval_loss": 0.8290252089500427,
+      "eval_runtime": 1.5193,
+      "eval_samples_per_second": 57.922,
+      "eval_steps_per_second": 1.975,
+      "step": 100
+    },
+    {
+      "epoch": 21.0,
+      "eval_accuracy": 0.7272727272727273,
+      "eval_loss": 0.835405170917511,
+      "eval_runtime": 1.5635,
+      "eval_samples_per_second": 56.283,
+      "eval_steps_per_second": 1.919,
+      "step": 105
+    },
+    {
+      "epoch": 21.666666666666668,
+      "grad_norm": 6.013044834136963,
+      "learning_rate": 4.479399826059472e-06,
+      "loss": 0.5504,
+      "step": 108
+    },
+    {
+      "epoch": 22.0,
+      "eval_accuracy": 0.7159090909090909,
+      "eval_loss": 0.8160114884376526,
+      "eval_runtime": 1.5625,
+      "eval_samples_per_second": 56.322,
+      "eval_steps_per_second": 1.92,
+      "step": 110
+    },
+    {
+      "epoch": 23.0,
+      "eval_accuracy": 0.7159090909090909,
+      "eval_loss": 0.8073451519012451,
+      "eval_runtime": 1.5277,
+      "eval_samples_per_second": 57.602,
+      "eval_steps_per_second": 1.964,
+      "step": 115
+    },
+    {
+      "epoch": 24.0,
+      "grad_norm": 9.277298927307129,
+      "learning_rate": 2.805573958619557e-06,
+      "loss": 0.507,
+      "step": 120
+    },
+    {
+      "epoch": 24.0,
+      "eval_accuracy": 0.7045454545454546,
+      "eval_loss": 0.8071027994155884,
+      "eval_runtime": 1.8337,
+      "eval_samples_per_second": 47.99,
+      "eval_steps_per_second": 1.636,
+      "step": 120
+    },
+    {
+      "epoch": 25.0,
+      "eval_accuracy": 0.6931818181818182,
+      "eval_loss": 0.8071491122245789,
+      "eval_runtime": 2.228,
+      "eval_samples_per_second": 39.498,
+      "eval_steps_per_second": 1.347,
+      "step": 125
+    },
+    {
+      "epoch": 26.0,
+      "eval_accuracy": 0.7045454545454546,
+      "eval_loss": 0.804684579372406,
+      "eval_runtime": 2.3318,
+      "eval_samples_per_second": 37.74,
+      "eval_steps_per_second": 1.287,
+      "step": 130
+    },
+    {
+      "epoch": 26.444444444444443,
+      "grad_norm": 6.318145751953125,
+      "learning_rate": 1.45166546799388e-06,
+      "loss": 0.5226,
+      "step": 132
+    },
+    {
+      "epoch": 27.0,
+      "eval_accuracy": 0.7045454545454546,
+      "eval_loss": 0.8000316023826599,
+      "eval_runtime": 1.5537,
+      "eval_samples_per_second": 56.64,
+      "eval_steps_per_second": 1.931,
+      "step": 135
+    },
+    {
+      "epoch": 28.0,
+      "eval_accuracy": 0.7159090909090909,
+      "eval_loss": 0.7987006902694702,
+      "eval_runtime": 1.565,
+      "eval_samples_per_second": 56.23,
+      "eval_steps_per_second": 1.917,
+      "step": 140
+    },
+    {
+      "epoch": 28.88888888888889,
+      "grad_norm": 8.081077575683594,
+      "learning_rate": 5.099409803790118e-07,
+      "loss": 0.5144,
+      "step": 144
+    },
+    {
+      "epoch": 29.0,
+      "eval_accuracy": 0.7159090909090909,
+      "eval_loss": 0.7999542355537415,
+      "eval_runtime": 1.5379,
+      "eval_samples_per_second": 57.222,
+      "eval_steps_per_second": 1.951,
+      "step": 145
+    },
+    {
+      "epoch": 30.0,
+      "eval_accuracy": 0.7159090909090909,
+      "eval_loss": 0.8002479672431946,
+      "eval_runtime": 1.5402,
+      "eval_samples_per_second": 57.135,
+      "eval_steps_per_second": 1.948,
+      "step": 150
+    },
+    {
+      "epoch": 31.0,
+      "eval_accuracy": 0.7159090909090909,
+      "eval_loss": 0.8008334636688232,
+      "eval_runtime": 1.5554,
+      "eval_samples_per_second": 56.577,
+      "eval_steps_per_second": 1.929,
+      "step": 155
+    },
+    {
+      "epoch": 31.22222222222222,
+      "grad_norm": 9.235086441040039,
+      "learning_rate": 4.4577463332602655e-08,
+      "loss": 0.4862,
+      "step": 156
+    },
+    {
+      "epoch": 32.0,
+      "eval_accuracy": 0.7159090909090909,
+      "eval_loss": 0.8007765412330627,
+      "eval_runtime": 1.9339,
+      "eval_samples_per_second": 45.503,
+      "eval_steps_per_second": 1.551,
+      "step": 160
+    },
+    {
+      "epoch": 32.0,
+      "step": 160,
+      "total_flos": 5.861874920136376e+17,
+      "train_loss": 0.7368841990828514,
+      "train_runtime": 406.4801,
+      "train_samples_per_second": 55.402,
+      "train_steps_per_second": 0.394
+    }
+  ],
+  "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
+}