Training in progress, epoch 1

all_results.json

@@ -1,5 +1,5 @@
 {
-    "epoch": 8.571428571428571,
+    "epoch": 10.0,
     "eval_accuracy": 0.9032258064516129,
     "eval_auc": 0.5,
     "eval_f1": 0.4745762711864407,
@@ -9,9 +9,9 @@
     "eval_runtime": 0.6301,
     "eval_samples_per_second": 295.208,
     "eval_steps_per_second": 4.761,
-    "total_flos": 1.8691703717363712e+17,
-    "train_loss": 0.2957582632700602,
-    "train_runtime": 47.4143,
-    "train_samples_per_second": 184.754,
-    "train_steps_per_second": 0.633
+    "total_flos": 1.1375723808188006e+18,
+    "train_loss": 0.21184425609452384,
+    "train_runtime": 168.3293,
+    "train_samples_per_second": 52.041,
+    "train_steps_per_second": 0.416
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4d07e7c41d6e3d022c28c3c64002a95b5a168904c992038f6dda631436efb1a4
+oid sha256:4ba391ebd43096538016e9666601807201abdf2a330498c741679a39b656df7d
 size 110342832

runs/May27_16-05-26_RTX3090/events.out.tfevents.1716793533.RTX3090.11668.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6300258e5aae69528f8891f392f6e6e86f1fb4ba969bbab08b06122104977ad
+size 5787

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 8.571428571428571,
-    "total_flos": 1.8691703717363712e+17,
-    "train_loss": 0.2957582632700602,
-    "train_runtime": 47.4143,
-    "train_samples_per_second": 184.754,
-    "train_steps_per_second": 0.633
+    "epoch": 10.0,
+    "total_flos": 1.1375723808188006e+18,
+    "train_loss": 0.21184425609452384,
+    "train_runtime": 168.3293,
+    "train_samples_per_second": 52.041,
+    "train_steps_per_second": 0.416
 }

trainer_state.json

@@ -1,163 +1,204 @@
 {
   "best_metric": 0.9032258064516129,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-RCC/checkpoint-3",
-  "epoch": 8.571428571428571,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-RCC/checkpoint-7",
+  "epoch": 10.0,
   "eval_steps": 500,
-  "global_step": 30,
+  "global_step": 70,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.8571428571428571,
+      "epoch": 1.0,
       "eval_accuracy": 0.9032258064516129,
       "eval_auc": 0.5,
       "eval_f1": 0.4745762711864407,
-      "eval_loss": 0.3023151159286499,
+      "eval_loss": 0.3295924961566925,
       "eval_precision": 0.9032258064516129,
       "eval_recall": 1.0,
-      "eval_runtime": 0.5993,
-      "eval_samples_per_second": 310.347,
-      "eval_steps_per_second": 5.006,
-      "step": 3
+      "eval_runtime": 1.1464,
+      "eval_samples_per_second": 162.249,
+      "eval_steps_per_second": 5.234,
+      "step": 7
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 0.9726035594940186,
+      "learning_rate": 4.761904761904762e-05,
+      "loss": 0.4795,
+      "step": 10
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 0.9032258064516129,
       "eval_auc": 0.5,
       "eval_f1": 0.4745762711864407,
-      "eval_loss": 0.36768200993537903,
+      "eval_loss": 0.31294792890548706,
       "eval_precision": 0.9032258064516129,
       "eval_recall": 1.0,
-      "eval_runtime": 0.6162,
-      "eval_samples_per_second": 301.832,
-      "eval_steps_per_second": 4.868,
-      "step": 7
+      "eval_runtime": 1.0901,
+      "eval_samples_per_second": 170.621,
+      "eval_steps_per_second": 5.504,
+      "step": 14
     },
     {
       "epoch": 2.857142857142857,
-      "grad_norm": 2.596181631088257,
-      "learning_rate": 3.7037037037037037e-05,
-      "loss": 0.4028,
-      "step": 10
+      "grad_norm": 0.6543672680854797,
+      "learning_rate": 3.968253968253968e-05,
+      "loss": 0.2503,
+      "step": 20
     },
     {
-      "epoch": 2.857142857142857,
+      "epoch": 3.0,
       "eval_accuracy": 0.9032258064516129,
       "eval_auc": 0.5,
       "eval_f1": 0.4745762711864407,
-      "eval_loss": 0.32757532596588135,
+      "eval_loss": 0.31820589303970337,
       "eval_precision": 0.9032258064516129,
       "eval_recall": 1.0,
-      "eval_runtime": 0.6008,
-      "eval_samples_per_second": 309.59,
-      "eval_steps_per_second": 4.993,
-      "step": 10
+      "eval_runtime": 1.1148,
+      "eval_samples_per_second": 166.845,
+      "eval_steps_per_second": 5.382,
+      "step": 21
     },
     {
       "epoch": 4.0,
       "eval_accuracy": 0.9032258064516129,
       "eval_auc": 0.5,
       "eval_f1": 0.4745762711864407,
-      "eval_loss": 0.32863086462020874,
+      "eval_loss": 0.2973339259624481,
       "eval_precision": 0.9032258064516129,
       "eval_recall": 1.0,
-      "eval_runtime": 0.6313,
-      "eval_samples_per_second": 294.647,
-      "eval_steps_per_second": 4.752,
-      "step": 14
+      "eval_runtime": 1.1837,
+      "eval_samples_per_second": 157.138,
+      "eval_steps_per_second": 5.069,
+      "step": 28
+    },
+    {
+      "epoch": 4.285714285714286,
+      "grad_norm": 1.603426218032837,
+      "learning_rate": 3.1746031746031745e-05,
+      "loss": 0.2231,
+      "step": 30
     },
     {
-      "epoch": 4.857142857142857,
+      "epoch": 5.0,
       "eval_accuracy": 0.9032258064516129,
       "eval_auc": 0.5,
       "eval_f1": 0.4745762711864407,
-      "eval_loss": 0.3271201252937317,
+      "eval_loss": 0.3274933993816376,
       "eval_precision": 0.9032258064516129,
       "eval_recall": 1.0,
-      "eval_runtime": 0.6283,
-      "eval_samples_per_second": 296.051,
-      "eval_steps_per_second": 4.775,
-      "step": 17
+      "eval_runtime": 1.1347,
+      "eval_samples_per_second": 163.913,
+      "eval_steps_per_second": 5.288,
+      "step": 35
     },
     {
       "epoch": 5.714285714285714,
-      "grad_norm": 1.759938359260559,
-      "learning_rate": 1.8518518518518518e-05,
-      "loss": 0.2494,
-      "step": 20
+      "grad_norm": 1.8515961170196533,
+      "learning_rate": 2.380952380952381e-05,
+      "loss": 0.1791,
+      "step": 40
     },
     {
       "epoch": 6.0,
       "eval_accuracy": 0.9032258064516129,
       "eval_auc": 0.5,
       "eval_f1": 0.4745762711864407,
-      "eval_loss": 0.32589343190193176,
+      "eval_loss": 0.31467387080192566,
       "eval_precision": 0.9032258064516129,
       "eval_recall": 1.0,
-      "eval_runtime": 0.607,
-      "eval_samples_per_second": 306.423,
-      "eval_steps_per_second": 4.942,
-      "step": 21
+      "eval_runtime": 1.1182,
+      "eval_samples_per_second": 166.344,
+      "eval_steps_per_second": 5.366,
+      "step": 42
     },
     {
-      "epoch": 6.857142857142857,
-      "eval_accuracy": 0.9032258064516129,
-      "eval_auc": 0.5,
-      "eval_f1": 0.4745762711864407,
-      "eval_loss": 0.3279106020927429,
-      "eval_precision": 0.9032258064516129,
-      "eval_recall": 1.0,
-      "eval_runtime": 0.6103,
-      "eval_samples_per_second": 304.771,
-      "eval_steps_per_second": 4.916,
-      "step": 24
+      "epoch": 7.0,
+      "eval_accuracy": 0.8978494623655914,
+      "eval_auc": 0.5218253968253969,
+      "eval_f1": 0.5205535205535206,
+      "eval_loss": 0.34014198184013367,
+      "eval_precision": 0.907103825136612,
+      "eval_recall": 0.9880952380952381,
+      "eval_runtime": 1.1252,
+      "eval_samples_per_second": 165.308,
+      "eval_steps_per_second": 5.333,
+      "step": 49
+    },
+    {
+      "epoch": 7.142857142857143,
+      "grad_norm": 1.4963322877883911,
+      "learning_rate": 1.5873015873015872e-05,
+      "loss": 0.1361,
+      "step": 50
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.9032258064516129,
-      "eval_auc": 0.5,
-      "eval_f1": 0.4745762711864407,
-      "eval_loss": 0.3266947269439697,
-      "eval_precision": 0.9032258064516129,
-      "eval_recall": 1.0,
-      "eval_runtime": 0.6353,
-      "eval_samples_per_second": 292.77,
-      "eval_steps_per_second": 4.722,
-      "step": 28
+      "eval_accuracy": 0.7849462365591398,
+      "eval_auc": 0.5833333333333335,
+      "eval_f1": 0.5528846153846154,
+      "eval_loss": 0.3884872794151306,
+      "eval_precision": 0.9210526315789473,
+      "eval_recall": 0.8333333333333334,
+      "eval_runtime": 1.1007,
+      "eval_samples_per_second": 168.985,
+      "eval_steps_per_second": 5.451,
+      "step": 56
     },
     {
       "epoch": 8.571428571428571,
-      "grad_norm": 2.163374423980713,
+      "grad_norm": 2.6748523712158203,
+      "learning_rate": 7.936507936507936e-06,
+      "loss": 0.1245,
+      "step": 60
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.8817204301075269,
+      "eval_auc": 0.5873015873015872,
+      "eval_f1": 0.6011695906432748,
+      "eval_loss": 0.31924736499786377,
+      "eval_precision": 0.9195402298850575,
+      "eval_recall": 0.9523809523809523,
+      "eval_runtime": 1.117,
+      "eval_samples_per_second": 166.524,
+      "eval_steps_per_second": 5.372,
+      "step": 63
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 3.4235951900482178,
       "learning_rate": 0.0,
-      "loss": 0.2351,
-      "step": 30
+      "loss": 0.0902,
+      "step": 70
     },
     {
-      "epoch": 8.571428571428571,
-      "eval_accuracy": 0.9032258064516129,
-      "eval_auc": 0.5,
-      "eval_f1": 0.4745762711864407,
-      "eval_loss": 0.32613006234169006,
-      "eval_precision": 0.9032258064516129,
-      "eval_recall": 1.0,
-      "eval_runtime": 0.6179,
-      "eval_samples_per_second": 301.013,
-      "eval_steps_per_second": 4.855,
-      "step": 30
+      "epoch": 10.0,
+      "eval_accuracy": 0.8225806451612904,
+      "eval_auc": 0.6041666666666667,
+      "eval_f1": 0.5828746177370031,
+      "eval_loss": 0.35260289907455444,
+      "eval_precision": 0.9245283018867925,
+      "eval_recall": 0.875,
+      "eval_runtime": 1.1366,
+      "eval_samples_per_second": 163.644,
+      "eval_steps_per_second": 5.279,
+      "step": 70
     },
     {
-      "epoch": 8.571428571428571,
-      "step": 30,
-      "total_flos": 1.8691703717363712e+17,
-      "train_loss": 0.2957582632700602,
-      "train_runtime": 47.4143,
-      "train_samples_per_second": 184.754,
-      "train_steps_per_second": 0.633
+      "epoch": 10.0,
+      "step": 70,
+      "total_flos": 1.1375723808188006e+18,
+      "train_loss": 0.21184425609452384,
+      "train_runtime": 168.3293,
+      "train_samples_per_second": 52.041,
+      "train_steps_per_second": 0.416
     }
   ],
   "logging_steps": 10,
-  "max_steps": 30,
+  "max_steps": 70,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 10,
   "save_steps": 500,
@@ -173,8 +214,8 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.8691703717363712e+17,
-  "train_batch_size": 64,
+  "total_flos": 1.1375723808188006e+18,
+  "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2e96781d201b4962023e2deb6df7f41bd5345729797e7384d64800c6fd5a32f6
+oid sha256:f1b26dc6bdef927a082d55156b77826a6cdd5096fa7c2241429ac0ddb18ef1bc
 size 4731