🍻 cheers

README.md

@@ -2,6 +2,7 @@
 license: apache-2.0
 base_model: google/vit-base-patch16-224
 tags:
+- image-classification
 - generated_from_trainer
 metrics:
 - accuracy
@@ -18,13 +19,13 @@ should probably proofread and complete it, then remove this comment. -->
 
 # vit-lr-0.0001
 
-This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on an unknown dataset.
+This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the skin-cancer dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0203
-- Accuracy: 0.8457
-- Precision: 0.8488
-- Recall: 0.8457
-- F1: 0.8416
+- Loss: 0.4778
+- Accuracy: 0.8311
+- Precision: 0.8433
+- Recall: 0.8311
+- F1: 0.8269
 
 ## Model description
 

all_results.json

@@ -1,16 +1,16 @@
 {
     "epoch": 12.0,
-    "eval_accuracy": 0.8283633841886269,
-    "eval_f1": 0.8213379813460607,
-    "eval_loss": 0.4716978371143341,
-    "eval_precision": 0.8230999577196753,
-    "eval_recall": 0.8283633841886269,
-    "eval_runtime": 38.426,
-    "eval_samples_per_second": 75.053,
-    "eval_steps_per_second": 9.395,
+    "eval_accuracy": 0.8311373092926491,
+    "eval_f1": 0.8268544179848872,
+    "eval_loss": 0.4778192639350891,
+    "eval_precision": 0.8433264041291242,
+    "eval_recall": 0.8311373092926491,
+    "eval_runtime": 38.5646,
+    "eval_samples_per_second": 74.784,
+    "eval_steps_per_second": 9.361,
     "total_flos": 4.768760767819088e+18,
-    "train_loss": 0.16453982563157443,
-    "train_runtime": 1738.7619,
-    "train_samples_per_second": 294.922,
-    "train_steps_per_second": 18.461
+    "train_loss": 0.15186076372211968,
+    "train_runtime": 1747.1919,
+    "train_samples_per_second": 293.5,
+    "train_steps_per_second": 18.372
 }

eval_results.json

@@ -1,11 +1,11 @@
 {
     "epoch": 12.0,
-    "eval_accuracy": 0.8283633841886269,
-    "eval_f1": 0.8213379813460607,
-    "eval_loss": 0.4716978371143341,
-    "eval_precision": 0.8230999577196753,
-    "eval_recall": 0.8283633841886269,
-    "eval_runtime": 38.426,
-    "eval_samples_per_second": 75.053,
-    "eval_steps_per_second": 9.395
+    "eval_accuracy": 0.8311373092926491,
+    "eval_f1": 0.8268544179848872,
+    "eval_loss": 0.4778192639350891,
+    "eval_precision": 0.8433264041291242,
+    "eval_recall": 0.8311373092926491,
+    "eval_runtime": 38.5646,
+    "eval_samples_per_second": 74.784,
+    "eval_steps_per_second": 9.361
 }

runs/Mar19_13-16-22_befed1cbff90/events.out.tfevents.1710855985.befed1cbff90.172.6

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a48d22f2a847c500cf946bfe1a8808b24bb9afcf39f36517c9ea66f7a5fda93
+size 560

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 12.0,
     "total_flos": 4.768760767819088e+18,
-    "train_loss": 0.16453982563157443,
-    "train_runtime": 1738.7619,
-    "train_samples_per_second": 294.922,
-    "train_steps_per_second": 18.461
+    "train_loss": 0.15186076372211968,
+    "train_runtime": 1747.1919,
+    "train_samples_per_second": 293.5,
+    "train_steps_per_second": 18.372
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.4716978371143341,
+  "best_metric": 0.4778192639350891,
   "best_model_checkpoint": "./vit-lr-0.0001/checkpoint-642",
   "epoch": 12.0,
   "eval_steps": 500,
@@ -10,240 +10,240 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 5.450945854187012,
-      "learning_rate": 9.549150281252611e-06,
-      "loss": 0.682,
+      "grad_norm": 5.167571067810059,
+      "learning_rate": 9.997548148581541e-05,
+      "loss": 0.6718,
       "step": 321
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.8099861303744799,
-      "eval_f1": 0.798120450266025,
-      "eval_loss": 0.5192127823829651,
-      "eval_precision": 0.8034262669589041,
-      "eval_recall": 0.8099861303744799,
-      "eval_runtime": 37.3933,
-      "eval_samples_per_second": 77.126,
-      "eval_steps_per_second": 9.654,
+      "eval_accuracy": 0.8242024965325936,
+      "eval_f1": 0.8162630829948523,
+      "eval_loss": 0.4934074282646179,
+      "eval_precision": 0.8254417232097437,
+      "eval_recall": 0.8242024965325936,
+      "eval_runtime": 37.6292,
+      "eval_samples_per_second": 76.643,
+      "eval_steps_per_second": 9.594,
       "step": 321
     },
     {
       "epoch": 2.0,
-      "grad_norm": 5.513205051422119,
-      "learning_rate": 6.394955530196143e-05,
-      "loss": 0.4386,
+      "grad_norm": 6.953319072723389,
+      "learning_rate": 9.990164344447551e-05,
+      "loss": 0.4045,
       "step": 642
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8283633841886269,
-      "eval_f1": 0.8213379813460607,
-      "eval_loss": 0.4716978371143341,
-      "eval_precision": 0.8230999577196753,
-      "eval_recall": 0.8283633841886269,
-      "eval_runtime": 37.886,
-      "eval_samples_per_second": 76.123,
-      "eval_steps_per_second": 9.529,
+      "eval_accuracy": 0.8311373092926491,
+      "eval_f1": 0.8268544179848872,
+      "eval_loss": 0.4778192639350891,
+      "eval_precision": 0.8433264041291242,
+      "eval_recall": 0.8311373092926491,
+      "eval_runtime": 37.5244,
+      "eval_samples_per_second": 76.857,
+      "eval_steps_per_second": 9.62,
       "step": 642
     },
     {
       "epoch": 3.0,
-      "grad_norm": 0.9458593726158142,
-      "learning_rate": 6.840622763423387e-05,
-      "loss": 0.2621,
+      "grad_norm": 7.354413986206055,
+      "learning_rate": 9.977855850575433e-05,
+      "loss": 0.2419,
       "step": 963
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8349514563106796,
-      "eval_f1": 0.8201100816778963,
-      "eval_loss": 0.5676096081733704,
-      "eval_precision": 0.8262616496556073,
-      "eval_recall": 0.8349514563106796,
-      "eval_runtime": 37.2214,
-      "eval_samples_per_second": 77.482,
-      "eval_steps_per_second": 9.699,
+      "eval_accuracy": 0.840499306518724,
+      "eval_f1": 0.8409562025706393,
+      "eval_loss": 0.5132780075073242,
+      "eval_precision": 0.8524873378932092,
+      "eval_recall": 0.840499306518724,
+      "eval_runtime": 36.8758,
+      "eval_samples_per_second": 78.209,
+      "eval_steps_per_second": 9.79,
       "step": 963
     },
     {
       "epoch": 4.0,
-      "grad_norm": 7.709182262420654,
-      "learning_rate": 6.9628986498044885e-06,
-      "loss": 0.1891,
+      "grad_norm": 0.17630255222320557,
+      "learning_rate": 9.960634813962617e-05,
+      "loss": 0.1267,
       "step": 1284
     },
     {
       "epoch": 4.0,
       "eval_accuracy": 0.8495145631067961,
-      "eval_f1": 0.8436304900485414,
-      "eval_loss": 0.5022268891334534,
-      "eval_precision": 0.8555501567834226,
+      "eval_f1": 0.8447806876150558,
+      "eval_loss": 0.6153807044029236,
+      "eval_precision": 0.8490665425851771,
       "eval_recall": 0.8495145631067961,
-      "eval_runtime": 37.4592,
-      "eval_samples_per_second": 76.99,
-      "eval_steps_per_second": 9.637,
+      "eval_runtime": 37.1161,
+      "eval_samples_per_second": 77.702,
+      "eval_steps_per_second": 9.726,
       "step": 1284
     },
     {
       "epoch": 5.0,
-      "grad_norm": 0.030552733689546585,
-      "learning_rate": 9.960573506570156e-05,
-      "loss": 0.1052,
+      "grad_norm": 0.5389394760131836,
+      "learning_rate": 9.938518229693118e-05,
+      "loss": 0.0733,
       "step": 1605
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.8398058252427184,
-      "eval_f1": 0.8357173097291434,
-      "eval_loss": 0.6660671830177307,
-      "eval_precision": 0.8482768361910064,
-      "eval_recall": 0.8398058252427184,
-      "eval_runtime": 39.2791,
-      "eval_samples_per_second": 73.423,
-      "eval_steps_per_second": 9.191,
+      "eval_accuracy": 0.8422330097087378,
+      "eval_f1": 0.8361023278787632,
+      "eval_loss": 0.7844527959823608,
+      "eval_precision": 0.8420613182520028,
+      "eval_recall": 0.8422330097087378,
+      "eval_runtime": 37.3092,
+      "eval_samples_per_second": 77.3,
+      "eval_steps_per_second": 9.676,
       "step": 1605
     },
     {
       "epoch": 6.0,
-      "grad_norm": 1.1932728290557861,
-      "learning_rate": 1.4644660940675861e-05,
-      "loss": 0.0785,
+      "grad_norm": 24.989282608032227,
+      "learning_rate": 9.911527924165445e-05,
+      "loss": 0.0446,
       "step": 1926
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.8592233009708737,
-      "eval_f1": 0.8574256259898136,
-      "eval_loss": 0.5570405125617981,
-      "eval_precision": 0.8573736727029027,
-      "eval_recall": 0.8592233009708737,
-      "eval_runtime": 37.1007,
-      "eval_samples_per_second": 77.734,
-      "eval_steps_per_second": 9.73,
+      "eval_accuracy": 0.8470873786407767,
+      "eval_f1": 0.840751584936853,
+      "eval_loss": 0.8766722679138184,
+      "eval_precision": 0.8407202080357687,
+      "eval_recall": 0.8470873786407767,
+      "eval_runtime": 37.9466,
+      "eval_samples_per_second": 76.002,
+      "eval_steps_per_second": 9.513,
       "step": 1926
     },
     {
       "epoch": 7.0,
-      "grad_norm": 17.325477600097656,
-      "learning_rate": 5.626666167820289e-05,
-      "loss": 0.0481,
+      "grad_norm": 0.16091497242450714,
+      "learning_rate": 9.879690533552573e-05,
+      "loss": 0.0523,
       "step": 2247
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.8547156726768377,
-      "eval_f1": 0.8523115263952025,
-      "eval_loss": 0.6496189832687378,
-      "eval_precision": 0.8547876753873573,
-      "eval_recall": 0.8547156726768377,
-      "eval_runtime": 36.7758,
-      "eval_samples_per_second": 78.421,
-      "eval_steps_per_second": 9.816,
+      "eval_accuracy": 0.84500693481276,
+      "eval_f1": 0.8406651419661996,
+      "eval_loss": 0.8674203753471375,
+      "eval_precision": 0.8491745633887275,
+      "eval_recall": 0.84500693481276,
+      "eval_runtime": 37.6505,
+      "eval_samples_per_second": 76.599,
+      "eval_steps_per_second": 9.588,
       "step": 2247
     },
     {
       "epoch": 8.0,
-      "grad_norm": 0.6180940270423889,
-      "learning_rate": 7.545207078756922e-05,
-      "loss": 0.0281,
+      "grad_norm": 0.0028496491722762585,
+      "learning_rate": 9.843159102999166e-05,
+      "loss": 0.0388,
       "step": 2568
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.8571428571428571,
-      "eval_f1": 0.8544924977438809,
-      "eval_loss": 0.7423234581947327,
-      "eval_precision": 0.8569592454687236,
-      "eval_recall": 0.8571428571428571,
-      "eval_runtime": 37.383,
-      "eval_samples_per_second": 77.147,
-      "eval_steps_per_second": 9.657,
+      "eval_accuracy": 0.8398058252427184,
+      "eval_f1": 0.8387317635385871,
+      "eval_loss": 0.9753792881965637,
+      "eval_precision": 0.8565844319804308,
+      "eval_recall": 0.8398058252427184,
+      "eval_runtime": 37.8044,
+      "eval_samples_per_second": 76.287,
+      "eval_steps_per_second": 9.549,
       "step": 2568
     },
     {
       "epoch": 9.0,
-      "grad_norm": 0.014829314313828945,
-      "learning_rate": 3.5111757055883184e-06,
-      "loss": 0.0439,
+      "grad_norm": 0.0019187598954886198,
+      "learning_rate": 9.801741382013225e-05,
+      "loss": 0.0402,
       "step": 2889
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.8540221914008321,
-      "eval_f1": 0.8518021160859013,
-      "eval_loss": 0.7677786946296692,
-      "eval_precision": 0.851139101545856,
-      "eval_recall": 0.8540221914008321,
-      "eval_runtime": 37.2162,
-      "eval_samples_per_second": 77.493,
-      "eval_steps_per_second": 9.7,
+      "eval_accuracy": 0.8491678224687933,
+      "eval_f1": 0.8461066193854649,
+      "eval_loss": 0.9369620084762573,
+      "eval_precision": 0.8547133286785668,
+      "eval_recall": 0.8491678224687933,
+      "eval_runtime": 37.5745,
+      "eval_samples_per_second": 76.754,
+      "eval_steps_per_second": 9.608,
       "step": 2889
     },
     {
       "epoch": 10.0,
-      "grad_norm": 0.003476586891338229,
-      "learning_rate": 9.801468428386933e-05,
-      "loss": 0.0297,
+      "grad_norm": 0.00408038217574358,
+      "learning_rate": 9.755584922004499e-05,
+      "loss": 0.0283,
       "step": 3210
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.8467406380027739,
-      "eval_f1": 0.8464070946689463,
-      "eval_loss": 0.8647756576538086,
-      "eval_precision": 0.8478472152581428,
-      "eval_recall": 0.8467406380027739,
-      "eval_runtime": 37.1575,
-      "eval_samples_per_second": 77.616,
-      "eval_steps_per_second": 9.715,
+      "eval_accuracy": 0.8509015256588072,
+      "eval_f1": 0.8482666902922076,
+      "eval_loss": 0.9217829704284668,
+      "eval_precision": 0.8495520865912046,
+      "eval_recall": 0.8509015256588072,
+      "eval_runtime": 37.6867,
+      "eval_samples_per_second": 76.526,
+      "eval_steps_per_second": 9.579,
       "step": 3210
     },
     {
       "epoch": 11.0,
-      "grad_norm": 0.09397422522306442,
-      "learning_rate": 2.0610737385380886e-05,
-      "loss": 0.037,
+      "grad_norm": 2.3583085536956787,
+      "learning_rate": 9.70473527382648e-05,
+      "loss": 0.0451,
       "step": 3531
     },
     {
       "epoch": 11.0,
-      "eval_accuracy": 0.8561026352288488,
-      "eval_f1": 0.8548878398804619,
-      "eval_loss": 0.7682856321334839,
-      "eval_precision": 0.8552181984518148,
-      "eval_recall": 0.8561026352288488,
-      "eval_runtime": 37.5045,
-      "eval_samples_per_second": 76.898,
-      "eval_steps_per_second": 9.626,
+      "eval_accuracy": 0.8474341192787794,
+      "eval_f1": 0.8401062715209746,
+      "eval_loss": 0.987198531627655,
+      "eval_precision": 0.8400167652295873,
+      "eval_recall": 0.8474341192787794,
+      "eval_runtime": 37.5611,
+      "eval_samples_per_second": 76.782,
+      "eval_steps_per_second": 9.611,
       "step": 3531
     },
     {
       "epoch": 12.0,
-      "grad_norm": 0.002134101465344429,
-      "learning_rate": 4.999999999998897e-05,
-      "loss": 0.0322,
+      "grad_norm": 11.217573165893555,
+      "learning_rate": 9.649242619942769e-05,
+      "loss": 0.0549,
       "step": 3852
     },
     {
       "epoch": 12.0,
-      "eval_accuracy": 0.8512482662968099,
-      "eval_f1": 0.8396847788052201,
-      "eval_loss": 1.0125652551651,
-      "eval_precision": 0.8561110421693547,
-      "eval_recall": 0.8512482662968099,
-      "eval_runtime": 37.3877,
-      "eval_samples_per_second": 77.138,
-      "eval_steps_per_second": 9.656,
+      "eval_accuracy": 0.8457004160887656,
+      "eval_f1": 0.8415779150905001,
+      "eval_loss": 1.0202850103378296,
+      "eval_precision": 0.8487681253829549,
+      "eval_recall": 0.8457004160887656,
+      "eval_runtime": 37.4542,
+      "eval_samples_per_second": 77.001,
+      "eval_steps_per_second": 9.638,
       "step": 3852
     },
     {
       "epoch": 12.0,
       "step": 3852,
       "total_flos": 4.768760767819088e+18,
-      "train_loss": 0.16453982563157443,
-      "train_runtime": 1738.7619,
-      "train_samples_per_second": 294.922,
-      "train_steps_per_second": 18.461
+      "train_loss": 0.15186076372211968,
+      "train_runtime": 1747.1919,
+      "train_samples_per_second": 293.5,
+      "train_steps_per_second": 18.372
     }
   ],
   "logging_steps": 500,