End of training

all_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 9.89,
-    "total_flos": 1.3244442196920238e+18,
-    "train_loss": 0.9847671677668889,
-    "train_runtime": 282.6595,
-    "train_samples_per_second": 61.134,
-    "train_steps_per_second": 0.212
+    "epoch": 19.89,
+    "total_flos": 2.6637698351109243e+18,
+    "train_loss": 0.012146842836712798,
+    "train_runtime": 613.8258,
+    "train_samples_per_second": 56.303,
+    "train_steps_per_second": 0.195
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:034114a0efe89d79a82c07eb97fde02ef1dfd83832450c7003a93a1da84dfaf7
+oid sha256:74cb5e8f5d78070c153615d7560f7c2a8e140588589df74270cb3f3e6b92973e
 size 343334641

runs/Jan18_23-37-50_teesta/events.out.tfevents.1674065278.teesta.15918.2

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7f4b076c59870f2500a2565d5f933453ba8b162363cbc5ab6f96f9ad64e25867
-size 5541
+oid sha256:d07c2b0b551aff8ab1e371ef374baa7d4db26cda47fcaa404328a2e93be2f7ef
+size 16534

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 9.89,
-    "total_flos": 1.3244442196920238e+18,
-    "train_loss": 0.9847671677668889,
-    "train_runtime": 282.6595,
-    "train_samples_per_second": 61.134,
-    "train_steps_per_second": 0.212
+    "epoch": 19.89,
+    "total_flos": 2.6637698351109243e+18,
+    "train_loss": 0.012146842836712798,
+    "train_runtime": 613.8258,
+    "train_samples_per_second": 56.303,
+    "train_steps_per_second": 0.195
 }

trainer_state.json

@@ -1,181 +1,337 @@
 {
-  "best_metric": 0.875,
-  "best_model_checkpoint": "google-vit-base-patch16-224-cartoon-face-recognition/checkpoint-48",
-  "epoch": 9.88888888888889,
-  "global_step": 60,
+  "best_metric": 0.9050925925925926,
+  "best_model_checkpoint": "google-vit-base-patch16-224-cartoon-face-recognition/checkpoint-84",
+  "epoch": 19.88888888888889,
+  "global_step": 120,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.89,
-      "eval_accuracy": 0.24074074074074073,
-      "eval_f1": 0.1643869480887188,
-      "eval_loss": 2.792280435562134,
-      "eval_precision": 0.20461543357832349,
-      "eval_recall": 0.24074074074074073,
-      "eval_runtime": 7.7909,
-      "eval_samples_per_second": 55.449,
-      "eval_steps_per_second": 0.898,
+      "eval_accuracy": 0.8611111111111112,
+      "eval_f1": 0.8577219426933996,
+      "eval_loss": 0.5458505749702454,
+      "eval_precision": 0.8682746241112194,
+      "eval_recall": 0.8611111111111112,
+      "eval_runtime": 7.819,
+      "eval_samples_per_second": 55.25,
+      "eval_steps_per_second": 0.895,
       "step": 6
     },
     {
       "epoch": 1.59,
-      "learning_rate": 0.00011111111111111112,
-      "loss": 3.0606,
+      "learning_rate": 0.0001,
+      "loss": 0.0812,
       "step": 10
     },
     {
       "epoch": 1.89,
-      "eval_accuracy": 0.5,
-      "eval_f1": 0.4341708558476282,
-      "eval_loss": 2.0185298919677734,
-      "eval_precision": 0.50652873651949,
-      "eval_recall": 0.5,
-      "eval_runtime": 6.5876,
-      "eval_samples_per_second": 65.578,
-      "eval_steps_per_second": 1.063,
+      "eval_accuracy": 0.8796296296296297,
+      "eval_f1": 0.8763585304815898,
+      "eval_loss": 0.47026392817497253,
+      "eval_precision": 0.8832897696668488,
+      "eval_recall": 0.8796296296296297,
+      "eval_runtime": 6.6723,
+      "eval_samples_per_second": 64.746,
+      "eval_steps_per_second": 1.049,
       "step": 12
     },
     {
       "epoch": 2.89,
-      "eval_accuracy": 0.6620370370370371,
-      "eval_f1": 0.6279977278642492,
-      "eval_loss": 1.4164462089538574,
-      "eval_precision": 0.6524558387748216,
-      "eval_recall": 0.6620370370370371,
-      "eval_runtime": 6.5774,
-      "eval_samples_per_second": 65.679,
-      "eval_steps_per_second": 1.064,
+      "eval_accuracy": 0.8935185185185185,
+      "eval_f1": 0.890600542893898,
+      "eval_loss": 0.44299086928367615,
+      "eval_precision": 0.896940900151719,
+      "eval_recall": 0.8935185185185185,
+      "eval_runtime": 7.4266,
+      "eval_samples_per_second": 58.169,
+      "eval_steps_per_second": 0.943,
       "step": 18
     },
     {
       "epoch": 3.3,
-      "learning_rate": 8.888888888888888e-05,
-      "loss": 1.688,
+      "learning_rate": 0.00011111111111111112,
+      "loss": 0.0307,
       "step": 20
     },
     {
       "epoch": 3.89,
-      "eval_accuracy": 0.7685185185185185,
-      "eval_f1": 0.7520520641207192,
-      "eval_loss": 1.0344423055648804,
-      "eval_precision": 0.7687545261419606,
-      "eval_recall": 0.7685185185185185,
-      "eval_runtime": 6.6855,
-      "eval_samples_per_second": 64.618,
-      "eval_steps_per_second": 1.047,
+      "eval_accuracy": 0.8819444444444444,
+      "eval_f1": 0.876747873541574,
+      "eval_loss": 0.40446680784225464,
+      "eval_precision": 0.8848741291373776,
+      "eval_recall": 0.8819444444444444,
+      "eval_runtime": 6.4316,
+      "eval_samples_per_second": 67.168,
+      "eval_steps_per_second": 1.088,
       "step": 24
     },
     {
       "epoch": 4.89,
-      "learning_rate": 6.666666666666667e-05,
-      "loss": 0.6694,
+      "learning_rate": 0.0001,
+      "loss": 0.0091,
       "step": 30
     },
     {
       "epoch": 4.89,
-      "eval_accuracy": 0.8310185185185185,
-      "eval_f1": 0.8204789843927855,
-      "eval_loss": 0.8109782338142395,
-      "eval_precision": 0.8419593068227221,
-      "eval_recall": 0.8310185185185185,
-      "eval_runtime": 6.4692,
-      "eval_samples_per_second": 66.778,
-      "eval_steps_per_second": 1.082,
+      "eval_accuracy": 0.9004629629629629,
+      "eval_f1": 0.8979540324742182,
+      "eval_loss": 0.36718857288360596,
+      "eval_precision": 0.9025419548328308,
+      "eval_recall": 0.9004629629629629,
+      "eval_runtime": 6.6517,
+      "eval_samples_per_second": 64.946,
+      "eval_steps_per_second": 1.052,
       "step": 30
     },
     {
       "epoch": 5.89,
-      "eval_accuracy": 0.8495370370370371,
-      "eval_f1": 0.8431889196512385,
-      "eval_loss": 0.680189847946167,
-      "eval_precision": 0.8577619498091635,
-      "eval_recall": 0.8495370370370371,
-      "eval_runtime": 6.5265,
-      "eval_samples_per_second": 66.192,
-      "eval_steps_per_second": 1.073,
+      "eval_accuracy": 0.9027777777777778,
+      "eval_f1": 0.9010548586749486,
+      "eval_loss": 0.3840581178665161,
+      "eval_precision": 0.9124659232717037,
+      "eval_recall": 0.9027777777777778,
+      "eval_runtime": 6.6486,
+      "eval_samples_per_second": 64.976,
+      "eval_steps_per_second": 1.053,
       "step": 36
     },
     {
       "epoch": 6.59,
-      "learning_rate": 4.444444444444444e-05,
-      "loss": 0.285,
+      "learning_rate": 8.888888888888888e-05,
+      "loss": 0.0043,
       "step": 40
     },
     {
       "epoch": 6.89,
-      "eval_accuracy": 0.8495370370370371,
-      "eval_f1": 0.8436215206983917,
-      "eval_loss": 0.6132100224494934,
-      "eval_precision": 0.8540014281192293,
-      "eval_recall": 0.8495370370370371,
-      "eval_runtime": 6.6157,
-      "eval_samples_per_second": 65.299,
-      "eval_steps_per_second": 1.058,
+      "eval_accuracy": 0.9004629629629629,
+      "eval_f1": 0.8971983578645838,
+      "eval_loss": 0.39255601167678833,
+      "eval_precision": 0.9073430487686229,
+      "eval_recall": 0.9004629629629629,
+      "eval_runtime": 6.6126,
+      "eval_samples_per_second": 65.329,
+      "eval_steps_per_second": 1.059,
       "step": 42
     },
     {
       "epoch": 7.89,
-      "eval_accuracy": 0.875,
-      "eval_f1": 0.8707588213347532,
-      "eval_loss": 0.5606003403663635,
-      "eval_precision": 0.8799090872795438,
-      "eval_recall": 0.875,
-      "eval_runtime": 6.6491,
-      "eval_samples_per_second": 64.971,
-      "eval_steps_per_second": 1.053,
+      "eval_accuracy": 0.8958333333333334,
+      "eval_f1": 0.8931276138165111,
+      "eval_loss": 0.3785696029663086,
+      "eval_precision": 0.9004579824364817,
+      "eval_recall": 0.8958333333333334,
+      "eval_runtime": 6.493,
+      "eval_samples_per_second": 66.533,
+      "eval_steps_per_second": 1.078,
       "step": 48
     },
     {
       "epoch": 8.3,
-      "learning_rate": 2.222222222222222e-05,
-      "loss": 0.1277,
+      "learning_rate": 7.777777777777778e-05,
+      "loss": 0.0031,
       "step": 50
     },
     {
       "epoch": 8.89,
-      "eval_accuracy": 0.8680555555555556,
-      "eval_f1": 0.864629496735875,
-      "eval_loss": 0.5443260669708252,
-      "eval_precision": 0.8753500264010703,
-      "eval_recall": 0.8680555555555556,
-      "eval_runtime": 6.7428,
-      "eval_samples_per_second": 64.068,
-      "eval_steps_per_second": 1.038,
+      "eval_accuracy": 0.9027777777777778,
+      "eval_f1": 0.9007047995444771,
+      "eval_loss": 0.3791252374649048,
+      "eval_precision": 0.9090957347391172,
+      "eval_recall": 0.9027777777777778,
+      "eval_runtime": 6.5347,
+      "eval_samples_per_second": 66.109,
+      "eval_steps_per_second": 1.071,
       "step": 54
     },
     {
       "epoch": 9.89,
-      "learning_rate": 0.0,
-      "loss": 0.0779,
+      "learning_rate": 6.666666666666667e-05,
+      "loss": 0.002,
       "step": 60
     },
     {
       "epoch": 9.89,
-      "eval_accuracy": 0.8611111111111112,
-      "eval_f1": 0.8553095447735699,
-      "eval_loss": 0.5382117033004761,
-      "eval_precision": 0.8637558340439363,
-      "eval_recall": 0.8611111111111112,
-      "eval_runtime": 6.6834,
-      "eval_samples_per_second": 64.638,
-      "eval_steps_per_second": 1.047,
+      "eval_accuracy": 0.9027777777777778,
+      "eval_f1": 0.9000566322189608,
+      "eval_loss": 0.3677087724208832,
+      "eval_precision": 0.9105962655404846,
+      "eval_recall": 0.9027777777777778,
+      "eval_runtime": 6.4291,
+      "eval_samples_per_second": 67.194,
+      "eval_steps_per_second": 1.089,
       "step": 60
     },
     {
-      "epoch": 9.89,
-      "step": 60,
-      "total_flos": 1.3244442196920238e+18,
-      "train_loss": 0.9847671677668889,
-      "train_runtime": 282.6595,
-      "train_samples_per_second": 61.134,
-      "train_steps_per_second": 0.212
+      "epoch": 10.89,
+      "eval_accuracy": 0.9027777777777778,
+      "eval_f1": 0.9007267676235234,
+      "eval_loss": 0.37403494119644165,
+      "eval_precision": 0.9098558471011668,
+      "eval_recall": 0.9027777777777778,
+      "eval_runtime": 7.5533,
+      "eval_samples_per_second": 57.193,
+      "eval_steps_per_second": 0.927,
+      "step": 66
+    },
+    {
+      "epoch": 11.59,
+      "learning_rate": 5.555555555555556e-05,
+      "loss": 0.0027,
+      "step": 70
+    },
+    {
+      "epoch": 11.89,
+      "eval_accuracy": 0.8981481481481481,
+      "eval_f1": 0.8955661460989992,
+      "eval_loss": 0.3868783414363861,
+      "eval_precision": 0.904346797052174,
+      "eval_recall": 0.8981481481481481,
+      "eval_runtime": 6.6537,
+      "eval_samples_per_second": 64.926,
+      "eval_steps_per_second": 1.052,
+      "step": 72
+    },
+    {
+      "epoch": 12.89,
+      "eval_accuracy": 0.8981481481481481,
+      "eval_f1": 0.8953859158931289,
+      "eval_loss": 0.38008958101272583,
+      "eval_precision": 0.9021237317078286,
+      "eval_recall": 0.8981481481481481,
+      "eval_runtime": 6.4045,
+      "eval_samples_per_second": 67.453,
+      "eval_steps_per_second": 1.093,
+      "step": 78
+    },
+    {
+      "epoch": 13.3,
+      "learning_rate": 4.444444444444444e-05,
+      "loss": 0.004,
+      "step": 80
+    },
+    {
+      "epoch": 13.89,
+      "eval_accuracy": 0.9050925925925926,
+      "eval_f1": 0.9027775427273881,
+      "eval_loss": 0.3674483001232147,
+      "eval_precision": 0.9112982137223617,
+      "eval_recall": 0.9050925925925926,
+      "eval_runtime": 6.5221,
+      "eval_samples_per_second": 66.237,
+      "eval_steps_per_second": 1.073,
+      "step": 84
+    },
+    {
+      "epoch": 14.89,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 0.0024,
+      "step": 90
+    },
+    {
+      "epoch": 14.89,
+      "eval_accuracy": 0.9050925925925926,
+      "eval_f1": 0.9027035229916751,
+      "eval_loss": 0.36199745535850525,
+      "eval_precision": 0.9096292505458038,
+      "eval_recall": 0.9050925925925926,
+      "eval_runtime": 6.6297,
+      "eval_samples_per_second": 65.161,
+      "eval_steps_per_second": 1.056,
+      "step": 90
+    },
+    {
+      "epoch": 15.89,
+      "eval_accuracy": 0.9027777777777778,
+      "eval_f1": 0.9005670379013408,
+      "eval_loss": 0.36699026823043823,
+      "eval_precision": 0.9088959728872165,
+      "eval_recall": 0.9027777777777778,
+      "eval_runtime": 7.1759,
+      "eval_samples_per_second": 60.202,
+      "eval_steps_per_second": 0.975,
+      "step": 96
+    },
+    {
+      "epoch": 16.59,
+      "learning_rate": 2.222222222222222e-05,
+      "loss": 0.0021,
+      "step": 100
+    },
+    {
+      "epoch": 16.89,
+      "eval_accuracy": 0.9004629629629629,
+      "eval_f1": 0.8980450039465582,
+      "eval_loss": 0.38269728422164917,
+      "eval_precision": 0.9064532394818394,
+      "eval_recall": 0.9004629629629629,
+      "eval_runtime": 8.0097,
+      "eval_samples_per_second": 53.935,
+      "eval_steps_per_second": 0.874,
+      "step": 102
+    },
+    {
+      "epoch": 17.89,
+      "eval_accuracy": 0.8981481481481481,
+      "eval_f1": 0.8957872163681325,
+      "eval_loss": 0.37475430965423584,
+      "eval_precision": 0.9049223762152907,
+      "eval_recall": 0.8981481481481481,
+      "eval_runtime": 7.8764,
+      "eval_samples_per_second": 54.848,
+      "eval_steps_per_second": 0.889,
+      "step": 108
+    },
+    {
+      "epoch": 18.3,
+      "learning_rate": 1.111111111111111e-05,
+      "loss": 0.0022,
+      "step": 110
+    },
+    {
+      "epoch": 18.89,
+      "eval_accuracy": 0.9027777777777778,
+      "eval_f1": 0.9005599048903711,
+      "eval_loss": 0.3825390636920929,
+      "eval_precision": 0.9100892619763822,
+      "eval_recall": 0.9027777777777778,
+      "eval_runtime": 6.4897,
+      "eval_samples_per_second": 66.567,
+      "eval_steps_per_second": 1.079,
+      "step": 114
+    },
+    {
+      "epoch": 19.89,
+      "learning_rate": 0.0,
+      "loss": 0.0019,
+      "step": 120
+    },
+    {
+      "epoch": 19.89,
+      "eval_accuracy": 0.9004629629629629,
+      "eval_f1": 0.8984296743444529,
+      "eval_loss": 0.370717316865921,
+      "eval_precision": 0.9066341895316832,
+      "eval_recall": 0.9004629629629629,
+      "eval_runtime": 6.4737,
+      "eval_samples_per_second": 66.732,
+      "eval_steps_per_second": 1.081,
+      "step": 120
+    },
+    {
+      "epoch": 19.89,
+      "step": 120,
+      "total_flos": 2.6637698351109243e+18,
+      "train_loss": 0.012146842836712798,
+      "train_runtime": 613.8258,
+      "train_samples_per_second": 56.303,
+      "train_steps_per_second": 0.195
     }
   ],
-  "max_steps": 60,
-  "num_train_epochs": 10,
-  "total_flos": 1.3244442196920238e+18,
+  "max_steps": 120,
+  "num_train_epochs": 20,
+  "total_flos": 2.6637698351109243e+18,
   "trial_name": null,
   "trial_params": null
 }