End of training

all_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 19.97,
-    "total_flos": 3.206521023309693e+18,
-    "train_loss": 0.16520703649148344,
-    "train_runtime": 938.1956,
-    "train_samples_per_second": 44.127,
-    "train_steps_per_second": 0.171
+    "epoch": 9.97,
+    "total_flos": 1.6024080844892897e+18,
+    "train_loss": 0.2579587835818529,
+    "train_runtime": 415.4218,
+    "train_samples_per_second": 49.829,
+    "train_steps_per_second": 0.193
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e242c8b30e0a510cee8ab34bc99f6200743177c577ca20e631eccb5d398ffe45
+oid sha256:18035e54e0e275c3715e720560bc3565ca6e888b15a76bf06d5479d62e779226
 size 343274861

runs/Feb02_16-46-42_teesta/events.out.tfevents.1675336619.teesta.29648.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e3d0249f4b7107ba1ed3611d2c02832dd67553f67ab06966f5edd935c317a4a3
-size 9150
+oid sha256:98d707ebc5bdb0270988ce092912f49c03d7f5566cdb523f4693fc3cffe5d4d4
+size 10115

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 19.97,
-    "total_flos": 3.206521023309693e+18,
-    "train_loss": 0.16520703649148344,
-    "train_runtime": 938.1956,
-    "train_samples_per_second": 44.127,
-    "train_steps_per_second": 0.171
+    "epoch": 9.97,
+    "total_flos": 1.6024080844892897e+18,
+    "train_loss": 0.2579587835818529,
+    "train_runtime": 415.4218,
+    "train_samples_per_second": 49.829,
+    "train_steps_per_second": 0.193
 }

trainer_state.json

@@ -1,361 +1,193 @@
 {
-  "best_metric": 0.908256880733945,
-  "best_model_checkpoint": "google-vit-base-patch16-224-cartoon-emotion-detection/checkpoint-128",
-  "epoch": 19.96969696969697,
-  "global_step": 160,
+  "best_metric": 0.8990825688073395,
+  "best_model_checkpoint": "google-vit-base-patch16-224-cartoon-emotion-detection/checkpoint-56",
+  "epoch": 9.969696969696969,
+  "global_step": 80,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.97,
-      "eval_accuracy": 0.5779816513761468,
-      "eval_f1": 0.5496044478283086,
-      "eval_loss": 1.0942180156707764,
-      "eval_precision": 0.6101578737429934,
-      "eval_recall": 0.5779816513761468,
-      "eval_runtime": 6.1545,
-      "eval_samples_per_second": 17.711,
-      "eval_steps_per_second": 0.325,
+      "eval_accuracy": 0.5596330275229358,
+      "eval_f1": 0.5360425549075453,
+      "eval_loss": 0.9901540279388428,
+      "eval_precision": 0.5505540964988554,
+      "eval_recall": 0.5596330275229358,
+      "eval_runtime": 5.4111,
+      "eval_samples_per_second": 20.144,
+      "eval_steps_per_second": 0.37,
       "step": 8
     },
     {
       "epoch": 1.24,
-      "learning_rate": 7.500000000000001e-05,
-      "loss": 1.3198,
+      "learning_rate": 0.00011666666666666667,
+      "loss": 1.242,
       "step": 10
     },
     {
       "epoch": 1.97,
-      "eval_accuracy": 0.7614678899082569,
-      "eval_f1": 0.7493394355240011,
-      "eval_loss": 0.6914369463920593,
-      "eval_precision": 0.7497958611640302,
-      "eval_recall": 0.7614678899082569,
-      "eval_runtime": 4.6109,
-      "eval_samples_per_second": 23.64,
-      "eval_steps_per_second": 0.434,
+      "eval_accuracy": 0.8165137614678899,
+      "eval_f1": 0.8131755577433994,
+      "eval_loss": 0.5157326459884644,
+      "eval_precision": 0.819457058574903,
+      "eval_recall": 0.8165137614678899,
+      "eval_runtime": 4.6435,
+      "eval_samples_per_second": 23.473,
+      "eval_steps_per_second": 0.431,
       "step": 16
     },
     {
       "epoch": 2.48,
-      "learning_rate": 0.00011666666666666667,
-      "loss": 0.6694,
+      "learning_rate": 0.0001,
+      "loss": 0.4438,
       "step": 20
     },
     {
       "epoch": 2.97,
-      "eval_accuracy": 0.7889908256880734,
-      "eval_f1": 0.7780956547838787,
-      "eval_loss": 0.47017431259155273,
-      "eval_precision": 0.7808396335677842,
-      "eval_recall": 0.7889908256880734,
-      "eval_runtime": 5.0072,
-      "eval_samples_per_second": 21.769,
-      "eval_steps_per_second": 0.399,
+      "eval_accuracy": 0.8440366972477065,
+      "eval_f1": 0.8446079913969823,
+      "eval_loss": 0.3870726525783539,
+      "eval_precision": 0.8516382699868937,
+      "eval_recall": 0.8440366972477065,
+      "eval_runtime": 4.4313,
+      "eval_samples_per_second": 24.598,
+      "eval_steps_per_second": 0.451,
       "step": 24
     },
     {
       "epoch": 3.73,
-      "learning_rate": 0.00010833333333333334,
-      "loss": 0.2725,
+      "learning_rate": 8.333333333333333e-05,
+      "loss": 0.1768,
       "step": 30
     },
     {
       "epoch": 3.97,
-      "eval_accuracy": 0.8532110091743119,
-      "eval_f1": 0.852182928635528,
-      "eval_loss": 0.3957468569278717,
-      "eval_precision": 0.8513973454558302,
-      "eval_recall": 0.8532110091743119,
-      "eval_runtime": 5.1072,
-      "eval_samples_per_second": 21.342,
-      "eval_steps_per_second": 0.392,
+      "eval_accuracy": 0.8623853211009175,
+      "eval_f1": 0.8585073146382761,
+      "eval_loss": 0.353137731552124,
+      "eval_precision": 0.865271038673443,
+      "eval_recall": 0.8623853211009175,
+      "eval_runtime": 6.7356,
+      "eval_samples_per_second": 16.183,
+      "eval_steps_per_second": 0.297,
       "step": 32
     },
     {
       "epoch": 4.97,
-      "learning_rate": 0.0001,
-      "loss": 0.1116,
+      "learning_rate": 6.666666666666667e-05,
+      "loss": 0.0661,
       "step": 40
     },
     {
       "epoch": 4.97,
       "eval_accuracy": 0.8715596330275229,
-      "eval_f1": 0.8693196656152253,
-      "eval_loss": 0.3428177833557129,
-      "eval_precision": 0.8696674261696613,
+      "eval_f1": 0.8673695820239424,
+      "eval_loss": 0.3780345022678375,
+      "eval_precision": 0.8693191511585397,
       "eval_recall": 0.8715596330275229,
-      "eval_runtime": 4.6559,
-      "eval_samples_per_second": 23.411,
-      "eval_steps_per_second": 0.43,
+      "eval_runtime": 4.3244,
+      "eval_samples_per_second": 25.206,
+      "eval_steps_per_second": 0.462,
       "step": 40
     },
     {
       "epoch": 5.97,
-      "eval_accuracy": 0.8532110091743119,
-      "eval_f1": 0.852182928635528,
-      "eval_loss": 0.38648203015327454,
-      "eval_precision": 0.8513973454558302,
-      "eval_recall": 0.8532110091743119,
-      "eval_runtime": 4.7748,
-      "eval_samples_per_second": 22.828,
-      "eval_steps_per_second": 0.419,
+      "eval_accuracy": 0.8623853211009175,
+      "eval_f1": 0.8631844663686221,
+      "eval_loss": 0.374713659286499,
+      "eval_precision": 0.864939209175348,
+      "eval_recall": 0.8623853211009175,
+      "eval_runtime": 4.479,
+      "eval_samples_per_second": 24.336,
+      "eval_steps_per_second": 0.447,
       "step": 48
     },
     {
       "epoch": 6.24,
-      "learning_rate": 9.166666666666667e-05,
-      "loss": 0.0486,
+      "learning_rate": 5e-05,
+      "loss": 0.0375,
       "step": 50
     },
     {
       "epoch": 6.97,
-      "eval_accuracy": 0.8532110091743119,
-      "eval_f1": 0.8506716755357531,
-      "eval_loss": 0.3445301055908203,
-      "eval_precision": 0.8494739768520552,
-      "eval_recall": 0.8532110091743119,
-      "eval_runtime": 4.611,
-      "eval_samples_per_second": 23.639,
-      "eval_steps_per_second": 0.434,
+      "eval_accuracy": 0.8990825688073395,
+      "eval_f1": 0.8970800861485594,
+      "eval_loss": 0.37597015500068665,
+      "eval_precision": 0.8961258037611989,
+      "eval_recall": 0.8990825688073395,
+      "eval_runtime": 4.4765,
+      "eval_samples_per_second": 24.349,
+      "eval_steps_per_second": 0.447,
       "step": 56
     },
     {
       "epoch": 7.48,
-      "learning_rate": 8.333333333333333e-05,
-      "loss": 0.0346,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 0.0362,
       "step": 60
     },
     {
       "epoch": 7.97,
-      "eval_accuracy": 0.8807339449541285,
-      "eval_f1": 0.8831061019446504,
-      "eval_loss": 0.35539454221725464,
-      "eval_precision": 0.8921137072355041,
-      "eval_recall": 0.8807339449541285,
-      "eval_runtime": 5.4806,
-      "eval_samples_per_second": 19.888,
-      "eval_steps_per_second": 0.365,
+      "eval_accuracy": 0.8715596330275229,
+      "eval_f1": 0.8680866720523425,
+      "eval_loss": 0.4092080295085907,
+      "eval_precision": 0.8684143462632116,
+      "eval_recall": 0.8715596330275229,
+      "eval_runtime": 4.3876,
+      "eval_samples_per_second": 24.843,
+      "eval_steps_per_second": 0.456,
       "step": 64
     },
     {
       "epoch": 8.73,
-      "learning_rate": 7.500000000000001e-05,
-      "loss": 0.0304,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.0322,
       "step": 70
     },
     {
       "epoch": 8.97,
-      "eval_accuracy": 0.8623853211009175,
-      "eval_f1": 0.8605019312208758,
-      "eval_loss": 0.3100130259990692,
-      "eval_precision": 0.859227121302747,
-      "eval_recall": 0.8623853211009175,
-      "eval_runtime": 4.4858,
-      "eval_samples_per_second": 24.299,
-      "eval_steps_per_second": 0.446,
+      "eval_accuracy": 0.8899082568807339,
+      "eval_f1": 0.8888099976565806,
+      "eval_loss": 0.3499366044998169,
+      "eval_precision": 0.8880198424442283,
+      "eval_recall": 0.8899082568807339,
+      "eval_runtime": 5.3381,
+      "eval_samples_per_second": 20.419,
+      "eval_steps_per_second": 0.375,
       "step": 72
     },
     {
       "epoch": 9.97,
-      "learning_rate": 6.666666666666667e-05,
-      "loss": 0.0215,
+      "learning_rate": 0.0,
+      "loss": 0.029,
       "step": 80
     },
     {
       "epoch": 9.97,
-      "eval_accuracy": 0.8715596330275229,
-      "eval_f1": 0.8706535281458829,
-      "eval_loss": 0.3717760443687439,
-      "eval_precision": 0.8700050641259672,
-      "eval_recall": 0.8715596330275229,
-      "eval_runtime": 4.4596,
-      "eval_samples_per_second": 24.442,
-      "eval_steps_per_second": 0.448,
-      "step": 80
-    },
-    {
-      "epoch": 10.97,
-      "eval_accuracy": 0.8899082568807339,
-      "eval_f1": 0.8896213803351865,
-      "eval_loss": 0.3945842683315277,
-      "eval_precision": 0.8901151962474995,
-      "eval_recall": 0.8899082568807339,
-      "eval_runtime": 4.3806,
-      "eval_samples_per_second": 24.882,
-      "eval_steps_per_second": 0.457,
-      "step": 88
-    },
-    {
-      "epoch": 11.24,
-      "learning_rate": 5.833333333333333e-05,
-      "loss": 0.0201,
-      "step": 90
-    },
-    {
-      "epoch": 11.97,
-      "eval_accuracy": 0.8532110091743119,
-      "eval_f1": 0.8523968233405174,
-      "eval_loss": 0.45045793056488037,
-      "eval_precision": 0.8557594291539246,
-      "eval_recall": 0.8532110091743119,
-      "eval_runtime": 4.5094,
-      "eval_samples_per_second": 24.172,
-      "eval_steps_per_second": 0.444,
-      "step": 96
-    },
-    {
-      "epoch": 12.48,
-      "learning_rate": 5e-05,
-      "loss": 0.02,
-      "step": 100
-    },
-    {
-      "epoch": 12.97,
-      "eval_accuracy": 0.8715596330275229,
-      "eval_f1": 0.8718071382402676,
-      "eval_loss": 0.45429742336273193,
-      "eval_precision": 0.8734266859810075,
-      "eval_recall": 0.8715596330275229,
-      "eval_runtime": 4.3136,
-      "eval_samples_per_second": 25.269,
-      "eval_steps_per_second": 0.464,
-      "step": 104
-    },
-    {
-      "epoch": 13.73,
-      "learning_rate": 4.1666666666666665e-05,
-      "loss": 0.0181,
-      "step": 110
-    },
-    {
-      "epoch": 13.97,
-      "eval_accuracy": 0.8899082568807339,
-      "eval_f1": 0.8884082118557669,
-      "eval_loss": 0.38368093967437744,
-      "eval_precision": 0.887776250789288,
-      "eval_recall": 0.8899082568807339,
-      "eval_runtime": 5.0241,
-      "eval_samples_per_second": 21.695,
-      "eval_steps_per_second": 0.398,
-      "step": 112
-    },
-    {
-      "epoch": 14.97,
-      "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.0158,
-      "step": 120
-    },
-    {
-      "epoch": 14.97,
-      "eval_accuracy": 0.8715596330275229,
-      "eval_f1": 0.8690629924501888,
-      "eval_loss": 0.39043566584587097,
-      "eval_precision": 0.8676131770089379,
-      "eval_recall": 0.8715596330275229,
-      "eval_runtime": 4.6243,
-      "eval_samples_per_second": 23.571,
-      "eval_steps_per_second": 0.432,
-      "step": 120
-    },
-    {
-      "epoch": 15.97,
-      "eval_accuracy": 0.908256880733945,
-      "eval_f1": 0.9076765848444429,
-      "eval_loss": 0.38814032077789307,
-      "eval_precision": 0.9077606207554442,
-      "eval_recall": 0.908256880733945,
-      "eval_runtime": 4.4737,
-      "eval_samples_per_second": 24.364,
-      "eval_steps_per_second": 0.447,
-      "step": 128
-    },
-    {
-      "epoch": 16.24,
-      "learning_rate": 2.5e-05,
-      "loss": 0.0147,
-      "step": 130
-    },
-    {
-      "epoch": 16.97,
       "eval_accuracy": 0.8807339449541285,
-      "eval_f1": 0.8785402173459473,
-      "eval_loss": 0.4233301281929016,
-      "eval_precision": 0.877296042811741,
+      "eval_f1": 0.8782945902988435,
+      "eval_loss": 0.37059423327445984,
+      "eval_precision": 0.8768597487153273,
       "eval_recall": 0.8807339449541285,
-      "eval_runtime": 5.5943,
-      "eval_samples_per_second": 19.484,
-      "eval_steps_per_second": 0.358,
-      "step": 136
-    },
-    {
-      "epoch": 17.48,
-      "learning_rate": 1.6666666666666667e-05,
-      "loss": 0.0138,
-      "step": 140
-    },
-    {
-      "epoch": 17.97,
-      "eval_accuracy": 0.8715596330275229,
-      "eval_f1": 0.8706535281458829,
-      "eval_loss": 0.4334617853164673,
-      "eval_precision": 0.8700050641259672,
-      "eval_recall": 0.8715596330275229,
-      "eval_runtime": 4.4249,
-      "eval_samples_per_second": 24.634,
-      "eval_steps_per_second": 0.452,
-      "step": 144
-    },
-    {
-      "epoch": 18.73,
-      "learning_rate": 8.333333333333334e-06,
-      "loss": 0.0166,
-      "step": 150
-    },
-    {
-      "epoch": 18.97,
-      "eval_accuracy": 0.8715596330275229,
-      "eval_f1": 0.8701110574601642,
-      "eval_loss": 0.44918692111968994,
-      "eval_precision": 0.8689788374764814,
-      "eval_recall": 0.8715596330275229,
-      "eval_runtime": 4.6129,
-      "eval_samples_per_second": 23.629,
-      "eval_steps_per_second": 0.434,
-      "step": 152
-    },
-    {
-      "epoch": 19.97,
-      "learning_rate": 0.0,
-      "loss": 0.016,
-      "step": 160
-    },
-    {
-      "epoch": 19.97,
-      "eval_accuracy": 0.8715596330275229,
-      "eval_f1": 0.871683140929764,
-      "eval_loss": 0.4169996976852417,
-      "eval_precision": 0.8725197999744695,
-      "eval_recall": 0.8715596330275229,
-      "eval_runtime": 4.5712,
-      "eval_samples_per_second": 23.845,
-      "eval_steps_per_second": 0.438,
-      "step": 160
+      "eval_runtime": 4.4763,
+      "eval_samples_per_second": 24.35,
+      "eval_steps_per_second": 0.447,
+      "step": 80
     },
     {
-      "epoch": 19.97,
-      "step": 160,
-      "total_flos": 3.206521023309693e+18,
-      "train_loss": 0.16520703649148344,
-      "train_runtime": 938.1956,
-      "train_samples_per_second": 44.127,
-      "train_steps_per_second": 0.171
+      "epoch": 9.97,
+      "step": 80,
+      "total_flos": 1.6024080844892897e+18,
+      "train_loss": 0.2579587835818529,
+      "train_runtime": 415.4218,
+      "train_samples_per_second": 49.829,
+      "train_steps_per_second": 0.193
     }
   ],
-  "max_steps": 160,
-  "num_train_epochs": 20,
-  "total_flos": 3.206521023309693e+18,
+  "max_steps": 80,
+  "num_train_epochs": 10,
+  "total_flos": 1.6024080844892897e+18,
   "trial_name": null,
   "trial_params": null
 }