End of training

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 9.1,
-    "eval_accuracy": 0.065,
-    "eval_loss": 8.357227325439453,
-    "eval_runtime": 45.341,
-    "eval_samples_per_second": 4.411,
-    "eval_steps_per_second": 2.206
+    "eval_accuracy": 0.0,
+    "eval_loss": 8.188952445983887,
+    "eval_runtime": 1.8553,
+    "eval_samples_per_second": 4.312,
+    "eval_steps_per_second": 2.156
 }

runs/Jun28_09-15-43_d7d6e6a71cfc/events.out.tfevents.1687946329.d7d6e6a71cfc.28.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:323dc234232021dc460a19c80045cc381d6c5f5a9973896c61f45878336236d5
-size 411
+oid sha256:4cd3cd6364c6ed1920e7374bc0c1813d79bd0694e7a123be68d41a3227e27e6a
+size 734

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 9.1,
-    "eval_accuracy": 0.065,
-    "eval_loss": 8.357227325439453,
-    "eval_runtime": 45.341,
-    "eval_samples_per_second": 4.411,
-    "eval_steps_per_second": 2.206
+    "eval_accuracy": 0.0,
+    "eval_loss": 8.188952445983887,
+    "eval_runtime": 1.8553,
+    "eval_samples_per_second": 4.312,
+    "eval_steps_per_second": 2.156
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.9532710280373832,
-  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset-buddhika-weerasinghe/checkpoint-1800",
+  "best_metric": 0.9763313609467456,
+  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset-buddhika-weerasinghe/checkpoint-1620",
   "epoch": 9.1,
   "global_step": 1800,
   "is_hyper_param_search": false,
@@ -10,198 +10,198 @@
     {
       "epoch": 0.06,
       "learning_rate": 2.777777777777778e-05,
-      "loss": 2.4197,
+      "loss": 2.4105,
       "step": 100
     },
     {
       "epoch": 0.1,
-      "eval_accuracy": 0.5327102803738317,
-      "eval_loss": 1.6377586126327515,
-      "eval_runtime": 24.7176,
-      "eval_samples_per_second": 4.329,
-      "eval_steps_per_second": 2.185,
+      "eval_accuracy": 0.46153846153846156,
+      "eval_loss": 1.6308622360229492,
+      "eval_runtime": 77.3029,
+      "eval_samples_per_second": 4.372,
+      "eval_steps_per_second": 2.186,
       "step": 180
     },
     {
       "epoch": 1.01,
       "learning_rate": 4.938271604938271e-05,
-      "loss": 1.8361,
+      "loss": 1.8958,
       "step": 200
     },
     {
       "epoch": 1.07,
       "learning_rate": 4.62962962962963e-05,
-      "loss": 0.8573,
+      "loss": 0.9175,
       "step": 300
     },
     {
       "epoch": 1.1,
-      "eval_accuracy": 0.8411214953271028,
-      "eval_loss": 0.6645746827125549,
-      "eval_runtime": 25.158,
-      "eval_samples_per_second": 4.253,
-      "eval_steps_per_second": 2.146,
+      "eval_accuracy": 0.878698224852071,
+      "eval_loss": 0.4374070167541504,
+      "eval_runtime": 78.2783,
+      "eval_samples_per_second": 4.318,
+      "eval_steps_per_second": 2.159,
       "step": 360
     },
     {
       "epoch": 2.02,
       "learning_rate": 4.3209876543209875e-05,
-      "loss": 0.5808,
+      "loss": 0.6119,
       "step": 400
     },
     {
       "epoch": 2.08,
       "learning_rate": 4.012345679012346e-05,
-      "loss": 0.3788,
+      "loss": 0.5086,
       "step": 500
     },
     {
       "epoch": 2.1,
-      "eval_accuracy": 0.8691588785046729,
-      "eval_loss": 0.47293204069137573,
-      "eval_runtime": 24.7596,
-      "eval_samples_per_second": 4.322,
-      "eval_steps_per_second": 2.181,
+      "eval_accuracy": 0.8905325443786982,
+      "eval_loss": 0.38013187050819397,
+      "eval_runtime": 74.0577,
+      "eval_samples_per_second": 4.564,
+      "eval_steps_per_second": 2.282,
       "step": 540
     },
     {
       "epoch": 3.03,
       "learning_rate": 3.7037037037037037e-05,
-      "loss": 0.3202,
+      "loss": 0.4112,
       "step": 600
     },
     {
       "epoch": 3.09,
       "learning_rate": 3.395061728395062e-05,
-      "loss": 0.3128,
+      "loss": 0.2994,
       "step": 700
     },
     {
       "epoch": 3.1,
-      "eval_accuracy": 0.8785046728971962,
-      "eval_loss": 0.4594048857688904,
-      "eval_runtime": 24.6582,
-      "eval_samples_per_second": 4.339,
-      "eval_steps_per_second": 2.19,
+      "eval_accuracy": 0.8816568047337278,
+      "eval_loss": 0.346200168132782,
+      "eval_runtime": 77.9704,
+      "eval_samples_per_second": 4.335,
+      "eval_steps_per_second": 2.167,
       "step": 720
     },
     {
       "epoch": 4.04,
       "learning_rate": 3.08641975308642e-05,
-      "loss": 0.1997,
+      "loss": 0.2183,
       "step": 800
     },
     {
       "epoch": 4.1,
       "learning_rate": 2.777777777777778e-05,
-      "loss": 0.1661,
+      "loss": 0.1555,
       "step": 900
     },
     {
       "epoch": 4.1,
-      "eval_accuracy": 0.8317757009345794,
-      "eval_loss": 0.5953821539878845,
-      "eval_runtime": 24.4322,
-      "eval_samples_per_second": 4.379,
-      "eval_steps_per_second": 2.21,
+      "eval_accuracy": 0.9230769230769231,
+      "eval_loss": 0.3273673951625824,
+      "eval_runtime": 77.1136,
+      "eval_samples_per_second": 4.383,
+      "eval_steps_per_second": 2.192,
       "step": 900
     },
     {
       "epoch": 5.06,
       "learning_rate": 2.4691358024691357e-05,
-      "loss": 0.1321,
+      "loss": 0.1337,
       "step": 1000
     },
     {
       "epoch": 5.1,
-      "eval_accuracy": 0.9345794392523364,
-      "eval_loss": 0.19723989069461823,
-      "eval_runtime": 24.5315,
-      "eval_samples_per_second": 4.362,
-      "eval_steps_per_second": 2.201,
+      "eval_accuracy": 0.9615384615384616,
+      "eval_loss": 0.1434585154056549,
+      "eval_runtime": 77.4654,
+      "eval_samples_per_second": 4.363,
+      "eval_steps_per_second": 2.182,
       "step": 1080
     },
     {
       "epoch": 6.01,
       "learning_rate": 2.1604938271604937e-05,
-      "loss": 0.0066,
+      "loss": 0.0749,
       "step": 1100
     },
     {
       "epoch": 6.07,
       "learning_rate": 1.8518518518518518e-05,
-      "loss": 0.0602,
+      "loss": 0.021,
       "step": 1200
     },
     {
       "epoch": 6.1,
-      "eval_accuracy": 0.9252336448598131,
-      "eval_loss": 0.3471812903881073,
-      "eval_runtime": 24.5462,
-      "eval_samples_per_second": 4.359,
-      "eval_steps_per_second": 2.2,
+      "eval_accuracy": 0.9615384615384616,
+      "eval_loss": 0.18787238001823425,
+      "eval_runtime": 77.5975,
+      "eval_samples_per_second": 4.356,
+      "eval_steps_per_second": 2.178,
       "step": 1260
     },
     {
       "epoch": 7.02,
       "learning_rate": 1.54320987654321e-05,
-      "loss": 0.0025,
+      "loss": 0.003,
       "step": 1300
     },
     {
       "epoch": 7.08,
       "learning_rate": 1.2345679012345678e-05,
-      "loss": 0.0269,
+      "loss": 0.0485,
       "step": 1400
     },
     {
       "epoch": 7.1,
-      "eval_accuracy": 0.9252336448598131,
-      "eval_loss": 0.3087226152420044,
-      "eval_runtime": 24.5511,
-      "eval_samples_per_second": 4.358,
-      "eval_steps_per_second": 2.199,
+      "eval_accuracy": 0.9674556213017751,
+      "eval_loss": 0.10547798126935959,
+      "eval_runtime": 76.9785,
+      "eval_samples_per_second": 4.391,
+      "eval_steps_per_second": 2.195,
       "step": 1440
     },
     {
       "epoch": 8.03,
       "learning_rate": 9.259259259259259e-06,
-      "loss": 0.0116,
+      "loss": 0.0021,
       "step": 1500
     },
     {
       "epoch": 8.09,
       "learning_rate": 6.172839506172839e-06,
-      "loss": 0.0058,
+      "loss": 0.0019,
       "step": 1600
     },
     {
       "epoch": 8.1,
-      "eval_accuracy": 0.9439252336448598,
-      "eval_loss": 0.31687241792678833,
-      "eval_runtime": 24.5581,
-      "eval_samples_per_second": 4.357,
-      "eval_steps_per_second": 2.199,
+      "eval_accuracy": 0.9763313609467456,
+      "eval_loss": 0.08644197881221771,
+      "eval_runtime": 78.2696,
+      "eval_samples_per_second": 4.318,
+      "eval_steps_per_second": 2.159,
       "step": 1620
     },
     {
       "epoch": 9.04,
       "learning_rate": 3.0864197530864196e-06,
-      "loss": 0.0029,
+      "loss": 0.0038,
       "step": 1700
     },
     {
       "epoch": 9.1,
       "learning_rate": 0.0,
-      "loss": 0.0061,
+      "loss": 0.0054,
       "step": 1800
     },
     {
       "epoch": 9.1,
-      "eval_accuracy": 0.9532710280373832,
-      "eval_loss": 0.2525083124637604,
-      "eval_runtime": 24.7778,
-      "eval_samples_per_second": 4.318,
+      "eval_accuracy": 0.9763313609467456,
+      "eval_loss": 0.08389929682016373,
+      "eval_runtime": 77.5761,
+      "eval_samples_per_second": 4.357,
       "eval_steps_per_second": 2.179,
       "step": 1800
     },
@@ -209,27 +209,27 @@
       "epoch": 9.1,
       "step": 1800,
       "total_flos": 4.4862325749448704e+18,
-      "train_loss": 0.40701433156927425,
-      "train_runtime": 1997.2528,
-      "train_samples_per_second": 1.802,
-      "train_steps_per_second": 0.901
+      "train_loss": 0.42906751138468585,
+      "train_runtime": 2457.281,
+      "train_samples_per_second": 1.465,
+      "train_steps_per_second": 0.733
     },
     {
       "epoch": 9.1,
-      "eval_accuracy": 0.065,
-      "eval_loss": 8.357227325439453,
-      "eval_runtime": 46.1165,
-      "eval_samples_per_second": 4.337,
-      "eval_steps_per_second": 2.168,
+      "eval_accuracy": 0.0,
+      "eval_loss": 8.188952445983887,
+      "eval_runtime": 2.195,
+      "eval_samples_per_second": 3.645,
+      "eval_steps_per_second": 1.822,
       "step": 1800
     },
     {
       "epoch": 9.1,
-      "eval_accuracy": 0.065,
-      "eval_loss": 8.357227325439453,
-      "eval_runtime": 45.341,
-      "eval_samples_per_second": 4.411,
-      "eval_steps_per_second": 2.206,
+      "eval_accuracy": 0.0,
+      "eval_loss": 8.188952445983887,
+      "eval_runtime": 1.8553,
+      "eval_samples_per_second": 4.312,
+      "eval_steps_per_second": 2.156,
       "step": 1800
     }
   ],