End of training

all_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 1.5,
-    "eval_accuracy": 0.055,
-    "eval_loss": 4.912134170532227,
-    "eval_runtime": 45.6456,
-    "eval_samples_per_second": 4.382,
-    "eval_steps_per_second": 2.191
+    "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
 }

runs/Jun28_08-08-17_4f5dfca5dd87/events.out.tfevents.1687941883.4f5dfca5dd87.28.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c805bb99dd5e265499088b0f489d2dc4b1b01c069a6d9e241e049b9d915c39fd
-size 411
+oid sha256:2b5ede4cfcdc7a299a6bb8188ca2ee203c84ba3a79448cc7a5ed87737e4a0e71
+size 734

test_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 1.5,
-    "eval_accuracy": 0.055,
-    "eval_loss": 4.912134170532227,
-    "eval_runtime": 45.6456,
-    "eval_samples_per_second": 4.382,
-    "eval_steps_per_second": 2.191
+    "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
 }

trainer_state.json

@@ -1,79 +1,241 @@
 {
-  "best_metric": 0.822429906542056,
-  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset-buddhika-weerasinghe/checkpoint-360",
-  "epoch": 1.5,
-  "global_step": 360,
+  "best_metric": 0.9532710280373832,
+  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset-buddhika-weerasinghe/checkpoint-1800",
+  "epoch": 9.1,
+  "global_step": 1800,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.28,
-      "learning_rate": 4.012345679012346e-05,
-      "loss": 2.2265,
+      "epoch": 0.06,
+      "learning_rate": 2.777777777777778e-05,
+      "loss": 2.4197,
       "step": 100
     },
     {
-      "epoch": 0.5,
-      "eval_accuracy": 0.719626168224299,
-      "eval_loss": 1.0680640935897827,
-      "eval_runtime": 25.2225,
-      "eval_samples_per_second": 4.242,
-      "eval_steps_per_second": 2.141,
+      "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,
       "step": 180
     },
     {
-      "epoch": 1.06,
-      "learning_rate": 2.4691358024691357e-05,
-      "loss": 1.3376,
+      "epoch": 1.01,
+      "learning_rate": 4.938271604938271e-05,
+      "loss": 1.8361,
       "step": 200
     },
     {
-      "epoch": 1.33,
-      "learning_rate": 9.259259259259259e-06,
-      "loss": 0.5234,
+      "epoch": 1.07,
+      "learning_rate": 4.62962962962963e-05,
+      "loss": 0.8573,
       "step": 300
     },
     {
-      "epoch": 1.5,
-      "eval_accuracy": 0.822429906542056,
-      "eval_loss": 0.4782926142215729,
-      "eval_runtime": 25.0552,
-      "eval_samples_per_second": 4.271,
-      "eval_steps_per_second": 2.155,
+      "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,
       "step": 360
     },
     {
-      "epoch": 1.5,
-      "step": 360,
-      "total_flos": 8.972465149889741e+17,
-      "train_loss": 1.1890253172980414,
-      "train_runtime": 471.6798,
-      "train_samples_per_second": 1.526,
-      "train_steps_per_second": 0.763
+      "epoch": 2.02,
+      "learning_rate": 4.3209876543209875e-05,
+      "loss": 0.5808,
+      "step": 400
     },
     {
-      "epoch": 1.5,
-      "eval_accuracy": 0.055,
-      "eval_loss": 4.912134170532227,
-      "eval_runtime": 47.46,
-      "eval_samples_per_second": 4.214,
-      "eval_steps_per_second": 2.107,
-      "step": 360
+      "epoch": 2.08,
+      "learning_rate": 4.012345679012346e-05,
+      "loss": 0.3788,
+      "step": 500
     },
     {
-      "epoch": 1.5,
-      "eval_accuracy": 0.055,
-      "eval_loss": 4.912134170532227,
-      "eval_runtime": 45.6456,
-      "eval_samples_per_second": 4.382,
-      "eval_steps_per_second": 2.191,
-      "step": 360
+      "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,
+      "step": 540
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 3.7037037037037037e-05,
+      "loss": 0.3202,
+      "step": 600
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 3.395061728395062e-05,
+      "loss": 0.3128,
+      "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,
+      "step": 720
+    },
+    {
+      "epoch": 4.04,
+      "learning_rate": 3.08641975308642e-05,
+      "loss": 0.1997,
+      "step": 800
+    },
+    {
+      "epoch": 4.1,
+      "learning_rate": 2.777777777777778e-05,
+      "loss": 0.1661,
+      "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,
+      "step": 900
+    },
+    {
+      "epoch": 5.06,
+      "learning_rate": 2.4691358024691357e-05,
+      "loss": 0.1321,
+      "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,
+      "step": 1080
+    },
+    {
+      "epoch": 6.01,
+      "learning_rate": 2.1604938271604937e-05,
+      "loss": 0.0066,
+      "step": 1100
+    },
+    {
+      "epoch": 6.07,
+      "learning_rate": 1.8518518518518518e-05,
+      "loss": 0.0602,
+      "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,
+      "step": 1260
+    },
+    {
+      "epoch": 7.02,
+      "learning_rate": 1.54320987654321e-05,
+      "loss": 0.0025,
+      "step": 1300
+    },
+    {
+      "epoch": 7.08,
+      "learning_rate": 1.2345679012345678e-05,
+      "loss": 0.0269,
+      "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,
+      "step": 1440
+    },
+    {
+      "epoch": 8.03,
+      "learning_rate": 9.259259259259259e-06,
+      "loss": 0.0116,
+      "step": 1500
+    },
+    {
+      "epoch": 8.09,
+      "learning_rate": 6.172839506172839e-06,
+      "loss": 0.0058,
+      "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,
+      "step": 1620
+    },
+    {
+      "epoch": 9.04,
+      "learning_rate": 3.0864197530864196e-06,
+      "loss": 0.0029,
+      "step": 1700
+    },
+    {
+      "epoch": 9.1,
+      "learning_rate": 0.0,
+      "loss": 0.0061,
+      "step": 1800
+    },
+    {
+      "epoch": 9.1,
+      "eval_accuracy": 0.9532710280373832,
+      "eval_loss": 0.2525083124637604,
+      "eval_runtime": 24.7778,
+      "eval_samples_per_second": 4.318,
+      "eval_steps_per_second": 2.179,
+      "step": 1800
+    },
+    {
+      "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
+    },
+    {
+      "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,
+      "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,
+      "step": 1800
     }
   ],
-  "max_steps": 360,
+  "max_steps": 1800,
   "num_train_epochs": 9223372036854775807,
-  "total_flos": 8.972465149889741e+17,
+  "total_flos": 4.4862325749448704e+18,
   "trial_name": null,
   "trial_params": null
 }