End of training

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.23,
+    "eval_accuracy": 0.7548387096774194,
+    "eval_loss": 0.6493480801582336,
+    "eval_runtime": 62.791,
+    "eval_samples_per_second": 2.469,
+    "eval_steps_per_second": 0.319
+}

runs/Feb26_08-31-39_numasvr01/events.out.tfevents.1708938342.numasvr01.108911.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7433185a71f032d3e8c7d22046ff0169ca9e6c47279f0869680e16931fded1f7
-size 411
+oid sha256:4367e768b296d53191eab3de9af2f6b8bc299b7860092a0543eb162f2c45bcbf
+size 734

test_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.23,
+    "eval_accuracy": 0.7548387096774194,
+    "eval_loss": 0.6493480801582336,
+    "eval_runtime": 62.791,
+    "eval_samples_per_second": 2.469,
+    "eval_steps_per_second": 0.319
+}

trainer_state.json

@@ -0,0 +1,182 @@
+{
+  "best_metric": 0.8,
+  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-114",
+  "epoch": 3.22972972972973,
+  "eval_steps": 500,
+  "global_step": 148,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.07,
+      "grad_norm": 6.32248592376709,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 2.3623,
+      "step": 10
+    },
+    {
+      "epoch": 0.14,
+      "grad_norm": 7.283350944519043,
+      "learning_rate": 4.81203007518797e-05,
+      "loss": 2.2168,
+      "step": 20
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 7.692025661468506,
+      "learning_rate": 4.43609022556391e-05,
+      "loss": 2.1563,
+      "step": 30
+    },
+    {
+      "epoch": 0.26,
+      "eval_accuracy": 0.5285714285714286,
+      "eval_loss": 1.7596344947814941,
+      "eval_runtime": 29.087,
+      "eval_samples_per_second": 2.407,
+      "eval_steps_per_second": 0.309,
+      "step": 38
+    },
+    {
+      "epoch": 1.01,
+      "grad_norm": 9.136490821838379,
+      "learning_rate": 4.0601503759398494e-05,
+      "loss": 1.9464,
+      "step": 40
+    },
+    {
+      "epoch": 1.08,
+      "grad_norm": 8.407499313354492,
+      "learning_rate": 3.6842105263157895e-05,
+      "loss": 1.5855,
+      "step": 50
+    },
+    {
+      "epoch": 1.15,
+      "grad_norm": 11.171178817749023,
+      "learning_rate": 3.3082706766917295e-05,
+      "loss": 1.24,
+      "step": 60
+    },
+    {
+      "epoch": 1.22,
+      "grad_norm": 16.1917724609375,
+      "learning_rate": 2.9323308270676693e-05,
+      "loss": 1.0928,
+      "step": 70
+    },
+    {
+      "epoch": 1.26,
+      "eval_accuracy": 0.6142857142857143,
+      "eval_loss": 0.9959077835083008,
+      "eval_runtime": 32.6823,
+      "eval_samples_per_second": 2.142,
+      "eval_steps_per_second": 0.275,
+      "step": 76
+    },
+    {
+      "epoch": 2.03,
+      "grad_norm": 10.195868492126465,
+      "learning_rate": 2.556390977443609e-05,
+      "loss": 0.7965,
+      "step": 80
+    },
+    {
+      "epoch": 2.09,
+      "grad_norm": 9.43371868133545,
+      "learning_rate": 2.1804511278195487e-05,
+      "loss": 0.6722,
+      "step": 90
+    },
+    {
+      "epoch": 2.16,
+      "grad_norm": 15.41746997833252,
+      "learning_rate": 1.8045112781954888e-05,
+      "loss": 0.5247,
+      "step": 100
+    },
+    {
+      "epoch": 2.23,
+      "grad_norm": 11.87302017211914,
+      "learning_rate": 1.4285714285714285e-05,
+      "loss": 0.5361,
+      "step": 110
+    },
+    {
+      "epoch": 2.26,
+      "eval_accuracy": 0.8,
+      "eval_loss": 0.6064900159835815,
+      "eval_runtime": 30.6719,
+      "eval_samples_per_second": 2.282,
+      "eval_steps_per_second": 0.293,
+      "step": 114
+    },
+    {
+      "epoch": 3.04,
+      "grad_norm": 11.358328819274902,
+      "learning_rate": 1.0526315789473684e-05,
+      "loss": 0.5129,
+      "step": 120
+    },
+    {
+      "epoch": 3.11,
+      "grad_norm": 2.6366848945617676,
+      "learning_rate": 6.766917293233083e-06,
+      "loss": 0.3151,
+      "step": 130
+    },
+    {
+      "epoch": 3.18,
+      "grad_norm": 5.886019706726074,
+      "learning_rate": 3.007518796992481e-06,
+      "loss": 0.3213,
+      "step": 140
+    },
+    {
+      "epoch": 3.23,
+      "eval_accuracy": 0.7714285714285715,
+      "eval_loss": 0.5719130039215088,
+      "eval_runtime": 32.2022,
+      "eval_samples_per_second": 2.174,
+      "eval_steps_per_second": 0.279,
+      "step": 148
+    },
+    {
+      "epoch": 3.23,
+      "step": 148,
+      "total_flos": 1.460491890402263e+18,
+      "train_loss": 1.1188414193488456,
+      "train_runtime": 1808.8912,
+      "train_samples_per_second": 0.655,
+      "train_steps_per_second": 0.082
+    },
+    {
+      "epoch": 3.23,
+      "eval_accuracy": 0.7548387096774194,
+      "eval_loss": 0.6493480801582336,
+      "eval_runtime": 63.4253,
+      "eval_samples_per_second": 2.444,
+      "eval_steps_per_second": 0.315,
+      "step": 148
+    },
+    {
+      "epoch": 3.23,
+      "eval_accuracy": 0.7548387096774194,
+      "eval_loss": 0.6493480801582336,
+      "eval_runtime": 62.791,
+      "eval_samples_per_second": 2.469,
+      "eval_steps_per_second": 0.319,
+      "step": 148
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 148,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 9223372036854775807,
+  "save_steps": 500,
+  "total_flos": 1.460491890402263e+18,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}