Model save

all_results.json

@@ -1,11 +1,11 @@
 {
     "epoch": 7.02,
-    "eval_accuracy": 0.44,
-    "eval_loss": 0.9982941150665283,
-    "eval_runtime": 16.8527,
+    "eval_accuracy": 0.5714285714285714,
+    "eval_loss": 0.9015251398086548,
+    "eval_runtime": 3.5127,
     "eval_samples": 6,
-    "eval_samples_per_second": 1.483,
-    "eval_steps_per_second": 0.178,
+    "eval_samples_per_second": 5.978,
+    "eval_steps_per_second": 0.569,
     "total_flos": 2.208046144034046e+18,
     "train_loss": 0.6113187122344971,
     "train_runtime": 861.0964,

runs/Oct15_02-59-37_226fb5eed268/events.out.tfevents.1697339652.226fb5eed268.363.8

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:56efc207b104c2115f9bf9fc4ebf506611606cd59d05f1aadfc57f3b85de3dc3
-size 734
+oid sha256:1aba308d4d6a03ef1a2ee44596b54cdeb05e837b56a0b046bb868c942280780e
+size 1057

test_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 7.02,
+    "eval_accuracy": 0.5714285714285714,
+    "eval_loss": 0.9015251398086548,
+    "eval_runtime": 3.5127,
+    "eval_samples_per_second": 5.978,
+    "eval_steps_per_second": 0.569
+}

trainer_state.json

@@ -0,0 +1,214 @@
+{
+  "best_metric": 0.44,
+  "best_model_checkpoint": "videomae-base-finetuned-SSBD/checkpoint-21",
+  "epoch": 7.02,
+  "global_step": 150,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.07,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 1.0927,
+      "step": 10
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.814814814814815e-05,
+      "loss": 1.1764,
+      "step": 20
+    },
+    {
+      "epoch": 0.14,
+      "eval_accuracy": 0.44,
+      "eval_loss": 0.9982941150665283,
+      "eval_runtime": 22.6723,
+      "eval_samples_per_second": 1.103,
+      "eval_steps_per_second": 0.132,
+      "step": 21
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 4.4444444444444447e-05,
+      "loss": 0.9538,
+      "step": 30
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 4.074074074074074e-05,
+      "loss": 1.2156,
+      "step": 40
+    },
+    {
+      "epoch": 1.14,
+      "eval_accuracy": 0.4,
+      "eval_loss": 1.1047239303588867,
+      "eval_runtime": 17.0623,
+      "eval_samples_per_second": 1.465,
+      "eval_steps_per_second": 0.176,
+      "step": 42
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 3.7037037037037037e-05,
+      "loss": 0.7382,
+      "step": 50
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 1.0214,
+      "step": 60
+    },
+    {
+      "epoch": 2.14,
+      "eval_accuracy": 0.2,
+      "eval_loss": 1.2947560548782349,
+      "eval_runtime": 16.4433,
+      "eval_samples_per_second": 1.52,
+      "eval_steps_per_second": 0.182,
+      "step": 63
+    },
+    {
+      "epoch": 3.05,
+      "learning_rate": 2.962962962962963e-05,
+      "loss": 0.8333,
+      "step": 70
+    },
+    {
+      "epoch": 3.11,
+      "learning_rate": 2.5925925925925925e-05,
+      "loss": 0.5585,
+      "step": 80
+    },
+    {
+      "epoch": 3.14,
+      "eval_accuracy": 0.44,
+      "eval_loss": 2.0399672985076904,
+      "eval_runtime": 16.7172,
+      "eval_samples_per_second": 1.495,
+      "eval_steps_per_second": 0.179,
+      "step": 84
+    },
+    {
+      "epoch": 4.04,
+      "learning_rate": 2.2222222222222223e-05,
+      "loss": 0.5306,
+      "step": 90
+    },
+    {
+      "epoch": 4.11,
+      "learning_rate": 1.8518518518518518e-05,
+      "loss": 0.3809,
+      "step": 100
+    },
+    {
+      "epoch": 4.14,
+      "eval_accuracy": 0.4,
+      "eval_loss": 2.2226576805114746,
+      "eval_runtime": 16.8941,
+      "eval_samples_per_second": 1.48,
+      "eval_steps_per_second": 0.178,
+      "step": 105
+    },
+    {
+      "epoch": 5.03,
+      "learning_rate": 1.4814814814814815e-05,
+      "loss": 0.2621,
+      "step": 110
+    },
+    {
+      "epoch": 5.1,
+      "learning_rate": 1.1111111111111112e-05,
+      "loss": 0.1665,
+      "step": 120
+    },
+    {
+      "epoch": 5.14,
+      "eval_accuracy": 0.12,
+      "eval_loss": 2.2652955055236816,
+      "eval_runtime": 16.6771,
+      "eval_samples_per_second": 1.499,
+      "eval_steps_per_second": 0.18,
+      "step": 126
+    },
+    {
+      "epoch": 6.03,
+      "learning_rate": 7.4074074074074075e-06,
+      "loss": 0.1161,
+      "step": 130
+    },
+    {
+      "epoch": 6.09,
+      "learning_rate": 3.7037037037037037e-06,
+      "loss": 0.0736,
+      "step": 140
+    },
+    {
+      "epoch": 6.14,
+      "eval_accuracy": 0.24,
+      "eval_loss": 2.824446201324463,
+      "eval_runtime": 17.0232,
+      "eval_samples_per_second": 1.469,
+      "eval_steps_per_second": 0.176,
+      "step": 147
+    },
+    {
+      "epoch": 7.02,
+      "learning_rate": 0.0,
+      "loss": 0.0501,
+      "step": 150
+    },
+    {
+      "epoch": 7.02,
+      "eval_accuracy": 0.2,
+      "eval_loss": 2.841770887374878,
+      "eval_runtime": 16.8793,
+      "eval_samples_per_second": 1.481,
+      "eval_steps_per_second": 0.178,
+      "step": 150
+    },
+    {
+      "epoch": 7.02,
+      "step": 150,
+      "total_flos": 2.208046144034046e+18,
+      "train_loss": 0.6113187122344971,
+      "train_runtime": 861.0964,
+      "train_samples_per_second": 2.09,
+      "train_steps_per_second": 0.174
+    },
+    {
+      "epoch": 7.02,
+      "eval_accuracy": 0.5714285714285714,
+      "eval_loss": 0.9015253186225891,
+      "eval_runtime": 7.2806,
+      "eval_samples_per_second": 2.884,
+      "eval_steps_per_second": 0.275,
+      "step": 150
+    },
+    {
+      "epoch": 7.02,
+      "eval_accuracy": 0.44,
+      "eval_loss": 0.9982941150665283,
+      "eval_runtime": 16.8527,
+      "eval_samples_per_second": 1.483,
+      "eval_steps_per_second": 0.178,
+      "step": 150
+    },
+    {
+      "epoch": 7.02,
+      "eval_accuracy": 0.5714285714285714,
+      "eval_loss": 0.9015251398086548,
+      "eval_runtime": 3.5127,
+      "eval_samples_per_second": 5.978,
+      "eval_steps_per_second": 0.569,
+      "step": 150
+    }
+  ],
+  "max_steps": 150,
+  "num_train_epochs": 9223372036854775807,
+  "total_flos": 2.208046144034046e+18,
+  "trial_name": null,
+  "trial_params": null
+}