End of training

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.22972972972973,
+    "eval_accuracy": 0.7741935483870968,
+    "eval_loss": 0.5534085631370544,
+    "eval_runtime": 10.9793,
+    "eval_samples_per_second": 14.117,
+    "eval_steps_per_second": 1.822
+}

test_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.22972972972973,
+    "eval_accuracy": 0.7741935483870968,
+    "eval_loss": 0.5534085631370544,
+    "eval_runtime": 10.9793,
+    "eval_samples_per_second": 14.117,
+    "eval_steps_per_second": 1.822
+}

trainer_state.json

@@ -0,0 +1,194 @@
+{
+  "best_metric": 0.8571428571428571,
+  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-148",
+  "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": 0.06756756756756757,
+      "grad_norm": 6.342924118041992,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 2.3015,
+      "step": 10
+    },
+    {
+      "epoch": 0.13513513513513514,
+      "grad_norm": 7.6155548095703125,
+      "learning_rate": 4.81203007518797e-05,
+      "loss": 2.1922,
+      "step": 20
+    },
+    {
+      "epoch": 0.20270270270270271,
+      "grad_norm": 7.791998863220215,
+      "learning_rate": 4.43609022556391e-05,
+      "loss": 2.1444,
+      "step": 30
+    },
+    {
+      "epoch": 0.25675675675675674,
+      "eval_accuracy": 0.45714285714285713,
+      "eval_loss": 1.8017991781234741,
+      "eval_runtime": 4.64,
+      "eval_samples_per_second": 15.086,
+      "eval_steps_per_second": 1.94,
+      "step": 38
+    },
+    {
+      "epoch": 1.0135135135135136,
+      "grad_norm": 14.773469924926758,
+      "learning_rate": 4.0601503759398494e-05,
+      "loss": 1.9126,
+      "step": 40
+    },
+    {
+      "epoch": 1.0810810810810811,
+      "grad_norm": 10.467906951904297,
+      "learning_rate": 3.6842105263157895e-05,
+      "loss": 1.5059,
+      "step": 50
+    },
+    {
+      "epoch": 1.1486486486486487,
+      "grad_norm": 21.419572830200195,
+      "learning_rate": 3.3082706766917295e-05,
+      "loss": 1.1834,
+      "step": 60
+    },
+    {
+      "epoch": 1.2162162162162162,
+      "grad_norm": 4.917880535125732,
+      "learning_rate": 2.9323308270676693e-05,
+      "loss": 0.876,
+      "step": 70
+    },
+    {
+      "epoch": 1.2567567567567568,
+      "eval_accuracy": 0.5857142857142857,
+      "eval_loss": 1.0229278802871704,
+      "eval_runtime": 4.8096,
+      "eval_samples_per_second": 14.554,
+      "eval_steps_per_second": 1.871,
+      "step": 76
+    },
+    {
+      "epoch": 2.027027027027027,
+      "grad_norm": 5.517823219299316,
+      "learning_rate": 2.556390977443609e-05,
+      "loss": 0.8999,
+      "step": 80
+    },
+    {
+      "epoch": 2.0945945945945947,
+      "grad_norm": 8.552562713623047,
+      "learning_rate": 2.1804511278195487e-05,
+      "loss": 0.7013,
+      "step": 90
+    },
+    {
+      "epoch": 2.1621621621621623,
+      "grad_norm": 21.381250381469727,
+      "learning_rate": 1.8045112781954888e-05,
+      "loss": 0.6283,
+      "step": 100
+    },
+    {
+      "epoch": 2.22972972972973,
+      "grad_norm": 5.894458293914795,
+      "learning_rate": 1.4285714285714285e-05,
+      "loss": 0.4885,
+      "step": 110
+    },
+    {
+      "epoch": 2.2567567567567566,
+      "eval_accuracy": 0.7428571428571429,
+      "eval_loss": 0.672834038734436,
+      "eval_runtime": 4.8713,
+      "eval_samples_per_second": 14.37,
+      "eval_steps_per_second": 1.848,
+      "step": 114
+    },
+    {
+      "epoch": 3.0405405405405403,
+      "grad_norm": 3.988964796066284,
+      "learning_rate": 1.0526315789473684e-05,
+      "loss": 0.4021,
+      "step": 120
+    },
+    {
+      "epoch": 3.108108108108108,
+      "grad_norm": 3.7764477729797363,
+      "learning_rate": 6.766917293233083e-06,
+      "loss": 0.2978,
+      "step": 130
+    },
+    {
+      "epoch": 3.175675675675676,
+      "grad_norm": 10.011561393737793,
+      "learning_rate": 3.007518796992481e-06,
+      "loss": 0.2917,
+      "step": 140
+    },
+    {
+      "epoch": 3.22972972972973,
+      "eval_accuracy": 0.8571428571428571,
+      "eval_loss": 0.49036097526550293,
+      "eval_runtime": 4.6895,
+      "eval_samples_per_second": 14.927,
+      "eval_steps_per_second": 1.919,
+      "step": 148
+    },
+    {
+      "epoch": 3.22972972972973,
+      "step": 148,
+      "total_flos": 1.460491890402263e+18,
+      "train_loss": 1.0841489240929887,
+      "train_runtime": 148.3557,
+      "train_samples_per_second": 7.981,
+      "train_steps_per_second": 0.998
+    },
+    {
+      "epoch": 3.22972972972973,
+      "eval_accuracy": 0.7741935483870968,
+      "eval_loss": 0.5534085035324097,
+      "eval_runtime": 10.6474,
+      "eval_samples_per_second": 14.558,
+      "eval_steps_per_second": 1.878,
+      "step": 148
+    },
+    {
+      "epoch": 3.22972972972973,
+      "eval_accuracy": 0.7741935483870968,
+      "eval_loss": 0.5534085631370544,
+      "eval_runtime": 10.9793,
+      "eval_samples_per_second": 14.117,
+      "eval_steps_per_second": 1.822,
+      "step": 148
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 148,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 9223372036854775807,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.460491890402263e+18,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}