End of training

README.md

@@ -18,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingface.co/MCG-NJU/videomae-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5786
+- Loss: 0.6020
 - Accuracy: 0.8852
 
 ## Model description

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 32.01538461538462,
+    "eval_accuracy": 0.8852459016393442,
+    "eval_loss": 0.6019696593284607,
+    "eval_runtime": 1.1018,
+    "eval_samples_per_second": 55.366,
+    "eval_steps_per_second": 1.815
+}

runs/Jun28_02-01-44_GAN-SVR/events.out.tfevents.1751043961.GAN-SVR.323061.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:44cf220a559b0114748aa4f766eba837233aa113daa08ca928e2e4d9abdb7683
-size 411
+oid sha256:fcc64b9691d7066d32ba016089a7bd9b037d2f99a2eb27093e8279fb6931d717
+size 734

test_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 32.01538461538462,
+    "eval_accuracy": 0.8852459016393442,
+    "eval_loss": 0.6019696593284607,
+    "eval_runtime": 1.1018,
+    "eval_samples_per_second": 55.366,
+    "eval_steps_per_second": 1.815
+}

trainer_state.json

@@ -0,0 +1,448 @@
+{
+  "best_metric": 0.8852459016393442,
+  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-112",
+  "epoch": 32.01538461538462,
+  "eval_steps": 500,
+  "global_step": 130,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03076923076923077,
+      "eval_accuracy": 0.04918032786885246,
+      "eval_loss": 3.0568840503692627,
+      "eval_runtime": 1.0821,
+      "eval_samples_per_second": 56.372,
+      "eval_steps_per_second": 1.848,
+      "step": 4
+    },
+    {
+      "epoch": 1.0307692307692307,
+      "eval_accuracy": 0.03278688524590164,
+      "eval_loss": 3.0837085247039795,
+      "eval_runtime": 1.0457,
+      "eval_samples_per_second": 58.335,
+      "eval_steps_per_second": 1.913,
+      "step": 8
+    },
+    {
+      "epoch": 2.0153846153846153,
+      "grad_norm": 6.3322553634643555,
+      "learning_rate": 3.846153846153846e-05,
+      "loss": 3.0022,
+      "step": 10
+    },
+    {
+      "epoch": 2.0307692307692307,
+      "eval_accuracy": 0.09836065573770492,
+      "eval_loss": 3.067941188812256,
+      "eval_runtime": 1.0536,
+      "eval_samples_per_second": 57.894,
+      "eval_steps_per_second": 1.898,
+      "step": 12
+    },
+    {
+      "epoch": 3.0307692307692307,
+      "eval_accuracy": 0.08196721311475409,
+      "eval_loss": 3.068866014480591,
+      "eval_runtime": 1.0514,
+      "eval_samples_per_second": 58.02,
+      "eval_steps_per_second": 1.902,
+      "step": 16
+    },
+    {
+      "epoch": 4.030769230769231,
+      "grad_norm": 7.941536903381348,
+      "learning_rate": 4.700854700854701e-05,
+      "loss": 2.9588,
+      "step": 20
+    },
+    {
+      "epoch": 4.030769230769231,
+      "eval_accuracy": 0.06557377049180328,
+      "eval_loss": 3.020595073699951,
+      "eval_runtime": 1.0479,
+      "eval_samples_per_second": 58.212,
+      "eval_steps_per_second": 1.909,
+      "step": 20
+    },
+    {
+      "epoch": 5.030769230769231,
+      "eval_accuracy": 0.18032786885245902,
+      "eval_loss": 2.9200916290283203,
+      "eval_runtime": 1.0555,
+      "eval_samples_per_second": 57.79,
+      "eval_steps_per_second": 1.895,
+      "step": 24
+    },
+    {
+      "epoch": 6.030769230769231,
+      "eval_accuracy": 0.14754098360655737,
+      "eval_loss": 2.871077060699463,
+      "eval_runtime": 1.0507,
+      "eval_samples_per_second": 58.055,
+      "eval_steps_per_second": 1.903,
+      "step": 28
+    },
+    {
+      "epoch": 7.015384615384615,
+      "grad_norm": 7.486019611358643,
+      "learning_rate": 4.2735042735042735e-05,
+      "loss": 2.7723,
+      "step": 30
+    },
+    {
+      "epoch": 7.030769230769231,
+      "eval_accuracy": 0.18032786885245902,
+      "eval_loss": 2.7757270336151123,
+      "eval_runtime": 1.299,
+      "eval_samples_per_second": 46.961,
+      "eval_steps_per_second": 1.54,
+      "step": 32
+    },
+    {
+      "epoch": 8.03076923076923,
+      "eval_accuracy": 0.39344262295081966,
+      "eval_loss": 2.4965713024139404,
+      "eval_runtime": 1.0435,
+      "eval_samples_per_second": 58.458,
+      "eval_steps_per_second": 1.917,
+      "step": 36
+    },
+    {
+      "epoch": 9.03076923076923,
+      "grad_norm": 12.012168884277344,
+      "learning_rate": 3.846153846153846e-05,
+      "loss": 2.3535,
+      "step": 40
+    },
+    {
+      "epoch": 9.03076923076923,
+      "eval_accuracy": 0.29508196721311475,
+      "eval_loss": 2.3920199871063232,
+      "eval_runtime": 1.0595,
+      "eval_samples_per_second": 57.577,
+      "eval_steps_per_second": 1.888,
+      "step": 40
+    },
+    {
+      "epoch": 10.03076923076923,
+      "eval_accuracy": 0.47540983606557374,
+      "eval_loss": 2.150529146194458,
+      "eval_runtime": 1.045,
+      "eval_samples_per_second": 58.374,
+      "eval_steps_per_second": 1.914,
+      "step": 44
+    },
+    {
+      "epoch": 11.03076923076923,
+      "eval_accuracy": 0.5081967213114754,
+      "eval_loss": 2.054077386856079,
+      "eval_runtime": 1.0446,
+      "eval_samples_per_second": 58.397,
+      "eval_steps_per_second": 1.915,
+      "step": 48
+    },
+    {
+      "epoch": 12.015384615384615,
+      "grad_norm": 10.503146171569824,
+      "learning_rate": 3.418803418803419e-05,
+      "loss": 1.7855,
+      "step": 50
+    },
+    {
+      "epoch": 12.03076923076923,
+      "eval_accuracy": 0.5901639344262295,
+      "eval_loss": 1.8268964290618896,
+      "eval_runtime": 1.0537,
+      "eval_samples_per_second": 57.891,
+      "eval_steps_per_second": 1.898,
+      "step": 52
+    },
+    {
+      "epoch": 13.03076923076923,
+      "eval_accuracy": 0.6557377049180327,
+      "eval_loss": 1.6476784944534302,
+      "eval_runtime": 1.075,
+      "eval_samples_per_second": 56.743,
+      "eval_steps_per_second": 1.86,
+      "step": 56
+    },
+    {
+      "epoch": 14.03076923076923,
+      "grad_norm": 8.819632530212402,
+      "learning_rate": 2.9914529914529915e-05,
+      "loss": 1.2907,
+      "step": 60
+    },
+    {
+      "epoch": 14.03076923076923,
+      "eval_accuracy": 0.7049180327868853,
+      "eval_loss": 1.5733423233032227,
+      "eval_runtime": 1.0559,
+      "eval_samples_per_second": 57.773,
+      "eval_steps_per_second": 1.894,
+      "step": 60
+    },
+    {
+      "epoch": 15.03076923076923,
+      "eval_accuracy": 0.8032786885245902,
+      "eval_loss": 1.3481733798980713,
+      "eval_runtime": 1.0463,
+      "eval_samples_per_second": 58.298,
+      "eval_steps_per_second": 1.911,
+      "step": 64
+    },
+    {
+      "epoch": 16.03076923076923,
+      "eval_accuracy": 0.7704918032786885,
+      "eval_loss": 1.2641762495040894,
+      "eval_runtime": 1.0633,
+      "eval_samples_per_second": 57.37,
+      "eval_steps_per_second": 1.881,
+      "step": 68
+    },
+    {
+      "epoch": 17.015384615384615,
+      "grad_norm": 6.274653911590576,
+      "learning_rate": 2.564102564102564e-05,
+      "loss": 0.8499,
+      "step": 70
+    },
+    {
+      "epoch": 17.03076923076923,
+      "eval_accuracy": 0.8688524590163934,
+      "eval_loss": 1.129570722579956,
+      "eval_runtime": 1.0682,
+      "eval_samples_per_second": 57.105,
+      "eval_steps_per_second": 1.872,
+      "step": 72
+    },
+    {
+      "epoch": 18.03076923076923,
+      "eval_accuracy": 0.819672131147541,
+      "eval_loss": 1.105241298675537,
+      "eval_runtime": 1.0651,
+      "eval_samples_per_second": 57.27,
+      "eval_steps_per_second": 1.878,
+      "step": 76
+    },
+    {
+      "epoch": 19.03076923076923,
+      "grad_norm": 7.309925556182861,
+      "learning_rate": 2.1367521367521368e-05,
+      "loss": 0.5723,
+      "step": 80
+    },
+    {
+      "epoch": 19.03076923076923,
+      "eval_accuracy": 0.819672131147541,
+      "eval_loss": 1.0154109001159668,
+      "eval_runtime": 1.0751,
+      "eval_samples_per_second": 56.738,
+      "eval_steps_per_second": 1.86,
+      "step": 80
+    },
+    {
+      "epoch": 20.03076923076923,
+      "eval_accuracy": 0.8524590163934426,
+      "eval_loss": 0.8963517546653748,
+      "eval_runtime": 1.085,
+      "eval_samples_per_second": 56.224,
+      "eval_steps_per_second": 1.843,
+      "step": 84
+    },
+    {
+      "epoch": 21.03076923076923,
+      "eval_accuracy": 0.8688524590163934,
+      "eval_loss": 0.8365468382835388,
+      "eval_runtime": 1.0948,
+      "eval_samples_per_second": 55.717,
+      "eval_steps_per_second": 1.827,
+      "step": 88
+    },
+    {
+      "epoch": 22.015384615384615,
+      "grad_norm": 5.605804443359375,
+      "learning_rate": 1.7094017094017095e-05,
+      "loss": 0.3825,
+      "step": 90
+    },
+    {
+      "epoch": 22.03076923076923,
+      "eval_accuracy": 0.8524590163934426,
+      "eval_loss": 0.772179901599884,
+      "eval_runtime": 1.068,
+      "eval_samples_per_second": 57.117,
+      "eval_steps_per_second": 1.873,
+      "step": 92
+    },
+    {
+      "epoch": 23.03076923076923,
+      "eval_accuracy": 0.8688524590163934,
+      "eval_loss": 0.7270120978355408,
+      "eval_runtime": 1.072,
+      "eval_samples_per_second": 56.904,
+      "eval_steps_per_second": 1.866,
+      "step": 96
+    },
+    {
+      "epoch": 24.03076923076923,
+      "grad_norm": 2.8808488845825195,
+      "learning_rate": 1.282051282051282e-05,
+      "loss": 0.2545,
+      "step": 100
+    },
+    {
+      "epoch": 24.03076923076923,
+      "eval_accuracy": 0.8524590163934426,
+      "eval_loss": 0.7208819389343262,
+      "eval_runtime": 1.0741,
+      "eval_samples_per_second": 56.794,
+      "eval_steps_per_second": 1.862,
+      "step": 100
+    },
+    {
+      "epoch": 25.03076923076923,
+      "eval_accuracy": 0.8688524590163934,
+      "eval_loss": 0.6552789807319641,
+      "eval_runtime": 1.0496,
+      "eval_samples_per_second": 58.115,
+      "eval_steps_per_second": 1.905,
+      "step": 104
+    },
+    {
+      "epoch": 26.03076923076923,
+      "eval_accuracy": 0.8688524590163934,
+      "eval_loss": 0.622446596622467,
+      "eval_runtime": 1.1046,
+      "eval_samples_per_second": 55.224,
+      "eval_steps_per_second": 1.811,
+      "step": 108
+    },
+    {
+      "epoch": 27.015384615384615,
+      "grad_norm": 1.8433109521865845,
+      "learning_rate": 8.547008547008548e-06,
+      "loss": 0.1936,
+      "step": 110
+    },
+    {
+      "epoch": 27.03076923076923,
+      "eval_accuracy": 0.8852459016393442,
+      "eval_loss": 0.6002484560012817,
+      "eval_runtime": 1.115,
+      "eval_samples_per_second": 54.711,
+      "eval_steps_per_second": 1.794,
+      "step": 112
+    },
+    {
+      "epoch": 28.03076923076923,
+      "eval_accuracy": 0.8852459016393442,
+      "eval_loss": 0.5697656869888306,
+      "eval_runtime": 1.0584,
+      "eval_samples_per_second": 57.635,
+      "eval_steps_per_second": 1.89,
+      "step": 116
+    },
+    {
+      "epoch": 29.03076923076923,
+      "grad_norm": 1.7687464952468872,
+      "learning_rate": 4.273504273504274e-06,
+      "loss": 0.1622,
+      "step": 120
+    },
+    {
+      "epoch": 29.03076923076923,
+      "eval_accuracy": 0.8688524590163934,
+      "eval_loss": 0.5786339044570923,
+      "eval_runtime": 1.0579,
+      "eval_samples_per_second": 57.66,
+      "eval_steps_per_second": 1.89,
+      "step": 120
+    },
+    {
+      "epoch": 30.03076923076923,
+      "eval_accuracy": 0.8688524590163934,
+      "eval_loss": 0.5774585008621216,
+      "eval_runtime": 1.0633,
+      "eval_samples_per_second": 57.371,
+      "eval_steps_per_second": 1.881,
+      "step": 124
+    },
+    {
+      "epoch": 31.03076923076923,
+      "eval_accuracy": 0.8688524590163934,
+      "eval_loss": 0.54356849193573,
+      "eval_runtime": 1.0573,
+      "eval_samples_per_second": 57.696,
+      "eval_steps_per_second": 1.892,
+      "step": 128
+    },
+    {
+      "epoch": 32.01538461538462,
+      "grad_norm": 2.1357920169830322,
+      "learning_rate": 0.0,
+      "loss": 0.1456,
+      "step": 130
+    },
+    {
+      "epoch": 32.01538461538462,
+      "eval_accuracy": 0.8688524590163934,
+      "eval_loss": 0.5693949460983276,
+      "eval_runtime": 1.0553,
+      "eval_samples_per_second": 57.806,
+      "eval_steps_per_second": 1.895,
+      "step": 130
+    },
+    {
+      "epoch": 32.01538461538462,
+      "step": 130,
+      "total_flos": 4.4666522108512174e+18,
+      "train_loss": 1.286447219665234,
+      "train_runtime": 202.2451,
+      "train_samples_per_second": 20.569,
+      "train_steps_per_second": 0.643
+    },
+    {
+      "epoch": 32.01538461538462,
+      "eval_accuracy": 0.8852459016393442,
+      "eval_loss": 0.5785899758338928,
+      "eval_runtime": 1.0845,
+      "eval_samples_per_second": 56.249,
+      "eval_steps_per_second": 1.844,
+      "step": 130
+    },
+    {
+      "epoch": 32.01538461538462,
+      "eval_accuracy": 0.8852459016393442,
+      "eval_loss": 0.6019696593284607,
+      "eval_runtime": 1.1018,
+      "eval_samples_per_second": 55.366,
+      "eval_steps_per_second": 1.815,
+      "step": 130
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 130,
+  "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": 4.4666522108512174e+18,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}