End of training

README.md

@@ -17,7 +17,7 @@ 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: 0.5305
+- Loss: 0.5324
 - Accuracy: 0.7871
 
 ## Model description

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 7.1,
+    "eval_accuracy": 0.7870967741935484,
+    "eval_loss": 0.5323519110679626,
+    "eval_runtime": 10.925,
+    "eval_samples_per_second": 14.188,
+    "eval_steps_per_second": 0.915
+}

test_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 7.1,
+    "eval_accuracy": 0.7870967741935484,
+    "eval_loss": 0.5323519110679626,
+    "eval_runtime": 10.925,
+    "eval_samples_per_second": 14.188,
+    "eval_steps_per_second": 0.915
+}

trainer_state.json

@@ -0,0 +1,202 @@
+{
+  "best_metric": 0.9142857142857143,
+  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-95",
+  "epoch": 7.101351351351352,
+  "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,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 2.3271,
+      "step": 10
+    },
+    {
+      "epoch": 0.13,
+      "eval_accuracy": 0.24285714285714285,
+      "eval_loss": 2.1349682807922363,
+      "eval_runtime": 5.8502,
+      "eval_samples_per_second": 11.965,
+      "eval_steps_per_second": 0.855,
+      "step": 19
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.81203007518797e-05,
+      "loss": 2.2407,
+      "step": 20
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 4.43609022556391e-05,
+      "loss": 1.9566,
+      "step": 30
+    },
+    {
+      "epoch": 1.13,
+      "eval_accuracy": 0.5857142857142857,
+      "eval_loss": 1.4260412454605103,
+      "eval_runtime": 5.6595,
+      "eval_samples_per_second": 12.369,
+      "eval_steps_per_second": 0.883,
+      "step": 38
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.0601503759398494e-05,
+      "loss": 1.5366,
+      "step": 40
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 3.6842105263157895e-05,
+      "loss": 1.0941,
+      "step": 50
+    },
+    {
+      "epoch": 2.13,
+      "eval_accuracy": 0.7142857142857143,
+      "eval_loss": 0.8856863379478455,
+      "eval_runtime": 5.7621,
+      "eval_samples_per_second": 12.148,
+      "eval_steps_per_second": 0.868,
+      "step": 57
+    },
+    {
+      "epoch": 3.02,
+      "learning_rate": 3.3082706766917295e-05,
+      "loss": 0.7917,
+      "step": 60
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 2.9323308270676693e-05,
+      "loss": 0.5363,
+      "step": 70
+    },
+    {
+      "epoch": 3.13,
+      "eval_accuracy": 0.7142857142857143,
+      "eval_loss": 0.6663923859596252,
+      "eval_runtime": 5.6929,
+      "eval_samples_per_second": 12.296,
+      "eval_steps_per_second": 0.878,
+      "step": 76
+    },
+    {
+      "epoch": 4.03,
+      "learning_rate": 2.556390977443609e-05,
+      "loss": 0.4737,
+      "step": 80
+    },
+    {
+      "epoch": 4.09,
+      "learning_rate": 2.1804511278195487e-05,
+      "loss": 0.3392,
+      "step": 90
+    },
+    {
+      "epoch": 4.13,
+      "eval_accuracy": 0.9142857142857143,
+      "eval_loss": 0.40226492285728455,
+      "eval_runtime": 5.0014,
+      "eval_samples_per_second": 13.996,
+      "eval_steps_per_second": 1.0,
+      "step": 95
+    },
+    {
+      "epoch": 5.03,
+      "learning_rate": 1.8045112781954888e-05,
+      "loss": 0.3211,
+      "step": 100
+    },
+    {
+      "epoch": 5.1,
+      "learning_rate": 1.4285714285714285e-05,
+      "loss": 0.1765,
+      "step": 110
+    },
+    {
+      "epoch": 5.13,
+      "eval_accuracy": 0.9,
+      "eval_loss": 0.348732054233551,
+      "eval_runtime": 5.6896,
+      "eval_samples_per_second": 12.303,
+      "eval_steps_per_second": 0.879,
+      "step": 114
+    },
+    {
+      "epoch": 6.04,
+      "learning_rate": 1.0526315789473684e-05,
+      "loss": 0.1674,
+      "step": 120
+    },
+    {
+      "epoch": 6.11,
+      "learning_rate": 6.766917293233083e-06,
+      "loss": 0.1318,
+      "step": 130
+    },
+    {
+      "epoch": 6.13,
+      "eval_accuracy": 0.9,
+      "eval_loss": 0.3181443214416504,
+      "eval_runtime": 5.105,
+      "eval_samples_per_second": 13.712,
+      "eval_steps_per_second": 0.979,
+      "step": 133
+    },
+    {
+      "epoch": 7.05,
+      "learning_rate": 3.007518796992481e-06,
+      "loss": 0.1092,
+      "step": 140
+    },
+    {
+      "epoch": 7.1,
+      "eval_accuracy": 0.9,
+      "eval_loss": 0.2886265814304352,
+      "eval_runtime": 5.8952,
+      "eval_samples_per_second": 11.874,
+      "eval_steps_per_second": 0.848,
+      "step": 148
+    },
+    {
+      "epoch": 7.1,
+      "step": 148,
+      "total_flos": 2.915999166844109e+18,
+      "train_loss": 0.8297193227587519,
+      "train_runtime": 366.4935,
+      "train_samples_per_second": 6.461,
+      "train_steps_per_second": 0.404
+    },
+    {
+      "epoch": 7.1,
+      "eval_accuracy": 0.7870967741935484,
+      "eval_loss": 0.530457615852356,
+      "eval_runtime": 11.3358,
+      "eval_samples_per_second": 13.673,
+      "eval_steps_per_second": 0.882,
+      "step": 148
+    },
+    {
+      "epoch": 7.1,
+      "eval_accuracy": 0.7870967741935484,
+      "eval_loss": 0.5323519110679626,
+      "eval_runtime": 10.925,
+      "eval_samples_per_second": 14.188,
+      "eval_steps_per_second": 0.915,
+      "step": 148
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 148,
+  "num_train_epochs": 9223372036854775807,
+  "save_steps": 500,
+  "total_flos": 2.915999166844109e+18,
+  "trial_name": null,
+  "trial_params": null
+}