End of training

Browse files

Files changed (4) hide show

README.md +1 -1
all_results.json +8 -0
test_results.json +8 -0
trainer_state.json +237 -0

README.md CHANGED Viewed

@@ -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: 1.2406
 - Accuracy: 0.6067
 ## Model description

 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: 1.2411
 - Accuracy: 0.6067
 ## Model description

all_results.json ADDED Viewed

	@@ -0,0 +1,8 @@

+{
+    "epoch": 7.102564102564102,
+    "eval_accuracy": 0.6066790352504638,
+    "eval_loss": 1.2410632371902466,
+    "eval_runtime": 1594.8747,
+    "eval_samples_per_second": 0.338,
+    "eval_steps_per_second": 0.043
+}

test_results.json ADDED Viewed

	@@ -0,0 +1,8 @@

+{
+    "epoch": 7.102564102564102,
+    "eval_accuracy": 0.6066790352504638,
+    "eval_loss": 1.2410632371902466,
+    "eval_runtime": 1594.8747,
+    "eval_samples_per_second": 0.338,
+    "eval_steps_per_second": 0.043
+}

trainer_state.json ADDED Viewed

	@@ -0,0 +1,237 @@

+{
+  "best_metric": 0.5360824742268041,
+  "best_model_checkpoint": "videomae-base-finetuned-lift-data/checkpoint-140",
+  "epoch": 7.102564102564102,
+  "eval_steps": 500,
+  "global_step": 156,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0641025641025641,
+      "grad_norm": 5.224398136138916,
+      "learning_rate": 3.125e-05,
+      "loss": 1.6104,
+      "step": 10
+    },
+    {
+      "epoch": 0.1282051282051282,
+      "grad_norm": 6.130595684051514,
+      "learning_rate": 4.8571428571428576e-05,
+      "loss": 1.6135,
+      "step": 20
+    },
+    {
+      "epoch": 0.1282051282051282,
+      "eval_accuracy": 0.22974963181148747,
+      "eval_loss": 1.653665542602539,
+      "eval_runtime": 2113.0859,
+      "eval_samples_per_second": 0.321,
+      "eval_steps_per_second": 0.04,
+      "step": 20
+    },
+    {
+      "epoch": 1.064102564102564,
+      "grad_norm": 4.675447940826416,
+      "learning_rate": 4.5e-05,
+      "loss": 1.4479,
+      "step": 30
+    },
+    {
+      "epoch": 1.1282051282051282,
+      "grad_norm": 9.729595184326172,
+      "learning_rate": 4.1428571428571437e-05,
+      "loss": 1.3333,
+      "step": 40
+    },
+    {
+      "epoch": 1.1282051282051282,
+      "eval_accuracy": 0.28718703976435933,
+      "eval_loss": 1.5571914911270142,
+      "eval_runtime": 2153.3219,
+      "eval_samples_per_second": 0.315,
+      "eval_steps_per_second": 0.039,
+      "step": 40
+    },
+    {
+      "epoch": 2.064102564102564,
+      "grad_norm": 9.443714141845703,
+      "learning_rate": 3.785714285714286e-05,
+      "loss": 1.2919,
+      "step": 50
+    },
+    {
+      "epoch": 2.128205128205128,
+      "grad_norm": 6.667464733123779,
+      "learning_rate": 3.428571428571429e-05,
+      "loss": 1.4219,
+      "step": 60
+    },
+    {
+      "epoch": 2.128205128205128,
+      "eval_accuracy": 0.30338733431516934,
+      "eval_loss": 1.446468472480774,
+      "eval_runtime": 2046.258,
+      "eval_samples_per_second": 0.332,
+      "eval_steps_per_second": 0.042,
+      "step": 60
+    },
+    {
+      "epoch": 3.064102564102564,
+      "grad_norm": 8.536487579345703,
+      "learning_rate": 3.071428571428572e-05,
+      "loss": 1.2711,
+      "step": 70
+    },
+    {
+      "epoch": 3.128205128205128,
+      "grad_norm": 9.417764663696289,
+      "learning_rate": 2.714285714285714e-05,
+      "loss": 1.1874,
+      "step": 80
+    },
+    {
+      "epoch": 3.128205128205128,
+      "eval_accuracy": 0.30633284241531666,
+      "eval_loss": 1.407287836074829,
+      "eval_runtime": 2182.1497,
+      "eval_samples_per_second": 0.311,
+      "eval_steps_per_second": 0.039,
+      "step": 80
+    },
+    {
+      "epoch": 4.064102564102564,
+      "grad_norm": 3.6362509727478027,
+      "learning_rate": 2.357142857142857e-05,
+      "loss": 1.1012,
+      "step": 90
+    },
+    {
+      "epoch": 4.128205128205128,
+      "grad_norm": 5.921740531921387,
+      "learning_rate": 2e-05,
+      "loss": 1.1121,
+      "step": 100
+    },
+    {
+      "epoch": 4.128205128205128,
+      "eval_accuracy": 0.29013254786450665,
+      "eval_loss": 1.324741244316101,
+      "eval_runtime": 2052.0114,
+      "eval_samples_per_second": 0.331,
+      "eval_steps_per_second": 0.041,
+      "step": 100
+    },
+    {
+      "epoch": 5.064102564102564,
+      "grad_norm": 9.157364845275879,
+      "learning_rate": 1.642857142857143e-05,
+      "loss": 1.1462,
+      "step": 110
+    },
+    {
+      "epoch": 5.128205128205128,
+      "grad_norm": 5.3379645347595215,
+      "learning_rate": 1.2857142857142857e-05,
+      "loss": 0.938,
+      "step": 120
+    },
+    {
+      "epoch": 5.128205128205128,
+      "eval_accuracy": 0.32547864506627394,
+      "eval_loss": 1.2014753818511963,
+      "eval_runtime": 2141.2727,
+      "eval_samples_per_second": 0.317,
+      "eval_steps_per_second": 0.04,
+      "step": 120
+    },
+    {
+      "epoch": 6.064102564102564,
+      "grad_norm": 7.765174388885498,
+      "learning_rate": 9.285714285714286e-06,
+      "loss": 0.9989,
+      "step": 130
+    },
+    {
+      "epoch": 6.128205128205128,
+      "grad_norm": 15.063241958618164,
+      "learning_rate": 5.7142857142857145e-06,
+      "loss": 1.1317,
+      "step": 140
+    },
+    {
+      "epoch": 6.128205128205128,
+      "eval_accuracy": 0.5360824742268041,
+      "eval_loss": 1.2458850145339966,
+      "eval_runtime": 2040.1861,
+      "eval_samples_per_second": 0.333,
+      "eval_steps_per_second": 0.042,
+      "step": 140
+    },
+    {
+      "epoch": 7.064102564102564,
+      "grad_norm": 6.829884052276611,
+      "learning_rate": 2.142857142857143e-06,
+      "loss": 1.0411,
+      "step": 150
+    },
+    {
+      "epoch": 7.102564102564102,
+      "eval_accuracy": 0.4683357879234168,
+      "eval_loss": 1.1582751274108887,
+      "eval_runtime": 2040.2319,
+      "eval_samples_per_second": 0.333,
+      "eval_steps_per_second": 0.042,
+      "step": 156
+    },
+    {
+      "epoch": 7.102564102564102,
+      "step": 156,
+      "total_flos": 1.528962156685394e+18,
+      "train_loss": 1.2294957026457176,
+      "train_runtime": 20749.4564,
+      "train_samples_per_second": 0.06,
+      "train_steps_per_second": 0.008
+    },
+    {
+      "epoch": 7.102564102564102,
+      "eval_accuracy": 0.6066790352504638,
+      "eval_loss": 1.2405734062194824,
+      "eval_runtime": 1591.1567,
+      "eval_samples_per_second": 0.339,
+      "eval_steps_per_second": 0.043,
+      "step": 156
+    },
+    {
+      "epoch": 7.102564102564102,
+      "eval_accuracy": 0.6066790352504638,
+      "eval_loss": 1.2410632371902466,
+      "eval_runtime": 1594.8747,
+      "eval_samples_per_second": 0.338,
+      "eval_steps_per_second": 0.043,
+      "step": 156
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 156,
+  "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": 1.528962156685394e+18,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}