Training in progress, epoch 0

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.2,
-    "eval_accuracy": 0.75,
-    "eval_loss": 0.9399035573005676,
-    "eval_runtime": 13.5901,
-    "eval_samples_per_second": 0.883,
-    "eval_steps_per_second": 0.441
+    "eval_accuracy": 0.8333333333333334,
+    "eval_loss": 0.565607488155365,
+    "eval_runtime": 20.5092,
+    "eval_samples_per_second": 0.585,
+    "eval_steps_per_second": 0.293
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6182a9bfb2373b41b74f4ac9629b324221606974f190129c923b99798086807a
+oid sha256:4f6fff2e07447ea557efe95f08184dc7f85dd73ff3753b3026f381b7d64b8aed
 size 344937368

runs/Aug13_19-26-02_3438a655bd59/events.out.tfevents.1723577881.3438a655bd59.1027.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fefe7170114f5e4412bbf26300f0bdf3d4033206c452b00d89d307d3f0ea36d3
+size 405

runs/Aug13_20-26-46_3438a655bd59/events.out.tfevents.1723580854.3438a655bd59.1027.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b45246ac6f828454ff947e623fdc63c529b92cf6306ead71f1f13c9afba015e
+size 5936

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.2,
-    "eval_accuracy": 0.75,
-    "eval_loss": 0.9399035573005676,
-    "eval_runtime": 13.5901,
-    "eval_samples_per_second": 0.883,
-    "eval_steps_per_second": 0.441
+    "eval_accuracy": 0.8333333333333334,
+    "eval_loss": 0.565607488155365,
+    "eval_runtime": 20.5092,
+    "eval_samples_per_second": 0.585,
+    "eval_steps_per_second": 0.293
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_metric": 1.0,
-  "best_model_checkpoint": "videomae-base-finetuned-Risky-situations/checkpoint-50",
+  "best_model_checkpoint": "videomae-base-finetuned-Risky-situations/checkpoint-25",
   "epoch": 4.2,
   "eval_steps": 500,
   "global_step": 125,
@@ -10,167 +10,149 @@
   "log_history": [
     {
       "epoch": 0.08,
-      "grad_norm": 8.420906066894531,
+      "grad_norm": 14.0280122756958,
       "learning_rate": 3.846153846153846e-05,
-      "loss": 0.8605,
+      "loss": 0.7118,
       "step": 10
     },
     {
       "epoch": 0.16,
-      "grad_norm": 16.786991119384766,
+      "grad_norm": 22.5030574798584,
       "learning_rate": 4.6875e-05,
-      "loss": 0.6079,
+      "loss": 0.7248,
       "step": 20
     },
     {
       "epoch": 0.2,
-      "eval_accuracy": 0.9166666666666666,
-      "eval_loss": 0.3069283962249756,
-      "eval_runtime": 13.6517,
-      "eval_samples_per_second": 0.879,
-      "eval_steps_per_second": 0.44,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.3613555431365967,
+      "eval_runtime": 13.5011,
+      "eval_samples_per_second": 0.889,
+      "eval_steps_per_second": 0.444,
       "step": 25
     },
     {
       "epoch": 1.04,
-      "grad_norm": 8.974565505981445,
+      "grad_norm": 7.6639909744262695,
       "learning_rate": 4.2410714285714285e-05,
-      "loss": 0.5951,
+      "loss": 0.4691,
       "step": 30
     },
     {
       "epoch": 1.12,
-      "grad_norm": 9.046051979064941,
+      "grad_norm": 1.105189323425293,
       "learning_rate": 3.794642857142857e-05,
-      "loss": 0.3517,
+      "loss": 0.148,
       "step": 40
     },
     {
       "epoch": 1.2,
-      "grad_norm": 0.05629388988018036,
+      "grad_norm": 0.10660043358802795,
       "learning_rate": 3.348214285714286e-05,
-      "loss": 0.1006,
+      "loss": 0.829,
       "step": 50
     },
     {
       "epoch": 1.2,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.00287891854532063,
-      "eval_runtime": 13.4227,
+      "eval_accuracy": 0.9166666666666666,
+      "eval_loss": 0.22277389466762543,
+      "eval_runtime": 13.4172,
       "eval_samples_per_second": 0.894,
       "eval_steps_per_second": 0.447,
       "step": 50
     },
     {
       "epoch": 2.08,
-      "grad_norm": 0.01983616128563881,
+      "grad_norm": 144.1703338623047,
       "learning_rate": 2.9017857142857146e-05,
-      "loss": 0.261,
+      "loss": 0.3704,
       "step": 60
     },
     {
       "epoch": 2.16,
-      "grad_norm": 0.09694314002990723,
+      "grad_norm": 0.0353066623210907,
       "learning_rate": 2.455357142857143e-05,
-      "loss": 0.2857,
+      "loss": 0.3176,
       "step": 70
     },
     {
       "epoch": 2.2,
-      "eval_accuracy": 0.6666666666666666,
-      "eval_loss": 1.6325932741165161,
-      "eval_runtime": 13.4779,
-      "eval_samples_per_second": 0.89,
-      "eval_steps_per_second": 0.445,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.006112185772508383,
+      "eval_runtime": 14.0771,
+      "eval_samples_per_second": 0.852,
+      "eval_steps_per_second": 0.426,
       "step": 75
     },
     {
       "epoch": 3.04,
-      "grad_norm": 159.1613006591797,
+      "grad_norm": 0.020980490371584892,
       "learning_rate": 2.0089285714285717e-05,
-      "loss": 0.6469,
+      "loss": 0.3022,
       "step": 80
     },
     {
       "epoch": 3.12,
-      "grad_norm": 0.024221086874604225,
+      "grad_norm": 0.20112857222557068,
       "learning_rate": 1.5625e-05,
-      "loss": 0.0029,
+      "loss": 0.252,
       "step": 90
     },
     {
       "epoch": 3.2,
-      "grad_norm": 0.013663964346051216,
+      "grad_norm": 0.01630263216793537,
       "learning_rate": 1.1160714285714287e-05,
-      "loss": 0.0044,
+      "loss": 0.0027,
       "step": 100
     },
     {
       "epoch": 3.2,
       "eval_accuracy": 1.0,
-      "eval_loss": 0.0005676120636053383,
-      "eval_runtime": 13.7149,
-      "eval_samples_per_second": 0.875,
-      "eval_steps_per_second": 0.437,
+      "eval_loss": 0.0014307590899989009,
+      "eval_runtime": 13.5819,
+      "eval_samples_per_second": 0.884,
+      "eval_steps_per_second": 0.442,
       "step": 100
     },
     {
       "epoch": 4.08,
-      "grad_norm": 385.0979309082031,
+      "grad_norm": 0.049904562532901764,
       "learning_rate": 6.696428571428572e-06,
-      "loss": 0.1138,
+      "loss": 0.0018,
       "step": 110
     },
     {
       "epoch": 4.16,
-      "grad_norm": 0.011926279403269291,
+      "grad_norm": 0.013597003184258938,
       "learning_rate": 2.2321428571428573e-06,
-      "loss": 0.0007,
+      "loss": 0.0011,
       "step": 120
     },
     {
       "epoch": 4.2,
       "eval_accuracy": 1.0,
-      "eval_loss": 0.00049240468069911,
-      "eval_runtime": 13.8419,
-      "eval_samples_per_second": 0.867,
-      "eval_steps_per_second": 0.433,
+      "eval_loss": 0.0012231767177581787,
+      "eval_runtime": 15.5498,
+      "eval_samples_per_second": 0.772,
+      "eval_steps_per_second": 0.386,
       "step": 125
     },
     {
       "epoch": 4.2,
       "step": 125,
-      "total_flos": 3.140082765468795e+17,
-      "train_loss": 0.3065260285772383,
-      "train_runtime": 512.4464,
-      "train_samples_per_second": 0.488,
-      "train_steps_per_second": 0.244
-    },
-    {
-      "epoch": 4.2,
-      "eval_accuracy": 0.75,
-      "eval_loss": 0.9399034380912781,
-      "eval_runtime": 13.7322,
-      "eval_samples_per_second": 0.874,
-      "eval_steps_per_second": 0.437,
-      "step": 125
-    },
-    {
-      "epoch": 4.2,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.00287891854532063,
-      "eval_runtime": 14.128,
-      "eval_samples_per_second": 0.849,
-      "eval_steps_per_second": 0.425,
-      "step": 125
+      "total_flos": 3.11516147367936e+17,
+      "train_loss": 0.3304869447611272,
+      "train_runtime": 473.2289,
+      "train_samples_per_second": 0.528,
+      "train_steps_per_second": 0.264
     },
     {
       "epoch": 4.2,
-      "eval_accuracy": 0.75,
-      "eval_loss": 0.9399035573005676,
-      "eval_runtime": 13.5901,
-      "eval_samples_per_second": 0.883,
-      "eval_steps_per_second": 0.441,
+      "eval_accuracy": 0.8333333333333334,
+      "eval_loss": 0.565607488155365,
+      "eval_runtime": 20.5092,
+      "eval_samples_per_second": 0.585,
+      "eval_steps_per_second": 0.293,
       "step": 125
     }
   ],
@@ -191,7 +173,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 3.140082765468795e+17,
+  "total_flos": 3.11516147367936e+17,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bd21f37360fd4be49f1cc0baf4fde0c84ee177bf033b91e793348d5cb8176f79
+oid sha256:94b3e8139b3b503633372f061bc637be0562cf8570b7ee62587f55669e461798
 size 5176