Training in progress, epoch 1

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 7.384615384615385,
-    "eval_accuracy": 0.6956521739130435,
-    "eval_loss": 1.0403441190719604,
-    "eval_runtime": 25.3548,
-    "eval_samples_per_second": 0.907,
-    "eval_steps_per_second": 0.118
+    "eval_accuracy": 0.6086956521739131,
+    "eval_loss": 0.8741648197174072,
+    "eval_runtime": 25.876,
+    "eval_samples_per_second": 0.889,
+    "eval_steps_per_second": 0.116
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2393e270b517f927f545a3fd4615f86e4a8dfce029ae5078bbcabc8e55c652ac
+oid sha256:5340e1d21e09c0a16de9de0f3a4d4330031d5928a5fad75e3429a4be7797d5c3
 size 344940444

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 7.384615384615385,
-    "eval_accuracy": 0.6956521739130435,
-    "eval_loss": 1.0403441190719604,
-    "eval_runtime": 25.3548,
-    "eval_samples_per_second": 0.907,
-    "eval_steps_per_second": 0.118
+    "eval_accuracy": 0.6086956521739131,
+    "eval_loss": 0.8741648197174072,
+    "eval_runtime": 25.876,
+    "eval_samples_per_second": 0.889,
+    "eval_steps_per_second": 0.116
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.9166666666666666,
-  "best_model_checkpoint": "videomae-base-finetuned-soccer\\checkpoint-65",
+  "best_metric": 0.5654115080833435,
+  "best_model_checkpoint": "videomae-base-finetuned-soccer\\checkpoint-91",
   "epoch": 7.384615384615385,
   "eval_steps": 500,
   "global_step": 96,
@@ -10,155 +10,155 @@
   "log_history": [
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 5.654183864593506,
+      "grad_norm": 6.056818008422852,
       "learning_rate": 5e-05,
-      "loss": 1.0527,
+      "loss": 1.018,
       "step": 10
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5833333333333334,
-      "eval_loss": 1.025175929069519,
-      "eval_runtime": 14.1316,
-      "eval_samples_per_second": 0.849,
-      "eval_steps_per_second": 0.142,
+      "eval_accuracy": 0.5,
+      "eval_loss": 0.9591365456581116,
+      "eval_runtime": 14.481,
+      "eval_samples_per_second": 0.829,
+      "eval_steps_per_second": 0.138,
       "step": 13
     },
     {
       "epoch": 1.5384615384615383,
-      "grad_norm": 9.78184986114502,
+      "grad_norm": 8.351778984069824,
       "learning_rate": 4.418604651162791e-05,
-      "loss": 0.6937,
+      "loss": 0.8837,
       "step": 20
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8333333333333334,
-      "eval_loss": 0.5850326418876648,
-      "eval_runtime": 13.5201,
-      "eval_samples_per_second": 0.888,
-      "eval_steps_per_second": 0.148,
+      "eval_accuracy": 0.5833333333333334,
+      "eval_loss": 1.0247520208358765,
+      "eval_runtime": 14.6436,
+      "eval_samples_per_second": 0.819,
+      "eval_steps_per_second": 0.137,
       "step": 26
     },
     {
       "epoch": 2.3076923076923075,
-      "grad_norm": 2.8956298828125,
+      "grad_norm": 6.016116619110107,
       "learning_rate": 3.837209302325582e-05,
-      "loss": 0.4002,
+      "loss": 0.7304,
       "step": 30
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.6666666666666666,
-      "eval_loss": 0.5491077303886414,
-      "eval_runtime": 13.511,
-      "eval_samples_per_second": 0.888,
-      "eval_steps_per_second": 0.148,
+      "eval_accuracy": 0.5,
+      "eval_loss": 0.9564730525016785,
+      "eval_runtime": 14.229,
+      "eval_samples_per_second": 0.843,
+      "eval_steps_per_second": 0.141,
       "step": 39
     },
     {
       "epoch": 3.076923076923077,
-      "grad_norm": 5.322183609008789,
+      "grad_norm": 7.938284397125244,
       "learning_rate": 3.2558139534883724e-05,
-      "loss": 0.3079,
+      "loss": 0.7537,
       "step": 40
     },
     {
       "epoch": 3.8461538461538463,
-      "grad_norm": 8.3604097366333,
+      "grad_norm": 4.581155300140381,
       "learning_rate": 2.674418604651163e-05,
-      "loss": 0.216,
+      "loss": 0.6652,
       "step": 50
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8333333333333334,
-      "eval_loss": 0.4196215867996216,
-      "eval_runtime": 13.9122,
-      "eval_samples_per_second": 0.863,
-      "eval_steps_per_second": 0.144,
+      "eval_accuracy": 0.5833333333333334,
+      "eval_loss": 0.7660763263702393,
+      "eval_runtime": 14.1514,
+      "eval_samples_per_second": 0.848,
+      "eval_steps_per_second": 0.141,
       "step": 52
     },
     {
       "epoch": 4.615384615384615,
-      "grad_norm": 9.210854530334473,
+      "grad_norm": 8.447895050048828,
       "learning_rate": 2.0930232558139536e-05,
-      "loss": 0.0978,
+      "loss": 0.4719,
       "step": 60
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.9166666666666666,
-      "eval_loss": 0.38431987166404724,
-      "eval_runtime": 13.7357,
-      "eval_samples_per_second": 0.874,
-      "eval_steps_per_second": 0.146,
+      "eval_accuracy": 0.6666666666666666,
+      "eval_loss": 0.9185061454772949,
+      "eval_runtime": 14.4168,
+      "eval_samples_per_second": 0.832,
+      "eval_steps_per_second": 0.139,
       "step": 65
     },
     {
       "epoch": 5.384615384615385,
-      "grad_norm": 2.0958056449890137,
+      "grad_norm": 9.912740707397461,
       "learning_rate": 1.5116279069767441e-05,
-      "loss": 0.0927,
+      "loss": 0.5202,
       "step": 70
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.9166666666666666,
-      "eval_loss": 0.28933361172676086,
-      "eval_runtime": 13.6757,
-      "eval_samples_per_second": 0.877,
-      "eval_steps_per_second": 0.146,
+      "eval_accuracy": 0.6666666666666666,
+      "eval_loss": 0.7269230484962463,
+      "eval_runtime": 14.2101,
+      "eval_samples_per_second": 0.844,
+      "eval_steps_per_second": 0.141,
       "step": 78
     },
     {
       "epoch": 6.153846153846154,
-      "grad_norm": 0.5293005108833313,
+      "grad_norm": 20.169536590576172,
       "learning_rate": 9.302325581395349e-06,
-      "loss": 0.0614,
+      "loss": 0.4159,
       "step": 80
     },
     {
       "epoch": 6.923076923076923,
-      "grad_norm": 0.14121398329734802,
+      "grad_norm": 5.597208499908447,
       "learning_rate": 3.488372093023256e-06,
-      "loss": 0.0588,
+      "loss": 0.3601,
       "step": 90
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.8333333333333334,
-      "eval_loss": 0.351638525724411,
-      "eval_runtime": 13.5976,
-      "eval_samples_per_second": 0.883,
-      "eval_steps_per_second": 0.147,
+      "eval_accuracy": 0.75,
+      "eval_loss": 0.5654115080833435,
+      "eval_runtime": 14.3617,
+      "eval_samples_per_second": 0.836,
+      "eval_steps_per_second": 0.139,
       "step": 91
     },
     {
       "epoch": 7.384615384615385,
-      "eval_accuracy": 0.8333333333333334,
-      "eval_loss": 0.35855498909950256,
-      "eval_runtime": 14.7901,
-      "eval_samples_per_second": 0.811,
-      "eval_steps_per_second": 0.135,
+      "eval_accuracy": 0.75,
+      "eval_loss": 0.5665368437767029,
+      "eval_runtime": 15.0544,
+      "eval_samples_per_second": 0.797,
+      "eval_steps_per_second": 0.133,
       "step": 96
     },
     {
       "epoch": 7.384615384615385,
       "step": 96,
       "total_flos": 8.959284297745367e+17,
-      "train_loss": 0.31444934258858365,
-      "train_runtime": 1126.2966,
-      "train_samples_per_second": 0.682,
-      "train_steps_per_second": 0.085
+      "train_loss": 0.6283697237571081,
+      "train_runtime": 4534.7888,
+      "train_samples_per_second": 0.169,
+      "train_steps_per_second": 0.021
     },
     {
       "epoch": 7.384615384615385,
-      "eval_accuracy": 0.6956521739130435,
-      "eval_loss": 1.0403441190719604,
-      "eval_runtime": 25.3548,
-      "eval_samples_per_second": 0.907,
-      "eval_steps_per_second": 0.118,
+      "eval_accuracy": 0.6086956521739131,
+      "eval_loss": 0.8741648197174072,
+      "eval_runtime": 25.876,
+      "eval_samples_per_second": 0.889,
+      "eval_steps_per_second": 0.116,
       "step": 96
     }
   ],

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cc5c27f6ca1d8cb5ac026b588ef54cf0579961f08d4257d922f5c4500a7ade71
+oid sha256:c4b0dd5cfcdb65c9312d18ea7b49c2aae747db7e4e56f76d397ff02b39fda25e
 size 5368