Training in progress, epoch 1

all_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.0,
+    "epoch": 3.6923076923076925,
     "eval_accuracy": 1.0,
-    "eval_loss": 0.24726401269435883,
-    "eval_runtime": 13.288,
-    "eval_samples_per_second": 0.677,
-    "eval_steps_per_second": 0.075
+    "eval_loss": 0.0028692723717540503,
+    "eval_runtime": 25.884,
+    "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:ae97de9d2f80c1f70886a6f3a2fff46bbfa607423ff450b3059f929f12622b8f
+oid sha256:674d55e0efe8bdad1c17e5fdfe988d9500c2d0322be92ae643e7ccabc1078c5a
 size 344940444

test_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.0,
+    "epoch": 3.6923076923076925,
     "eval_accuracy": 1.0,
-    "eval_loss": 0.24726401269435883,
-    "eval_runtime": 13.288,
-    "eval_samples_per_second": 0.677,
-    "eval_steps_per_second": 0.075
+    "eval_loss": 0.0028692723717540503,
+    "eval_runtime": 25.884,
+    "eval_samples_per_second": 0.889,
+    "eval_steps_per_second": 0.116
 }

trainer_state.json

@@ -1,79 +1,109 @@
 {
   "best_metric": 1.0,
-  "best_model_checkpoint": "videomae-base-finetuned-soccer\\checkpoint-4",
-  "epoch": 3.0,
+  "best_model_checkpoint": "videomae-base-finetuned-soccer\\checkpoint-13",
+  "epoch": 3.6923076923076925,
   "eval_steps": 500,
-  "global_step": 12,
+  "global_step": 48,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
+    {
+      "epoch": 0.7692307692307693,
+      "grad_norm": 0.3097344934940338,
+      "learning_rate": 4.418604651162791e-05,
+      "loss": 0.4706,
+      "step": 10
+    },
     {
       "epoch": 1.0,
       "eval_accuracy": 1.0,
-      "eval_loss": 0.20363736152648926,
-      "eval_runtime": 9.1182,
-      "eval_samples_per_second": 0.768,
-      "eval_steps_per_second": 0.11,
-      "step": 4
+      "eval_loss": 0.002405111910775304,
+      "eval_runtime": 14.963,
+      "eval_samples_per_second": 0.802,
+      "eval_steps_per_second": 0.134,
+      "step": 13
+    },
+    {
+      "epoch": 1.5384615384615383,
+      "grad_norm": 0.024141477420926094,
+      "learning_rate": 3.2558139534883724e-05,
+      "loss": 0.0027,
+      "step": 20
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 1.0,
-      "eval_loss": 0.010900563560426235,
-      "eval_runtime": 9.0053,
-      "eval_samples_per_second": 0.777,
-      "eval_steps_per_second": 0.111,
-      "step": 8
+      "eval_loss": 0.000666652515064925,
+      "eval_runtime": 14.6823,
+      "eval_samples_per_second": 0.817,
+      "eval_steps_per_second": 0.136,
+      "step": 26
     },
     {
-      "epoch": 2.5,
-      "grad_norm": 0.2005084902048111,
-      "learning_rate": 1e-05,
-      "loss": 0.4029,
-      "step": 10
+      "epoch": 2.3076923076923075,
+      "grad_norm": 0.013116070069372654,
+      "learning_rate": 2.0930232558139536e-05,
+      "loss": 0.0008,
+      "step": 30
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 1.0,
-      "eval_loss": 0.004224988166242838,
-      "eval_runtime": 9.5263,
-      "eval_samples_per_second": 0.735,
-      "eval_steps_per_second": 0.105,
-      "step": 12
+      "eval_loss": 0.0004932748270221055,
+      "eval_runtime": 15.7105,
+      "eval_samples_per_second": 0.764,
+      "eval_steps_per_second": 0.127,
+      "step": 39
     },
     {
-      "epoch": 3.0,
-      "step": 12,
-      "total_flos": 2.09338851031253e+17,
-      "train_loss": 0.3365984517149627,
-      "train_runtime": 293.1214,
-      "train_samples_per_second": 0.655,
-      "train_steps_per_second": 0.041
+      "epoch": 3.076923076923077,
+      "grad_norm": 0.011234880425035954,
+      "learning_rate": 9.302325581395349e-06,
+      "loss": 0.0005,
+      "step": 40
     },
     {
-      "epoch": 3.0,
+      "epoch": 3.6923076923076925,
       "eval_accuracy": 1.0,
-      "eval_loss": 0.24726401269435883,
-      "eval_runtime": 15.5309,
-      "eval_samples_per_second": 0.579,
-      "eval_steps_per_second": 0.064,
-      "step": 12
+      "eval_loss": 0.00047058644122444093,
+      "eval_runtime": 15.9668,
+      "eval_samples_per_second": 0.752,
+      "eval_steps_per_second": 0.125,
+      "step": 48
     },
     {
-      "epoch": 3.0,
+      "epoch": 3.6923076923076925,
+      "step": 48,
+      "total_flos": 4.523254798444462e+17,
+      "train_loss": 0.09896005064365454,
+      "train_runtime": 623.0335,
+      "train_samples_per_second": 0.616,
+      "train_steps_per_second": 0.077
+    },
+    {
+      "epoch": 3.6923076923076925,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.0028692723717540503,
+      "eval_runtime": 32.8199,
+      "eval_samples_per_second": 0.701,
+      "eval_steps_per_second": 0.091,
+      "step": 48
+    },
+    {
+      "epoch": 3.6923076923076925,
       "eval_accuracy": 1.0,
-      "eval_loss": 0.24726401269435883,
-      "eval_runtime": 13.288,
-      "eval_samples_per_second": 0.677,
-      "eval_steps_per_second": 0.075,
-      "step": 12
+      "eval_loss": 0.0028692723717540503,
+      "eval_runtime": 25.884,
+      "eval_samples_per_second": 0.889,
+      "eval_steps_per_second": 0.116,
+      "step": 48
     }
   ],
   "logging_steps": 10,
-  "max_steps": 12,
+  "max_steps": 48,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 3,
+  "num_train_epochs": 4,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -87,8 +117,8 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.09338851031253e+17,
-  "train_batch_size": 16,
+  "total_flos": 4.523254798444462e+17,
+  "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9f9b09f5c64af0e0c1bcc149348b0dbadf5b840eb6b6102ecd7c363cb80ddac4
+oid sha256:1ead260bf9961cab0d9166afe1d69a5f0f6c4275c04c12fd81197181a4a370ad
 size 5368