End of training

README.md

@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [facebook/timesformer-base-finetuned-k400](https://huggingface.co/facebook/timesformer-base-finetuned-k400) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7022
-- Accuracy: 0.7809
+- Loss: 0.6497
+- Accuracy: 0.7968
 
 ## Model description
 
@@ -51,8 +51,8 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 1.0336        | 0.5   | 183  | 0.0103          | 1.0      |
-| 0.0087        | 1.5   | 366  | 0.0042          | 1.0      |
+| 1.0603        | 0.5   | 183  | 0.0089          | 1.0      |
+| 0.0104        | 1.5   | 366  | 0.0022          | 1.0      |
 
 
 ### Framework versions

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.5,
-    "eval_accuracy": 0.7808764940239044,
-    "eval_loss": 0.7022287845611572,
-    "eval_runtime": 62.6519,
-    "eval_samples_per_second": 4.006,
-    "eval_steps_per_second": 0.511
+    "eval_accuracy": 0.796812749003984,
+    "eval_loss": 0.6496747136116028,
+    "eval_runtime": 59.1137,
+    "eval_samples_per_second": 4.246,
+    "eval_steps_per_second": 0.541
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6bd53dbe0ed93677a91afa0bbdf7adc903575419058d7bd7c7c9a7b35c7d1434
+oid sha256:352d5ee5482ff19f32f64f36d2ffa29df808c5984dfc139a6cff38a2638ab184
 size 485103024

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.5,
-    "eval_accuracy": 0.7808764940239044,
-    "eval_loss": 0.7022287845611572,
-    "eval_runtime": 62.6519,
-    "eval_samples_per_second": 4.006,
-    "eval_steps_per_second": 0.511
+    "eval_accuracy": 0.796812749003984,
+    "eval_loss": 0.6496747136116028,
+    "eval_runtime": 59.1137,
+    "eval_samples_per_second": 4.246,
+    "eval_steps_per_second": 0.541
 }

trainer_state.json

@@ -10,59 +10,59 @@
   "log_history": [
     {
       "epoch": 0.273224043715847,
-      "grad_norm": 0.9716752171516418,
+      "grad_norm": 3.352616310119629,
       "learning_rate": 4.0425531914893614e-05,
-      "loss": 1.0336,
+      "loss": 1.0603,
       "step": 100
     },
     {
       "epoch": 0.5,
       "eval_accuracy": 1.0,
-      "eval_loss": 0.010333087295293808,
-      "eval_runtime": 54.3545,
-      "eval_samples_per_second": 2.999,
-      "eval_steps_per_second": 0.386,
+      "eval_loss": 0.008912273682653904,
+      "eval_runtime": 37.8817,
+      "eval_samples_per_second": 4.303,
+      "eval_steps_per_second": 0.554,
       "step": 183
     },
     {
       "epoch": 1.046448087431694,
-      "grad_norm": 0.0508948378264904,
+      "grad_norm": 0.6868401765823364,
       "learning_rate": 2.5227963525835867e-05,
-      "loss": 0.0556,
+      "loss": 0.0552,
       "step": 200
     },
     {
       "epoch": 1.319672131147541,
-      "grad_norm": 0.0559578500688076,
+      "grad_norm": 0.05696626752614975,
       "learning_rate": 1.0030395136778116e-05,
-      "loss": 0.0087,
+      "loss": 0.0104,
       "step": 300
     },
     {
       "epoch": 1.5,
       "eval_accuracy": 1.0,
-      "eval_loss": 0.004173776134848595,
-      "eval_runtime": 53.366,
-      "eval_samples_per_second": 3.054,
-      "eval_steps_per_second": 0.394,
+      "eval_loss": 0.002216001506894827,
+      "eval_runtime": 41.5511,
+      "eval_samples_per_second": 3.923,
+      "eval_steps_per_second": 0.505,
       "step": 366
     },
     {
       "epoch": 1.5,
       "step": 366,
       "total_flos": 2.5655220082940314e+18,
-      "train_loss": 0.30036494380137957,
-      "train_runtime": 1349.0891,
-      "train_samples_per_second": 2.17,
-      "train_steps_per_second": 0.271
+      "train_loss": 0.30816985707465416,
+      "train_runtime": 1023.3387,
+      "train_samples_per_second": 2.861,
+      "train_steps_per_second": 0.358
     },
     {
       "epoch": 1.5,
-      "eval_accuracy": 0.7808764940239044,
-      "eval_loss": 0.7022287845611572,
-      "eval_runtime": 62.6519,
-      "eval_samples_per_second": 4.006,
-      "eval_steps_per_second": 0.511,
+      "eval_accuracy": 0.796812749003984,
+      "eval_loss": 0.6496747136116028,
+      "eval_runtime": 59.1137,
+      "eval_samples_per_second": 4.246,
+      "eval_steps_per_second": 0.541,
       "step": 366
     }
   ],