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: 0.7448
-- Accuracy: 0.8088
+- Loss: 0.7195
+- Accuracy: 0.7769
 
 ## Model description
 
@@ -51,8 +51,8 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 1.0487        | 0.5   | 183  | 0.0223          | 1.0      |
-| 0.0073        | 1.5   | 366  | 0.0031          | 1.0      |
+| 1.0601        | 0.5   | 183  | 0.0183          | 1.0      |
+| 0.0083        | 1.5   | 366  | 0.0128          | 0.9939   |
 
 
 ### Framework versions

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.5,
-    "eval_accuracy": 0.8087649402390438,
-    "eval_loss": 0.7448206543922424,
-    "eval_runtime": 59.7072,
-    "eval_samples_per_second": 4.204,
-    "eval_steps_per_second": 0.536
+    "eval_accuracy": 0.7768924302788844,
+    "eval_loss": 0.719463586807251,
+    "eval_runtime": 60.4264,
+    "eval_samples_per_second": 4.154,
+    "eval_steps_per_second": 0.53
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8ab674bd746b6934029bf68cb572d24ffb167847d20744fc3d7312a75f2c7488
+oid sha256:e9cc3db17b187ee40aca5836f148315b68252180d0310f6089c1d4662d6ed835
 size 485103024

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.5,
-    "eval_accuracy": 0.8087649402390438,
-    "eval_loss": 0.7448206543922424,
-    "eval_runtime": 59.7072,
-    "eval_samples_per_second": 4.204,
-    "eval_steps_per_second": 0.536
+    "eval_accuracy": 0.7768924302788844,
+    "eval_loss": 0.719463586807251,
+    "eval_runtime": 60.4264,
+    "eval_samples_per_second": 4.154,
+    "eval_steps_per_second": 0.53
 }

trainer_state.json

@@ -10,59 +10,59 @@
   "log_history": [
     {
       "epoch": 0.273224043715847,
-      "grad_norm": 3.1372175216674805,
+      "grad_norm": 4.7838592529296875,
       "learning_rate": 4.0425531914893614e-05,
-      "loss": 1.0487,
+      "loss": 1.0601,
       "step": 100
     },
     {
       "epoch": 0.5,
       "eval_accuracy": 1.0,
-      "eval_loss": 0.022322725504636765,
-      "eval_runtime": 39.6848,
-      "eval_samples_per_second": 4.107,
-      "eval_steps_per_second": 0.529,
+      "eval_loss": 0.01825304888188839,
+      "eval_runtime": 39.8932,
+      "eval_samples_per_second": 4.086,
+      "eval_steps_per_second": 0.526,
       "step": 183
     },
     {
       "epoch": 1.046448087431694,
-      "grad_norm": 0.4278085231781006,
+      "grad_norm": 0.07183460891246796,
       "learning_rate": 2.5227963525835867e-05,
-      "loss": 0.0582,
+      "loss": 0.051,
       "step": 200
     },
     {
       "epoch": 1.319672131147541,
-      "grad_norm": 0.04304807260632515,
+      "grad_norm": 0.08974410593509674,
       "learning_rate": 1.0030395136778116e-05,
-      "loss": 0.0073,
+      "loss": 0.0083,
       "step": 300
     },
     {
       "epoch": 1.5,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.0030893420334905386,
-      "eval_runtime": 39.5889,
-      "eval_samples_per_second": 4.117,
-      "eval_steps_per_second": 0.53,
+      "eval_accuracy": 0.9938650306748467,
+      "eval_loss": 0.012759056873619556,
+      "eval_runtime": 42.0931,
+      "eval_samples_per_second": 3.872,
+      "eval_steps_per_second": 0.499,
       "step": 366
     },
     {
       "epoch": 1.5,
       "step": 366,
       "total_flos": 2.5655220082940314e+18,
-      "train_loss": 0.3048882863072098,
-      "train_runtime": 1021.1018,
-      "train_samples_per_second": 2.867,
+      "train_loss": 0.3068508925985117,
+      "train_runtime": 1021.4706,
+      "train_samples_per_second": 2.866,
       "train_steps_per_second": 0.358
     },
     {
       "epoch": 1.5,
-      "eval_accuracy": 0.8087649402390438,
-      "eval_loss": 0.7448206543922424,
-      "eval_runtime": 59.7072,
-      "eval_samples_per_second": 4.204,
-      "eval_steps_per_second": 0.536,
+      "eval_accuracy": 0.7768924302788844,
+      "eval_loss": 0.719463586807251,
+      "eval_runtime": 60.4264,
+      "eval_samples_per_second": 4.154,
+      "eval_steps_per_second": 0.53,
       "step": 366
     }
   ],