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: 0.7582
-- Accuracy: 0.7928
+- Loss: 0.7022
+- Accuracy: 0.7809
 
 ## Model description
 
@@ -51,8 +51,8 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 1.0449        | 0.5   | 183  | 0.0172          | 0.9939   |
-| 0.0056        | 1.5   | 366  | 0.0024          | 1.0      |
+| 1.0336        | 0.5   | 183  | 0.0103          | 1.0      |
+| 0.0087        | 1.5   | 366  | 0.0042          | 1.0      |
 
 
 ### Framework versions

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.5,
-    "eval_accuracy": 0.7928286852589641,
-    "eval_loss": 0.758155345916748,
-    "eval_runtime": 81.0092,
-    "eval_samples_per_second": 3.098,
-    "eval_steps_per_second": 0.395
+    "eval_accuracy": 0.7808764940239044,
+    "eval_loss": 0.7022287845611572,
+    "eval_runtime": 62.6519,
+    "eval_samples_per_second": 4.006,
+    "eval_steps_per_second": 0.511
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7716485a7ee0c0fb708492d154c1318b0e0ae81d8b310372a07028d30e4db2a0
+oid sha256:5c0bf1414ce12baedd5613c23fd1d0f6999bf4e33b24843b7fa97712c585201b
 size 485103024

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.5,
-    "eval_accuracy": 0.7928286852589641,
-    "eval_loss": 0.758155345916748,
-    "eval_runtime": 81.0092,
-    "eval_samples_per_second": 3.098,
-    "eval_steps_per_second": 0.395
+    "eval_accuracy": 0.7808764940239044,
+    "eval_loss": 0.7022287845611572,
+    "eval_runtime": 62.6519,
+    "eval_samples_per_second": 4.006,
+    "eval_steps_per_second": 0.511
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_metric": 1.0,
-  "best_model_checkpoint": "/media/cse/HDD/Shawon/shawon/10 fold timesformer/Timesformer_default_fold_10_10_epoch_noAug_batch8_codecheck/checkpoint-366",
+  "best_model_checkpoint": "/media/cse/HDD/Shawon/shawon/10 fold timesformer/Timesformer_default_fold_10_10_epoch_noAug_batch8_codecheck/checkpoint-183",
   "epoch": 1.5,
   "eval_steps": 500,
   "global_step": 366,
@@ -10,59 +10,59 @@
   "log_history": [
     {
       "epoch": 0.273224043715847,
-      "grad_norm": 3.3262431621551514,
+      "grad_norm": 0.9716752171516418,
       "learning_rate": 4.0425531914893614e-05,
-      "loss": 1.0449,
+      "loss": 1.0336,
       "step": 100
     },
     {
       "epoch": 0.5,
-      "eval_accuracy": 0.9938650306748467,
-      "eval_loss": 0.017213916406035423,
-      "eval_runtime": 55.9924,
-      "eval_samples_per_second": 2.911,
-      "eval_steps_per_second": 0.375,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.010333087295293808,
+      "eval_runtime": 54.3545,
+      "eval_samples_per_second": 2.999,
+      "eval_steps_per_second": 0.386,
       "step": 183
     },
     {
       "epoch": 1.046448087431694,
-      "grad_norm": 0.36886924505233765,
+      "grad_norm": 0.0508948378264904,
       "learning_rate": 2.5227963525835867e-05,
-      "loss": 0.038,
+      "loss": 0.0556,
       "step": 200
     },
     {
       "epoch": 1.319672131147541,
-      "grad_norm": 0.04929419606924057,
+      "grad_norm": 0.0559578500688076,
       "learning_rate": 1.0030395136778116e-05,
-      "loss": 0.0056,
+      "loss": 0.0087,
       "step": 300
     },
     {
       "epoch": 1.5,
       "eval_accuracy": 1.0,
-      "eval_loss": 0.0024167420342564583,
-      "eval_runtime": 62.5343,
-      "eval_samples_per_second": 2.607,
-      "eval_steps_per_second": 0.336,
+      "eval_loss": 0.004173776134848595,
+      "eval_runtime": 53.366,
+      "eval_samples_per_second": 3.054,
+      "eval_steps_per_second": 0.394,
       "step": 366
     },
     {
       "epoch": 1.5,
       "step": 366,
       "total_flos": 2.5655220082940314e+18,
-      "train_loss": 0.2977769932623118,
-      "train_runtime": 1336.9205,
-      "train_samples_per_second": 2.19,
-      "train_steps_per_second": 0.274
+      "train_loss": 0.30036494380137957,
+      "train_runtime": 1349.0891,
+      "train_samples_per_second": 2.17,
+      "train_steps_per_second": 0.271
     },
     {
       "epoch": 1.5,
-      "eval_accuracy": 0.7928286852589641,
-      "eval_loss": 0.758155345916748,
-      "eval_runtime": 81.0092,
-      "eval_samples_per_second": 3.098,
-      "eval_steps_per_second": 0.395,
+      "eval_accuracy": 0.7808764940239044,
+      "eval_loss": 0.7022287845611572,
+      "eval_runtime": 62.6519,
+      "eval_samples_per_second": 4.006,
+      "eval_steps_per_second": 0.511,
       "step": 366
     }
   ],