End of training

README.md

@@ -18,7 +18,7 @@ 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: 0.7852
+- Loss: 0.8131
 - Accuracy: 0.7849
 
 ## Model description
@@ -51,7 +51,7 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.935         | 1.0   | 101  | 0.1722          | 0.9840   |
+| 0.8303        | 1.0   | 101  | 0.1456          | 0.9902   |
 
 
 ### Framework versions

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
     "eval_accuracy": 0.7848605577689243,
-    "eval_loss": 0.7851973176002502,
-    "eval_runtime": 60.2304,
-    "eval_samples_per_second": 4.167,
-    "eval_steps_per_second": 0.531
+    "eval_loss": 0.8131076693534851,
+    "eval_runtime": 56.6718,
+    "eval_samples_per_second": 4.429,
+    "eval_steps_per_second": 0.565
 }

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
     "eval_accuracy": 0.7848605577689243,
-    "eval_loss": 0.7851973176002502,
-    "eval_runtime": 60.2304,
-    "eval_samples_per_second": 4.167,
-    "eval_steps_per_second": 0.531
+    "eval_loss": 0.8131076693534851,
+    "eval_runtime": 56.6718,
+    "eval_samples_per_second": 4.429,
+    "eval_steps_per_second": 0.565
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.984029484029484,
+  "best_metric": 0.990159901599016,
   "best_model_checkpoint": "/media/cse/HDD/Shawon/shawon/10 fold timesformer/Timesformer_default_fold_10_10_epoch_noAug_batch8_codecheck/checkpoint-101",
   "epoch": 1.0,
   "eval_steps": 500,
@@ -10,36 +10,36 @@
   "log_history": [
     {
       "epoch": 0.9900990099009901,
-      "grad_norm": 2.462575912475586,
+      "grad_norm": 7.038530349731445,
       "learning_rate": 5.555555555555556e-07,
-      "loss": 0.935,
+      "loss": 0.8303,
       "step": 100
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.984029484029484,
-      "eval_loss": 0.1721714735031128,
-      "eval_runtime": 182.5078,
-      "eval_samples_per_second": 4.46,
-      "eval_steps_per_second": 0.559,
+      "eval_accuracy": 0.990159901599016,
+      "eval_loss": 0.14564795792102814,
+      "eval_runtime": 181.7345,
+      "eval_samples_per_second": 4.474,
+      "eval_steps_per_second": 0.561,
       "step": 101
     },
     {
       "epoch": 1.0,
       "step": 101,
       "total_flos": 7.07971920321577e+17,
-      "train_loss": 0.9269423101207998,
-      "train_runtime": 445.5113,
-      "train_samples_per_second": 1.814,
+      "train_loss": 0.8236100951043686,
+      "train_runtime": 444.4391,
+      "train_samples_per_second": 1.818,
       "train_steps_per_second": 0.227
     },
     {
       "epoch": 1.0,
       "eval_accuracy": 0.7848605577689243,
-      "eval_loss": 0.7851973176002502,
-      "eval_runtime": 60.2304,
-      "eval_samples_per_second": 4.167,
-      "eval_steps_per_second": 0.531,
+      "eval_loss": 0.8131076693534851,
+      "eval_runtime": 56.6718,
+      "eval_samples_per_second": 4.429,
+      "eval_steps_per_second": 0.565,
       "step": 101
     }
   ],