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: 1.1312
-- Accuracy: 0.7304
+- Loss: 1.2304
+- Accuracy: 0.7288
 
 ## Model description
 
@@ -45,13 +45,13 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
-- training_steps: 669
+- training_steps: 502
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.1824        | 1.0   | 669  | 0.0990          | 0.9933   |
+| 0.3006        | 1.0   | 502  | 0.2494          | 0.9806   |
 
 
 ### Framework versions

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_accuracy": 0.7304075235109718,
-    "eval_loss": 1.1312494277954102,
-    "eval_runtime": 377.402,
-    "eval_samples_per_second": 3.381,
-    "eval_steps_per_second": 0.424
+    "eval_accuracy": 0.7288401253918495,
+    "eval_loss": 1.230407953262329,
+    "eval_runtime": 286.6871,
+    "eval_samples_per_second": 4.451,
+    "eval_steps_per_second": 0.558
 }

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_accuracy": 0.7304075235109718,
-    "eval_loss": 1.1312494277954102,
-    "eval_runtime": 377.402,
-    "eval_samples_per_second": 3.381,
-    "eval_steps_per_second": 0.424
+    "eval_accuracy": 0.7288401253918495,
+    "eval_loss": 1.230407953262329,
+    "eval_runtime": 286.6871,
+    "eval_samples_per_second": 4.451,
+    "eval_steps_per_second": 0.558
 }

trainer_state.json

@@ -1,85 +1,78 @@
 {
-  "best_metric": 0.9932768924302788,
-  "best_model_checkpoint": "/media/cse/HDD/Shawon/shawon/10 fold timesformer/Timesformer_default_fold_10_10_epoch_noAug_batch8_codecheck/checkpoint-669",
+  "best_metric": 0.9805728518057285,
+  "best_model_checkpoint": "/media/cse/HDD/Shawon/shawon/10 fold timesformer/Timesformer_default_fold_10_10_epoch_noAug_batch8_codecheck/checkpoint-502",
   "epoch": 1.0,
   "eval_steps": 500,
-  "global_step": 669,
+  "global_step": 502,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.14947683109118087,
-      "grad_norm": 11.951628684997559,
-      "learning_rate": 4.725913621262459e-05,
-      "loss": 3.7222,
+      "epoch": 0.199203187250996,
+      "grad_norm": 11.452275276184082,
+      "learning_rate": 4.456762749445677e-05,
+      "loss": 3.5774,
       "step": 100
     },
     {
-      "epoch": 0.29895366218236175,
-      "grad_norm": 10.303458213806152,
-      "learning_rate": 3.895348837209303e-05,
-      "loss": 1.9162,
+      "epoch": 0.398406374501992,
+      "grad_norm": 10.499092102050781,
+      "learning_rate": 3.348115299334812e-05,
+      "loss": 1.8278,
       "step": 200
     },
     {
-      "epoch": 0.4484304932735426,
-      "grad_norm": 7.89475154876709,
-      "learning_rate": 3.064784053156146e-05,
-      "loss": 0.971,
+      "epoch": 0.5976095617529881,
+      "grad_norm": 7.440881729125977,
+      "learning_rate": 2.239467849223947e-05,
+      "loss": 0.9037,
       "step": 300
     },
     {
-      "epoch": 0.5979073243647235,
-      "grad_norm": 3.1657724380493164,
-      "learning_rate": 2.23421926910299e-05,
-      "loss": 0.478,
+      "epoch": 0.796812749003984,
+      "grad_norm": 2.855297088623047,
+      "learning_rate": 1.130820399113082e-05,
+      "loss": 0.4349,
       "step": 400
     },
     {
-      "epoch": 0.7473841554559043,
-      "grad_norm": 2.475733757019043,
-      "learning_rate": 1.403654485049834e-05,
-      "loss": 0.2504,
+      "epoch": 0.9960159362549801,
+      "grad_norm": 3.7258729934692383,
+      "learning_rate": 2.2172949002217296e-07,
+      "loss": 0.3006,
       "step": 500
     },
-    {
-      "epoch": 0.8968609865470852,
-      "grad_norm": 5.244026184082031,
-      "learning_rate": 5.730897009966778e-06,
-      "loss": 0.1824,
-      "step": 600
-    },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.9932768924302788,
-      "eval_loss": 0.09897388517856598,
-      "eval_runtime": 887.6237,
-      "eval_samples_per_second": 4.524,
-      "eval_steps_per_second": 0.566,
-      "step": 669
+      "eval_accuracy": 0.9805728518057285,
+      "eval_loss": 0.24935545027256012,
+      "eval_runtime": 1419.7674,
+      "eval_samples_per_second": 2.828,
+      "eval_steps_per_second": 0.354,
+      "step": 502
     },
     {
       "epoch": 1.0,
-      "step": 669,
-      "total_flos": 4.690865157815402e+18,
-      "train_loss": 1.136438287070692,
-      "train_runtime": 2516.9818,
-      "train_samples_per_second": 2.126,
-      "train_steps_per_second": 0.266
+      "step": 502,
+      "total_flos": 3.5199018075087176e+18,
+      "train_loss": 1.403769117427537,
+      "train_runtime": 3097.7448,
+      "train_samples_per_second": 1.296,
+      "train_steps_per_second": 0.162
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7304075235109718,
-      "eval_loss": 1.1312494277954102,
-      "eval_runtime": 377.402,
-      "eval_samples_per_second": 3.381,
-      "eval_steps_per_second": 0.424,
-      "step": 669
+      "eval_accuracy": 0.7288401253918495,
+      "eval_loss": 1.230407953262329,
+      "eval_runtime": 286.6871,
+      "eval_samples_per_second": 4.451,
+      "eval_steps_per_second": 0.558,
+      "step": 502
     }
   ],
   "logging_steps": 100,
-  "max_steps": 669,
+  "max_steps": 502,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 9223372036854775807,
   "save_steps": 500,
@@ -95,7 +88,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 4.690865157815402e+18,
+  "total_flos": 3.5199018075087176e+18,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null