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: 1.2463
-- Accuracy: 0.7241
+- Loss: 0.7724
+- Accuracy: 0.8111
 
 ## Model description
 
@@ -45,13 +45,14 @@ 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: 501
+- training_steps: 1002
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Accuracy |
-|:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.4047        | 1.0   | 501  | 0.4285          | 0.9604   |
+| Training Loss | Epoch  | Step | Validation Loss | Accuracy |
+|:-------------:|:------:|:----:|:---------------:|:--------:|
+| 0.3111        | 0.5010 | 502  | 0.2995          | 0.9457   |
+| 0.0629        | 1.4990 | 1002 | 0.0642          | 0.9918   |
 
 
 ### Framework versions

all_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 1.0,
-    "eval_accuracy": 0.7241379310344828,
-    "eval_loss": 1.246259093284607,
-    "eval_runtime": 500.0138,
-    "eval_samples_per_second": 2.552,
-    "eval_steps_per_second": 0.32
+    "epoch": 1.499001996007984,
+    "eval_accuracy": 0.8111285266457681,
+    "eval_loss": 0.7723951935768127,
+    "eval_runtime": 509.1163,
+    "eval_samples_per_second": 2.506,
+    "eval_steps_per_second": 0.314
 }

test_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 1.0,
-    "eval_accuracy": 0.7241379310344828,
-    "eval_loss": 1.246259093284607,
-    "eval_runtime": 500.0138,
-    "eval_samples_per_second": 2.552,
-    "eval_steps_per_second": 0.32
+    "epoch": 1.499001996007984,
+    "eval_accuracy": 0.8111285266457681,
+    "eval_loss": 0.7723951935768127,
+    "eval_runtime": 509.1163,
+    "eval_samples_per_second": 2.506,
+    "eval_steps_per_second": 0.314
 }

trainer_state.json

@@ -1,78 +1,122 @@
 {
-  "best_metric": 0.9604083665338645,
-  "best_model_checkpoint": "/media/cse/HDD/Shawon/shawon/10 fold timesformer/Timesformer_default_fold_10_10_epoch_noAug_batch8_codecheck/checkpoint-501",
-  "epoch": 1.0,
+  "best_metric": 0.9917828685258964,
+  "best_model_checkpoint": "/media/cse/HDD/Shawon/shawon/10 fold timesformer/Timesformer_default_fold_10_10_epoch_noAug_batch8_codecheck/checkpoint-1002",
+  "epoch": 1.499001996007984,
   "eval_steps": 500,
-  "global_step": 501,
+  "global_step": 1002,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.1996007984031936,
-      "grad_norm": 12.420443534851074,
-      "learning_rate": 4.4555555555555555e-05,
-      "loss": 3.6925,
+      "epoch": 0.0998003992015968,
+      "grad_norm": 12.41899585723877,
+      "learning_rate": 4.950495049504951e-05,
+      "loss": 3.8639,
       "step": 100
     },
     {
-      "epoch": 0.3992015968063872,
-      "grad_norm": 10.26618480682373,
-      "learning_rate": 3.3444444444444443e-05,
-      "loss": 2.0447,
+      "epoch": 0.1996007984031936,
+      "grad_norm": 11.086575508117676,
+      "learning_rate": 4.4506104328523865e-05,
+      "loss": 2.3042,
       "step": 200
     },
     {
-      "epoch": 0.5988023952095808,
-      "grad_norm": 11.14834976196289,
-      "learning_rate": 2.2333333333333335e-05,
-      "loss": 1.0555,
+      "epoch": 0.2994011976047904,
+      "grad_norm": 12.340323448181152,
+      "learning_rate": 3.895671476137625e-05,
+      "loss": 1.027,
       "step": 300
     },
     {
-      "epoch": 0.7984031936127745,
-      "grad_norm": 6.654690265655518,
-      "learning_rate": 1.1222222222222224e-05,
-      "loss": 0.6462,
+      "epoch": 0.3992015968063872,
+      "grad_norm": 6.579195499420166,
+      "learning_rate": 3.340732519422864e-05,
+      "loss": 0.6417,
       "step": 400
     },
     {
-      "epoch": 0.998003992015968,
-      "grad_norm": 6.689345836639404,
-      "learning_rate": 1.1111111111111112e-07,
-      "loss": 0.4047,
+      "epoch": 0.499001996007984,
+      "grad_norm": 8.828213691711426,
+      "learning_rate": 2.785793562708102e-05,
+      "loss": 0.3111,
       "step": 500
     },
     {
-      "epoch": 1.0,
-      "eval_accuracy": 0.9604083665338645,
-      "eval_loss": 0.4285348355770111,
-      "eval_runtime": 1730.4695,
-      "eval_samples_per_second": 2.321,
-      "eval_steps_per_second": 0.29,
-      "step": 501
+      "epoch": 0.500998003992016,
+      "eval_accuracy": 0.9457171314741036,
+      "eval_loss": 0.2995070815086365,
+      "eval_runtime": 1755.2973,
+      "eval_samples_per_second": 2.288,
+      "eval_steps_per_second": 0.286,
+      "step": 502
+    },
+    {
+      "epoch": 1.0978043912175648,
+      "grad_norm": 7.735666275024414,
+      "learning_rate": 2.230854605993341e-05,
+      "loss": 0.1745,
+      "step": 600
+    },
+    {
+      "epoch": 1.1976047904191618,
+      "grad_norm": 1.334542155265808,
+      "learning_rate": 1.6759156492785795e-05,
+      "loss": 0.1142,
+      "step": 700
+    },
+    {
+      "epoch": 1.2974051896207586,
+      "grad_norm": 6.124719619750977,
+      "learning_rate": 1.120976692563818e-05,
+      "loss": 0.0919,
+      "step": 800
+    },
+    {
+      "epoch": 1.3972055888223553,
+      "grad_norm": 7.422993183135986,
+      "learning_rate": 5.660377358490566e-06,
+      "loss": 0.0588,
+      "step": 900
+    },
+    {
+      "epoch": 1.4970059880239521,
+      "grad_norm": 1.5655475854873657,
+      "learning_rate": 1.1098779134295228e-07,
+      "loss": 0.0629,
+      "step": 1000
+    },
+    {
+      "epoch": 1.499001996007984,
+      "eval_accuracy": 0.9917828685258964,
+      "eval_loss": 0.06420521438121796,
+      "eval_runtime": 1671.9392,
+      "eval_samples_per_second": 2.402,
+      "eval_steps_per_second": 0.3,
+      "step": 1002
     },
     {
-      "epoch": 1.0,
-      "step": 501,
-      "total_flos": 3.512890050920055e+18,
-      "train_loss": 1.5669060153161694,
-      "train_runtime": 3832.4277,
-      "train_samples_per_second": 1.046,
-      "train_steps_per_second": 0.131
+      "epoch": 1.499001996007984,
+      "step": 1002,
+      "total_flos": 7.024903632266527e+18,
+      "train_loss": 0.863516083377564,
+      "train_runtime": 7438.8722,
+      "train_samples_per_second": 1.078,
+      "train_steps_per_second": 0.135
     },
     {
-      "epoch": 1.0,
-      "eval_accuracy": 0.7241379310344828,
-      "eval_loss": 1.246259093284607,
-      "eval_runtime": 500.0138,
-      "eval_samples_per_second": 2.552,
-      "eval_steps_per_second": 0.32,
-      "step": 501
+      "epoch": 1.499001996007984,
+      "eval_accuracy": 0.8111285266457681,
+      "eval_loss": 0.7723951935768127,
+      "eval_runtime": 509.1163,
+      "eval_samples_per_second": 2.506,
+      "eval_steps_per_second": 0.314,
+      "step": 1002
     }
   ],
   "logging_steps": 100,
-  "max_steps": 501,
+  "max_steps": 1002,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 9223372036854775807,
   "save_steps": 500,
@@ -88,7 +132,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 3.512890050920055e+18,
+  "total_flos": 7.024903632266527e+18,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null