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: 0.7724
-- Accuracy: 0.8111
+- Loss: 0.8394
+- Accuracy: 0.8017
 
 ## Model description
 
@@ -45,14 +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: 1002
+- training_steps: 1004
 
 ### Training results
 
-| 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   |
+| Training Loss | Epoch | Step | Validation Loss | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|
+| 0.3785        | 0.5   | 502  | 0.3069          | 0.9479   |
+| 0.0487        | 1.5   | 1004 | 0.0754          | 0.9883   |
 
 
 ### Framework versions

all_results.json

@@ -1,8 +1,8 @@
 {
-    "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
+    "epoch": 1.5,
+    "eval_accuracy": 0.8017241379310345,
+    "eval_loss": 0.8394390940666199,
+    "eval_runtime": 424.8162,
+    "eval_samples_per_second": 3.004,
+    "eval_steps_per_second": 0.377
 }

test_results.json

@@ -1,8 +1,8 @@
 {
-    "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
+    "epoch": 1.5,
+    "eval_accuracy": 0.8017241379310345,
+    "eval_loss": 0.8394390940666199,
+    "eval_runtime": 424.8162,
+    "eval_samples_per_second": 3.004,
+    "eval_steps_per_second": 0.377
 }

trainer_state.json

@@ -1,122 +1,122 @@
 {
-  "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,
+  "best_metric": 0.988293897882939,
+  "best_model_checkpoint": "/media/cse/HDD/Shawon/shawon/10 fold timesformer/Timesformer_default_fold_10_10_epoch_noAug_batch8_codecheck/checkpoint-1004",
+  "epoch": 1.5,
   "eval_steps": 500,
-  "global_step": 1002,
+  "global_step": 1004,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.0998003992015968,
-      "grad_norm": 12.41899585723877,
+      "epoch": 0.099601593625498,
+      "grad_norm": 12.111989974975586,
       "learning_rate": 4.950495049504951e-05,
-      "loss": 3.8639,
+      "loss": 3.7788,
       "step": 100
     },
     {
-      "epoch": 0.1996007984031936,
-      "grad_norm": 11.086575508117676,
-      "learning_rate": 4.4506104328523865e-05,
-      "loss": 2.3042,
+      "epoch": 0.199203187250996,
+      "grad_norm": 10.870170593261719,
+      "learning_rate": 4.451827242524917e-05,
+      "loss": 2.1601,
       "step": 200
     },
     {
-      "epoch": 0.2994011976047904,
-      "grad_norm": 12.340323448181152,
-      "learning_rate": 3.895671476137625e-05,
-      "loss": 1.027,
+      "epoch": 0.29880478087649404,
+      "grad_norm": 9.179058074951172,
+      "learning_rate": 3.89811738648948e-05,
+      "loss": 1.1244,
       "step": 300
     },
     {
-      "epoch": 0.3992015968063872,
-      "grad_norm": 6.579195499420166,
-      "learning_rate": 3.340732519422864e-05,
-      "loss": 0.6417,
+      "epoch": 0.398406374501992,
+      "grad_norm": 2.3942956924438477,
+      "learning_rate": 3.344407530454042e-05,
+      "loss": 0.5512,
       "step": 400
     },
     {
-      "epoch": 0.499001996007984,
-      "grad_norm": 8.828213691711426,
-      "learning_rate": 2.785793562708102e-05,
-      "loss": 0.3111,
+      "epoch": 0.49800796812749004,
+      "grad_norm": 3.746593952178955,
+      "learning_rate": 2.7906976744186048e-05,
+      "loss": 0.3785,
       "step": 500
     },
     {
-      "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,
+      "epoch": 0.5,
+      "eval_accuracy": 0.947945205479452,
+      "eval_loss": 0.30693259835243225,
+      "eval_runtime": 1802.7473,
+      "eval_samples_per_second": 2.227,
+      "eval_steps_per_second": 0.278,
       "step": 502
     },
     {
-      "epoch": 1.0978043912175648,
-      "grad_norm": 7.735666275024414,
-      "learning_rate": 2.230854605993341e-05,
-      "loss": 0.1745,
+      "epoch": 1.097609561752988,
+      "grad_norm": 3.173877716064453,
+      "learning_rate": 2.2369878183831673e-05,
+      "loss": 0.1585,
       "step": 600
     },
     {
-      "epoch": 1.1976047904191618,
-      "grad_norm": 1.334542155265808,
-      "learning_rate": 1.6759156492785795e-05,
-      "loss": 0.1142,
+      "epoch": 1.197211155378486,
+      "grad_norm": 3.913566827774048,
+      "learning_rate": 1.6832779623477298e-05,
+      "loss": 0.1062,
       "step": 700
     },
     {
-      "epoch": 1.2974051896207586,
-      "grad_norm": 6.124719619750977,
-      "learning_rate": 1.120976692563818e-05,
-      "loss": 0.0919,
+      "epoch": 1.296812749003984,
+      "grad_norm": 0.6361425518989563,
+      "learning_rate": 1.1295681063122925e-05,
+      "loss": 0.1037,
       "step": 800
     },
     {
-      "epoch": 1.3972055888223553,
-      "grad_norm": 7.422993183135986,
-      "learning_rate": 5.660377358490566e-06,
-      "loss": 0.0588,
+      "epoch": 1.3964143426294822,
+      "grad_norm": 0.8113210797309875,
+      "learning_rate": 5.75858250276855e-06,
+      "loss": 0.0565,
       "step": 900
     },
     {
-      "epoch": 1.4970059880239521,
-      "grad_norm": 1.5655475854873657,
-      "learning_rate": 1.1098779134295228e-07,
-      "loss": 0.0629,
+      "epoch": 1.49601593625498,
+      "grad_norm": 0.3181760609149933,
+      "learning_rate": 2.2148394241417497e-07,
+      "loss": 0.0487,
       "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": 1.5,
+      "eval_accuracy": 0.988293897882939,
+      "eval_loss": 0.07536407560110092,
+      "eval_runtime": 1562.8471,
+      "eval_samples_per_second": 2.569,
+      "eval_steps_per_second": 0.321,
+      "step": 1004
     },
     {
-      "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": 1.5,
+      "step": 1004,
+      "total_flos": 7.039803615017435e+18,
+      "train_loss": 0.8433415542145174,
+      "train_runtime": 7467.2826,
+      "train_samples_per_second": 1.076,
+      "train_steps_per_second": 0.134
     },
     {
-      "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
+      "epoch": 1.5,
+      "eval_accuracy": 0.8017241379310345,
+      "eval_loss": 0.8394390940666199,
+      "eval_runtime": 424.8162,
+      "eval_samples_per_second": 3.004,
+      "eval_steps_per_second": 0.377,
+      "step": 1004
     }
   ],
   "logging_steps": 100,
-  "max_steps": 1002,
+  "max_steps": 1004,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 9223372036854775807,
   "save_steps": 500,
@@ -132,7 +132,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 7.024903632266527e+18,
+  "total_flos": 7.039803615017435e+18,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null