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: 0.5098
-- Accuracy: 0.8367
+- Loss: 0.9561
+- Accuracy: 0.6773
 
 ## Model description
 
@@ -45,14 +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: 366
+- training_steps: 101
 
 ### Training results
 
-| Training Loss | Epoch  | Step | Validation Loss | Accuracy |
-|:-------------:|:------:|:----:|:---------------:|:--------:|
-| 1.0933        | 0.5027 | 184  | 0.0142          | 1.0      |
-| 0.011         | 1.4973 | 366  | 0.0109          | 0.9938   |
+| Training Loss | Epoch | Step | Validation Loss | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|
+| 0.8886        | 1.0   | 101  | 0.1661          | 0.9828   |
 
 
 ### Framework versions

all_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 1.4972677595628414,
-    "eval_accuracy": 0.8366533864541833,
-    "eval_loss": 0.5097954869270325,
-    "eval_runtime": 56.3553,
-    "eval_samples_per_second": 4.454,
-    "eval_steps_per_second": 0.568
+    "epoch": 1.0,
+    "eval_accuracy": 0.6772908366533864,
+    "eval_loss": 0.9560578465461731,
+    "eval_runtime": 55.4125,
+    "eval_samples_per_second": 4.53,
+    "eval_steps_per_second": 0.577
 }

test_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 1.4972677595628414,
-    "eval_accuracy": 0.8366533864541833,
-    "eval_loss": 0.5097954869270325,
-    "eval_runtime": 56.3553,
-    "eval_samples_per_second": 4.454,
-    "eval_steps_per_second": 0.568
+    "epoch": 1.0,
+    "eval_accuracy": 0.6772908366533864,
+    "eval_loss": 0.9560578465461731,
+    "eval_runtime": 55.4125,
+    "eval_samples_per_second": 4.53,
+    "eval_steps_per_second": 0.577
 }

trainer_state.json

@@ -1,73 +1,50 @@
 {
-  "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-184",
-  "epoch": 1.4972677595628414,
+  "best_metric": 0.9828009828009828,
+  "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,
-  "global_step": 366,
+  "global_step": 101,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.273224043715847,
-      "grad_norm": 2.266233444213867,
-      "learning_rate": 4.0425531914893614e-05,
-      "loss": 1.0933,
+      "epoch": 0.9900990099009901,
+      "grad_norm": 2.9471566677093506,
+      "learning_rate": 5.555555555555556e-07,
+      "loss": 0.8886,
       "step": 100
     },
     {
-      "epoch": 0.5027322404371585,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.014155284501612186,
-      "eval_runtime": 37.8295,
-      "eval_samples_per_second": 4.282,
-      "eval_steps_per_second": 0.555,
-      "step": 184
+      "epoch": 1.0,
+      "eval_accuracy": 0.9828009828009828,
+      "eval_loss": 0.16609854996204376,
+      "eval_runtime": 202.8409,
+      "eval_samples_per_second": 4.013,
+      "eval_steps_per_second": 0.503,
+      "step": 101
     },
     {
-      "epoch": 1.0437158469945356,
-      "grad_norm": 0.3162294626235962,
-      "learning_rate": 2.5227963525835867e-05,
-      "loss": 0.0458,
-      "step": 200
+      "epoch": 1.0,
+      "step": 101,
+      "total_flos": 7.07971920321577e+17,
+      "train_loss": 0.8806298740281917,
+      "train_runtime": 488.3542,
+      "train_samples_per_second": 1.655,
+      "train_steps_per_second": 0.207
     },
     {
-      "epoch": 1.3169398907103824,
-      "grad_norm": 0.509850263595581,
-      "learning_rate": 1.0030395136778116e-05,
-      "loss": 0.011,
-      "step": 300
-    },
-    {
-      "epoch": 1.4972677595628414,
-      "eval_accuracy": 0.9938271604938271,
-      "eval_loss": 0.010920087806880474,
-      "eval_runtime": 38.0242,
-      "eval_samples_per_second": 4.26,
-      "eval_steps_per_second": 0.552,
-      "step": 366
-    },
-    {
-      "epoch": 1.4972677595628414,
-      "step": 366,
-      "total_flos": 2.5593885881922355e+18,
-      "train_loss": 0.3163567338484884,
-      "train_runtime": 953.4888,
-      "train_samples_per_second": 3.071,
-      "train_steps_per_second": 0.384
-    },
-    {
-      "epoch": 1.4972677595628414,
-      "eval_accuracy": 0.8366533864541833,
-      "eval_loss": 0.5097954869270325,
-      "eval_runtime": 56.3553,
-      "eval_samples_per_second": 4.454,
-      "eval_steps_per_second": 0.568,
-      "step": 366
+      "epoch": 1.0,
+      "eval_accuracy": 0.6772908366533864,
+      "eval_loss": 0.9560578465461731,
+      "eval_runtime": 55.4125,
+      "eval_samples_per_second": 4.53,
+      "eval_steps_per_second": 0.577,
+      "step": 101
     }
   ],
   "logging_steps": 100,
-  "max_steps": 366,
+  "max_steps": 101,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 9223372036854775807,
   "save_steps": 500,
@@ -83,7 +60,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.5593885881922355e+18,
+  "total_flos": 7.07971920321577e+17,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null