Update README.md

README.md

@@ -15,29 +15,17 @@ This model is a fine-tuned version of [vinai/bertweet-covid19-base-uncased](http
 
 The model is intended to identify skepticism of COVID-19 policies (i.e. masks, social distancing, lockdowns, vaccines etc.). The model classifies as 0 (expressing skepticism/opposition to a COVID-19 policy or 1 (no opposition)
 
-It achieves the following results on the evaluation set:
-- Train Loss: 0.1007
-- Train Sparse Categorical Accuracy: 0.9591
-- Validation Loss: 0.0913
-- Validation Sparse Categorical Accuracy: 0.9627
-- Epoch: 3
 
+It's a pretty simple task but I used a grid search to optimize hyperparameters. The final model is achieves the following results and uses the following hyperparamters: 
 
-The following hyperparameters were used during training:
-- optimizer: {'name': 'Adam', 'learning_rate': 5e-07, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
-- training_precision: float32
+{'train_runtime': 174.3258, 'train_samples_per_second': 18.896, 'train_steps_per_second': 2.375, 'train_loss': 0.1576320076910194, 'epoch': 6.0}
+{'eval_loss': 0.8522606492042542, 'eval_runtime': 3.8368, 'eval_samples_per_second': 70.111, 'eval_steps_per_second': 8.862, 'epoch': 6.0}
 
-### Training results
+Optimized Hyperparameters
+----------------------------------------------------------------------------------------------------
+The best learning rate is: 5.4761828368201554e-05
+The best weight decay is: 0.0003655991822889909
+The best epoch is : 6
+The best train split is : 0.3284489429375188
 
-| Train Loss | Train Sparse Categorical Accuracy | Validation Loss | Validation Sparse Categorical Accuracy | Epoch |
-|:----------:|:---------------------------------:|:---------------:|:--------------------------------------:|:-----:|
-| 0.1822     | 0.9345                            | 0.1021          | 0.9584                                 | 0     |
-| 0.1007     | 0.9591                            | 0.0913          | 0.9627                                 | 1     |
 
-
-### Framework versions
-
-- Transformers 4.21.0
-- TensorFlow 2.8.2
-- Datasets 2.4.0
-- Tokenizers 0.12.1