Completed Filter Training With Full TP/FP/FN Distribution

README.md

@@ -20,11 +20,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/deberta-v3-small](https://huggingface.co/microsoft/deberta-v3-small) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1454
-- F1: 0.9819
-- Recall: 0.9854
-- Precision: 0.9784
-- Trash Caught: 0.5550
+- Loss: 0.1328
+- F1: 0.9811
+- Recall: 0.9969
+- Precision: 0.9658
+- Trash Caught: 0.2798
 
 ## Model description
 
@@ -43,9 +43,9 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 2e-05
-- train_batch_size: 16
-- eval_batch_size: 32
+- learning_rate: 1e-05
+- train_batch_size: 32
+- eval_batch_size: 64
 - seed: 42
 - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
@@ -55,11 +55,11 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | F1     | Recall | Precision | Trash Caught |
 |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:---------:|:------------:|
-| No log        | 1.0   | 499  | 0.1384          | 0.9777 | 0.9926 | 0.9632    | 0.2248       |
-| 0.3028        | 2.0   | 998  | 0.1125          | 0.9834 | 0.9948 | 0.9721    | 0.4174       |
-| 0.1833        | 3.0   | 1497 | 0.1465          | 0.9782 | 0.9782 | 0.9782    | 0.5550       |
-| 0.1336        | 4.0   | 1996 | 0.1366          | 0.9841 | 0.9917 | 0.9766    | 0.5138       |
-| 0.0987        | 5.0   | 2495 | 0.1454          | 0.9819 | 0.9854 | 0.9784    | 0.5550       |
+| No log        | 1.0   | 250  | 0.1488          | 0.9767 | 0.9978 | 0.9565    | 0.0734       |
+| 0.3046        | 2.0   | 500  | 0.1534          | 0.9774 | 0.9996 | 0.9562    | 0.0642       |
+| 0.3046        | 3.0   | 750  | 0.1259          | 0.9800 | 0.9969 | 0.9638    | 0.2339       |
+| 0.1678        | 4.0   | 1000 | 0.1344          | 0.9806 | 0.9973 | 0.9644    | 0.2477       |
+| 0.1678        | 5.0   | 1250 | 0.1328          | 0.9811 | 0.9969 | 0.9658    | 0.2798       |
 
 
 ### Framework versions