End of training

README.md

@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [judy93536/distilroberta-newsapi121k](https://huggingface.co/judy93536/distilroberta-newsapi121k) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1255
-- Accuracy: 0.9520
+- Loss: 0.1848
+- Accuracy: 0.9379
 
 ## Model description
 
@@ -37,22 +37,23 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 4.545606711088675e-05
+- learning_rate: 8.103335054745316e-06
 - train_batch_size: 32
-- eval_batch_size: 8
+- eval_batch_size: 32
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.28
-- num_epochs: 2
+- num_epochs: 3
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| No log        | 1.0   | 125  | 0.2073          | 0.9219   |
-| No log        | 2.0   | 250  | 0.1255          | 0.9520   |
+| No log        | 1.0   | 125  | 0.5800          | 0.7568   |
+| No log        | 2.0   | 250  | 0.2335          | 0.9219   |
+| No log        | 3.0   | 375  | 0.1848          | 0.9379   |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a0c230fd97b880c1bafc77b7f1d6789e4db178f71e24d6663bdad35e874bce89
+oid sha256:0ed6f80b6185f15fdf1faaa10b4fe53202883c74f01ba2333b5a4933a5704a76
 size 328495356

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bb033fb41f868407d4ad957458f4f0a338b785cd058400835a7dbac3bc0df573
+oid sha256:b74cc65bd4e6ffd89d30e2a346669d79b91755339972bc332710107b1185bdd1
 size 4536