Model save

README.md

@@ -18,13 +18,13 @@ should probably proofread and complete it, then remove this comment. -->
 
 # py38-pylingual-v1.1.1-segmenter
 
-This model is a fine-tuned version of [syssec-utd/py38-pylingual-v1.1.1-mlm](https://huggingface.co/syssec-utd/py38-pylingual-v1.1.1-mlm) on the syssec-utd/segmentation-py38-pylingual-v1.1-tokenized dataset.
+This model is a fine-tuned version of [syssec-utd/py38-pylingual-v1.1.1-mlm](https://huggingface.co/syssec-utd/py38-pylingual-v1.1.1-mlm) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0614
-- Precision: 0.8780
-- Recall: 0.8852
-- F1: 0.8816
-- Accuracy: 0.9694
+- Loss: 0.0811
+- Precision: 0.9316
+- Recall: 0.8862
+- F1: 0.9083
+- Accuracy: 0.9617
 
 ## Model description
 
@@ -51,7 +51,7 @@ The following hyperparameters were used during training:
 - num_devices: 3
 - total_train_batch_size: 144
 - total_eval_batch_size: 24
-- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- 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
 - num_epochs: 2
 - mixed_precision_training: Native AMP
@@ -60,13 +60,13 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| 0.2647        | 1.0   | 1490 | 0.1043          | 0.7752    | 0.8197 | 0.7968 | 0.9480   |
-| 0.1113        | 2.0   | 2980 | 0.0614          | 0.8780    | 0.8852 | 0.8816 | 0.9694   |
+| 0.2656        | 1.0   | 1490 | 0.1329          | 0.8099    | 0.7967 | 0.8033 | 0.9387   |
+| 0.1195        | 2.0   | 2980 | 0.0811          | 0.9316    | 0.8862 | 0.9083 | 0.9617   |
 
 
 ### Framework versions
 
-- Transformers 4.46.1
-- Pytorch 2.4.1+cu121
-- Datasets 3.0.1
-- Tokenizers 0.20.3
+- Transformers 4.54.1
+- Pytorch 2.8.0+cu128
+- Datasets 3.3.2
+- Tokenizers 0.21.4