tparng/roberta-base-lora-text-classification

README.md

@@ -1,11 +1,10 @@
 ---
 license: mit
-library_name: peft
+base_model: roberta-base
 tags:
 - generated_from_trainer
 metrics:
 - accuracy
-base_model: roberta-base
 model-index:
 - name: roberta-base-lora-text-classification
   results: []
@@ -18,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6390
-- Accuracy: {'accuracy': 0.934}
+- Loss: 0.7451
+- Accuracy: {'accuracy': 0.933}
 
 ## Model description
 
@@ -50,16 +49,16 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy            |
 |:-------------:|:-----:|:----:|:---------------:|:-------------------:|
-| No log        | 1.0   | 250  | 0.2597          | {'accuracy': 0.92}  |
-| 0.3563        | 2.0   | 500  | 0.3026          | {'accuracy': 0.924} |
-| 0.3563        | 3.0   | 750  | 0.3926          | {'accuracy': 0.923} |
-| 0.1723        | 4.0   | 1000 | 0.5263          | {'accuracy': 0.92}  |
-| 0.1723        | 5.0   | 1250 | 0.6424          | {'accuracy': 0.923} |
-| 0.0651        | 6.0   | 1500 | 0.5720          | {'accuracy': 0.936} |
-| 0.0651        | 7.0   | 1750 | 0.6541          | {'accuracy': 0.933} |
-| 0.0243        | 8.0   | 2000 | 0.7013          | {'accuracy': 0.931} |
-| 0.0243        | 9.0   | 2250 | 0.6402          | {'accuracy': 0.931} |
-| 0.02          | 10.0  | 2500 | 0.6390          | {'accuracy': 0.934} |
+| No log        | 1.0   | 250  | 0.3071          | {'accuracy': 0.919} |
+| 0.3665        | 2.0   | 500  | 0.3954          | {'accuracy': 0.922} |
+| 0.3665        | 3.0   | 750  | 0.3318          | {'accuracy': 0.937} |
+| 0.1483        | 4.0   | 1000 | 0.5179          | {'accuracy': 0.942} |
+| 0.1483        | 5.0   | 1250 | 0.5112          | {'accuracy': 0.933} |
+| 0.0829        | 6.0   | 1500 | 0.5775          | {'accuracy': 0.936} |
+| 0.0829        | 7.0   | 1750 | 0.6473          | {'accuracy': 0.931} |
+| 0.019         | 8.0   | 2000 | 0.6950          | {'accuracy': 0.937} |
+| 0.019         | 9.0   | 2250 | 0.7328          | {'accuracy': 0.931} |
+| 0.008         | 10.0  | 2500 | 0.7451          | {'accuracy': 0.933} |
 
 
 ### Framework versions
@@ -68,17 +67,3 @@ The following hyperparameters were used during training:
 - Pytorch 2.1.1+cu121
 - Datasets 2.15.0
 - Tokenizers 0.15.0
-## Training procedure
-
-
-### Framework versions
-
-
-- PEFT 0.6.2
-## Training procedure
-
-
-### Framework versions
-
-
-- PEFT 0.6.2

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:473a01dc223a84bc11c96d9c5f391732c8772e2d12d597190962b0974d5f17df
+oid sha256:751ffa3237a6722e852fd7a9c3726e074e8888d72cb6ba2b364ed0285b982f9b
 size 4600