rambodazimi
commited on
Commit
•
c2f0d9a
1
Parent(s):
f9a574e
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
datasets:
|
6 |
+
- glue
|
7 |
+
metrics:
|
8 |
+
- accuracy
|
9 |
+
- f1
|
10 |
+
model-index:
|
11 |
+
- name: distilroberta-base-finetuned-LoRA-MRPC
|
12 |
+
results:
|
13 |
+
- task:
|
14 |
+
name: Text Classification
|
15 |
+
type: text-classification
|
16 |
+
dataset:
|
17 |
+
name: glue
|
18 |
+
type: glue
|
19 |
+
args: mrpc
|
20 |
+
metrics:
|
21 |
+
- name: Accuracy
|
22 |
+
type: accuracy
|
23 |
+
value: 0.8455882352941176
|
24 |
+
- name: F1
|
25 |
+
type: f1
|
26 |
+
value: 0.8911917098445595
|
27 |
+
---
|
28 |
+
|
29 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
30 |
+
should probably proofread and complete it, then remove this comment. -->
|
31 |
+
|
32 |
+
# distilroberta-base-finetuned-lora-mrpc
|
33 |
+
|
34 |
+
This model is a fine-tuned version of [distilroberta-base](https://huggingface.co/distilbert/distilroberta-base) on the glue dataset.
|
35 |
+
It achieves the following results on the evaluation set:
|
36 |
+
- Accuracy: 0.8456
|
37 |
+
- F1: 0.8912
|
38 |
+
- trainable model parameters: 887042
|
39 |
+
- all model parameters: 83006980
|
40 |
+
- percentage of trainable model parameters: 1.07%
|
41 |
+
|
42 |
+
## Training procedure
|
43 |
+
|
44 |
+
### Training hyperparameters
|
45 |
+
|
46 |
+
The following hyperparameters were used during training:
|
47 |
+
- learning_rate: 5e-04
|
48 |
+
- train_batch_size: 32
|
49 |
+
- eval_batch_size: 32
|
50 |
+
- seed: 42
|
51 |
+
- weight_decay: 0.01
|
52 |
+
- rank: 16
|
53 |
+
- lora_alpha: 32
|
54 |
+
- lora_dropout: 0.05
|
55 |
+
- num_epochs: 4
|