Commit
•
bc05e37
1
Parent(s):
caeab9b
update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,84 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
license: apache-2.0
|
5 |
+
tags:
|
6 |
+
- generated_from_trainer
|
7 |
+
datasets:
|
8 |
+
- glue
|
9 |
+
metrics:
|
10 |
+
- accuracy
|
11 |
+
- f1
|
12 |
+
model-index:
|
13 |
+
- name: fnet-large-finetuned-mrpc
|
14 |
+
results:
|
15 |
+
- task:
|
16 |
+
name: Text Classification
|
17 |
+
type: text-classification
|
18 |
+
dataset:
|
19 |
+
name: GLUE MRPC
|
20 |
+
type: glue
|
21 |
+
args: mrpc
|
22 |
+
metrics:
|
23 |
+
- name: Accuracy
|
24 |
+
type: accuracy
|
25 |
+
value: 0.8259803921568627
|
26 |
+
- name: F1
|
27 |
+
type: f1
|
28 |
+
value: 0.8798646362098139
|
29 |
+
---
|
30 |
+
|
31 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
32 |
+
should probably proofread and complete it, then remove this comment. -->
|
33 |
+
|
34 |
+
# fnet-large-finetuned-mrpc
|
35 |
+
|
36 |
+
This model is a fine-tuned version of [google/fnet-large](https://huggingface.co/google/fnet-large) on the GLUE MRPC dataset.
|
37 |
+
It achieves the following results on the evaluation set:
|
38 |
+
- Loss: 1.0872
|
39 |
+
- Accuracy: 0.8260
|
40 |
+
- F1: 0.8799
|
41 |
+
- Combined Score: 0.8529
|
42 |
+
|
43 |
+
## Model description
|
44 |
+
|
45 |
+
More information needed
|
46 |
+
|
47 |
+
## Intended uses & limitations
|
48 |
+
|
49 |
+
More information needed
|
50 |
+
|
51 |
+
## Training and evaluation data
|
52 |
+
|
53 |
+
More information needed
|
54 |
+
|
55 |
+
## Training procedure
|
56 |
+
|
57 |
+
### Training hyperparameters
|
58 |
+
|
59 |
+
The following hyperparameters were used during training:
|
60 |
+
- learning_rate: 2e-05
|
61 |
+
- train_batch_size: 4
|
62 |
+
- eval_batch_size: 8
|
63 |
+
- seed: 42
|
64 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
65 |
+
- lr_scheduler_type: linear
|
66 |
+
- num_epochs: 5.0
|
67 |
+
|
68 |
+
### Training results
|
69 |
+
|
70 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Combined Score |
|
71 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:--------------:|
|
72 |
+
| 0.5656 | 1.0 | 917 | 0.6999 | 0.7843 | 0.8581 | 0.8212 |
|
73 |
+
| 0.3874 | 2.0 | 1834 | 0.7280 | 0.8088 | 0.8691 | 0.8390 |
|
74 |
+
| 0.1627 | 3.0 | 2751 | 1.1274 | 0.8162 | 0.8780 | 0.8471 |
|
75 |
+
| 0.0751 | 4.0 | 3668 | 1.0289 | 0.8333 | 0.8870 | 0.8602 |
|
76 |
+
| 0.0339 | 5.0 | 4585 | 1.0872 | 0.8260 | 0.8799 | 0.8529 |
|
77 |
+
|
78 |
+
|
79 |
+
### Framework versions
|
80 |
+
|
81 |
+
- Transformers 4.11.0.dev0
|
82 |
+
- Pytorch 1.9.0
|
83 |
+
- Datasets 1.12.1
|
84 |
+
- Tokenizers 0.10.3
|