add model
Browse files- README.md +76 -0
- metric.json +1 -0
- pytorch_model.bin +1 -1
README.md
ADDED
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
datasets:
|
3 |
+
- irony
|
4 |
+
metrics:
|
5 |
+
- f1
|
6 |
+
- accuracy
|
7 |
+
model-index:
|
8 |
+
- name: cardiffnlp/twitter-roberta-base-dec2021-irony
|
9 |
+
results:
|
10 |
+
- task:
|
11 |
+
type: text-classification
|
12 |
+
name: Text Classification
|
13 |
+
dataset:
|
14 |
+
name: irony
|
15 |
+
type: tweet_eval
|
16 |
+
split: test
|
17 |
+
metrics:
|
18 |
+
- name: F1
|
19 |
+
type: f1
|
20 |
+
value: 0.7959183673469388
|
21 |
+
- name: F1 (macro)
|
22 |
+
type: f1_macro
|
23 |
+
value: 0.791350632069195
|
24 |
+
- name: Accuracy
|
25 |
+
type: accuracy
|
26 |
+
value: 0.7959183673469388
|
27 |
+
pipeline_tag: text-classification
|
28 |
+
widget:
|
29 |
+
- text: Get the all-analog Classic Vinyl Edition of "Takin' Off" Album from {@herbiehancock@} via {@bluenoterecords@} link below: {{URL}}
|
30 |
+
example_title: "Example"
|
31 |
+
---
|
32 |
+
# cardiffnlp/twitter-roberta-base-dec2021-irony
|
33 |
+
|
34 |
+
This model is a fine-tuned version of [cardiffnlp/twitter-roberta-base-dec2021](https://huggingface.co/cardiffnlp/twitter-roberta-base-dec2021) on the
|
35 |
+
[`irony (tweet_eval))`](https://huggingface.co/datasets/irony)
|
36 |
+
via [`tweetnlp`](https://github.com/cardiffnlp/tweetnlp).
|
37 |
+
Training split is `train` and parameters have been tuned on the validation split `validation`.
|
38 |
+
|
39 |
+
Following metrics are achieved on the test split `test` ([link](https://huggingface.co/cardiffnlp/twitter-roberta-base-dec2021-irony/raw/main/metric.json)).
|
40 |
+
|
41 |
+
- F1 (micro): 0.7959183673469388
|
42 |
+
- F1 (macro): 0.791350632069195
|
43 |
+
- Accuracy: 0.7959183673469388
|
44 |
+
|
45 |
+
### Usage
|
46 |
+
Install tweetnlp via pip.
|
47 |
+
```shell
|
48 |
+
pip install tweetnlp
|
49 |
+
```
|
50 |
+
Load the model in python.
|
51 |
+
```python
|
52 |
+
import tweetnlp
|
53 |
+
model = tweetnlp.Classifier(cardiffnlp/twitter-roberta-base-dec2021-irony, max_length=128)
|
54 |
+
model.predict(Get the all-analog Classic Vinyl Edition of "Takin' Off" Album from {@herbiehancock@} via {@bluenoterecords@} link below: {{URL}})
|
55 |
+
```
|
56 |
+
|
57 |
+
### Reference
|
58 |
+
|
59 |
+
```
|
60 |
+
|
61 |
+
@inproceedings{dimosthenis-etal-2022-twitter,
|
62 |
+
title = "{T}witter {T}opic {C}lassification",
|
63 |
+
author = "Antypas, Dimosthenis and
|
64 |
+
Ushio, Asahi and
|
65 |
+
Camacho-Collados, Jose and
|
66 |
+
Neves, Leonardo and
|
67 |
+
Silva, Vitor and
|
68 |
+
Barbieri, Francesco",
|
69 |
+
booktitle = "Proceedings of the 29th International Conference on Computational Linguistics",
|
70 |
+
month = oct,
|
71 |
+
year = "2022",
|
72 |
+
address = "Gyeongju, Republic of Korea",
|
73 |
+
publisher = "International Committee on Computational Linguistics"
|
74 |
+
}
|
75 |
+
|
76 |
+
```
|
metric.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"eval_loss": 1.3228046894073486, "eval_f1": 0.7959183673469388, "eval_f1_macro": 0.791350632069195, "eval_accuracy": 0.7959183673469388, "eval_runtime": 2.2267, "eval_samples_per_second": 352.084, "eval_steps_per_second": 44.01}
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 498659253
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:28efa8055082cc6d82d1fa43a444f8eaa41112f443d92bf5ea27122195221d37
|
3 |
size 498659253
|