Instructions to use arpanghoshal/EkmanClassifier with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use arpanghoshal/EkmanClassifier with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="arpanghoshal/EkmanClassifier")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, BertForMultiLabelClassification tokenizer = AutoTokenizer.from_pretrained("arpanghoshal/EkmanClassifier") model = BertForMultiLabelClassification.from_pretrained("arpanghoshal/EkmanClassifier", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Commit ·
a6d477d
1
Parent(s): 869d738
Update README.md
Browse files
README.md
CHANGED
|
@@ -7,4 +7,45 @@ tags:
|
|
| 7 |
datasets:
|
| 8 |
- go_emotions
|
| 9 |
license: mit
|
| 10 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
datasets:
|
| 8 |
- go_emotions
|
| 9 |
license: mit
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
- [linkedin.com/in/arpanghoshal](https://www.linkedin.com/in/arpanghoshal)
|
| 13 |
+
|
| 14 |
+
|
| 15 |
+
## What is Ekman Emotions?
|
| 16 |
+
|
| 17 |
+
Ekman emotions refer to a specific set of six basic emotions that are thought to be universal across cultures. </br>
|
| 18 |
+
These emotions were identified by psychologist Paul Ekman, who conducted extensive research on facial expressions and emotional experience. </br>
|
| 19 |
+
|
| 20 |
+
The six Ekman emotions are:
|
| 21 |
+
|
| 22 |
+
- Happiness
|
| 23 |
+
- Sadness
|
| 24 |
+
- Anger
|
| 25 |
+
- Fear
|
| 26 |
+
- Disgust
|
| 27 |
+
- Surprise
|
| 28 |
+
|
| 29 |
+
Ekman's research found that these emotions are expressed through universal facial expressions, which are recognizable across cultures. </br>
|
| 30 |
+
According to Ekman, these emotions are biologically based and are fundamental to human social interaction. </br>
|
| 31 |
+
Ekman's work on basic emotions has had a significant impact on the field of psychology and has been widely influential in the study of emotional expression and experience.
|
| 32 |
+
|
| 33 |
+
## Usage
|
| 34 |
+
|
| 35 |
+
```python
|
| 36 |
+
|
| 37 |
+
from transformers import BertTokenizer
|
| 38 |
+
from model import BertForMultiLabelClassification
|
| 39 |
+
from multilabel_pipeline import MultiLabelPipeline
|
| 40 |
+
from pprint import pprint
|
| 41 |
+
|
| 42 |
+
tokenizer = BertTokenizer.from_pretrained('arpanghoshal/EkmanClassifier')
|
| 43 |
+
model = BertForMultiLabelClassification.from_pretrained('arpanghoshal/EkmanClassifier')
|
| 44 |
+
|
| 45 |
+
ekman = pipeline('sentiment-analysis', model='arpanghoshal/EmoRoBERTa')
|
| 46 |
+
|
| 47 |
+
ekman_labels = ekman("Thanks for using it.")
|
| 48 |
+
print(ekman_labels)
|
| 49 |
+
|
| 50 |
+
|
| 51 |
+
```
|