Downloads · 30 days
3.7K
1% of all-time downloads
rubert-tiny2-cedr-emotion-detection
cointegrated/rubert-tiny2-cedr-emotion-detection
rubert-tiny2-cedr-emotion-detection is a text classification model from cointegrated. Use it when you need a label for a piece of text. It is set up for transformers.
This is the cointegrated/rubert-tiny2 model fine-tuned for classification of emotions in Russian sentences. The task is multilabel classification, because one sentence can contain multiple emotions.
All-time downloads
453K
Public
Parameters
29.2M
234 MB on disk
Likes
40
Public
Repo makeup
Click a slice to open those files.
.bin117 MB · 49%
Parameter types
How the weights are stored.
F3229.2M · 100%
Try a prompt
Datasets
Model card
From the Hugging Face model README
This is the cointegrated/rubert-tiny2 model fine-tuned for classification of emotions in Russian sentences. The task is multilabel classification, because one sentence can contain multiple emotions.
The model on the CEDR dataset described in the paper "Data-Driven Model for Emotion Detection in Russian Texts" by Sboev et al.
The model has been trained with Adam optimizer for 40 epochs with learning rate 1e-5 and batch size 64 in this notebook.
The quality of the predicted probabilities on the test dataset is the following:
| label | no emotion | joy | sadness | surprise | fear | anger | mean | mean (emotions) |
|---|---|---|---|---|---|---|---|---|
| AUC | 0.9286 | 0.9512 | 0.9564 | 0.8908 | 0.8955 | 0.7511 | 0.8956 | 0.8890 |
| F1 micro | 0.8624 | 0.9389 | 0.9362 | 0.9469 | 0.9575 | 0.9261 | 0.9280 | 0.9411 |
| F1 macro | 0.8562 | 0.8962 | 0.9017 | 0.8366 | 0.8359 | 0.6820 | 0.8348 | 0.8305 |
- Task
- Text Classification
- Library
- transformers
- Type
- bert
- Languages
- ru
- Hf Inference
- live
- Created
- Mar 2, 2022
- Updated
- Mar 20, 2023