CoolFace
Modelpublic

seara/rubert-tiny2-russian-emotion-detection-ru-go-emotions

sourceHugging Facemitupdated 2y agoView on Hugging Face
13likes532downloads
Model Card

This is RuBERT-tiny2 model fine-tuned for _emotion classification of short Russian texts. The task is a multi-label classification_ with the following labels:

yaml
0: admiration
1: amusement
2: anger
3: annoyance
4: approval
5: caring
6: confusion
7: curiosity
8: desire
9: disappointment
10: disapproval
11: disgust
12: embarrassment
13: excitement
14: fear
15: gratitude
16: grief
17: joy
18: love
19: nervousness
20: optimism
21: pride
22: realization
23: relief
24: remorse
25: sadness
26: surprise
27: neutral

Label to Russian label:

yaml
admiration: восхищение
amusement: веселье
anger: злость
annoyance: раздражение
approval: одобрение
caring: забота
confusion: непонимание
curiosity: любопытство
desire: желание
disappointment: разочарование
disapproval: неодобрение
disgust: отвращение
embarrassment: смущение
excitement: возбуждение
fear: страх
gratitude: признательность
grief: горе
joy: радость
love: любовь
nervousness: нервозность
optimism: оптимизм
pride: гордость
realization: осознание
relief: облегчение
remorse: раскаяние
sadness: грусть
surprise: удивление
neutral: нейтральность

Usage

python
from transformers import pipeline
model = pipeline(model="seara/rubert-tiny2-ru-go-emotions")
model("Привет, ты мне нравишься!")
# [{'label': 'love', 'score': 0.5955629944801331}]

Dataset

This model was trained on translated GoEmotions dataset called ru_go_emotions.

An overview of the training data can be found on Hugging Face card and on Github repository.

Training

Training were done in this project with this parameters:

yaml
tokenizer.max_length: null
batch_size: 64
optimizer: adam
lr: 0.00001
weight_decay: 0
num_epochs: 31

Eval results (on test split)

precisionrecallf1-scoreauc-rocsupport
admiration0.680.610.640.92504
amusement0.80.840.820.96264
anger0.550.330.420.9198
annoyance0.560.030.060.81320
approval0.60.180.280.78351
caring0.50.040.070.84135
confusion0.770.070.120.9153
curiosity0.510.340.410.92284
desire0.710.180.290.8883
disappointment0.00.00.00.76151
disapproval0.480.10.170.85267
disgust0.940.120.220.9123
embarrassment0.00.00.00.8437
excitement0.810.20.330.88103
fear0.730.420.540.9278
gratitude0.950.890.920.99352
grief0.00.00.00.766
joy0.660.520.580.93161
love0.80.790.790.97238
nervousness0.00.00.00.8123
optimism0.670.410.510.89186
pride0.00.00.00.8916
realization0.00.00.00.7145
relief0.00.00.00.8411
remorse0.590.710.650.9956
sadness0.770.370.50.89156
surprise0.590.350.440.88141
neutral0.640.580.610.811787
micro avg0.680.430.530.936329
macro avg0.510.290.330.876329
weighted avg0.620.430.480.866329