Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,58 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
datasets:
|
| 4 |
+
- seara/ru_go_emotions
|
| 5 |
+
language:
|
| 6 |
+
- ru
|
| 7 |
+
base_model:
|
| 8 |
+
- cointegrated/rubert-tiny2
|
| 9 |
+
pipeline_tag: text-classification
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# RuBert-tiny2-EmotionsDetected
|
| 13 |
+
# RuBert-tyny2-EmotionsDetected
|
| 14 |
+
|
| 15 |
+
This model was obtained by fine-tuning the model [RuBert-tiny2](https://huggingface.co/cointegrated/rubert-tiny2) on the dataset [ru-goemotions](https://github.com/searayeah/ru-goemotions) containing 28 emotions:
|
| 16 |
+
```
|
| 17 |
+
0: admiration (восхищение)
|
| 18 |
+
1: amusement (веселье)
|
| 19 |
+
2: anger (злость)
|
| 20 |
+
3: annoyance (раздражение)
|
| 21 |
+
4: approval (одобрение)
|
| 22 |
+
5: caring (забота)
|
| 23 |
+
6: confusion (непонимание)
|
| 24 |
+
7: curiosity (любопытство)
|
| 25 |
+
8: desire (желание)
|
| 26 |
+
9: disappointment (разочарование)
|
| 27 |
+
10: disapproval (неодобрение)
|
| 28 |
+
11: disgust (отвращение)
|
| 29 |
+
12: embarrassment (смущение)
|
| 30 |
+
13: excitement (возбуждение)
|
| 31 |
+
14: fear (страх)
|
| 32 |
+
15: gratitude (признательность)
|
| 33 |
+
16: grief (горе)
|
| 34 |
+
17: joy (радость)
|
| 35 |
+
18: love (любовь)
|
| 36 |
+
19: nervousness (нервозность)
|
| 37 |
+
20: optimism (оптимизм)
|
| 38 |
+
21: pride (гордость)
|
| 39 |
+
22: realization (осознание)
|
| 40 |
+
23: relief (облегчение)
|
| 41 |
+
24: remorse (раскаяние)
|
| 42 |
+
25: sadness (грусть)
|
| 43 |
+
26: surprise (удивление)
|
| 44 |
+
27: neutral (нейтральность)
|
| 45 |
+
```
|
| 46 |
+
|
| 47 |
+
At the moment, the model has been trained for 40 epochs with the following hyperparameters:
|
| 48 |
+
```
|
| 49 |
+
per_device_train_batch_size=16,
|
| 50 |
+
per_device_eval_batch_size=16,
|
| 51 |
+
num_train_epochs=1,
|
| 52 |
+
weight_decay=0.01,
|
| 53 |
+
learning_rate=1e-5,
|
| 54 |
+
save_total_limit=2,
|
| 55 |
+
load_best_model_at_end=True,
|
| 56 |
+
metric_for_best_model="f1",
|
| 57 |
+
greater_is_better=True
|
| 58 |
+
```
|