Update README.md
Browse files
README.md
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
---
|
| 2 |
-
|
| 3 |
license: mit
|
| 4 |
-
language:
|
|
|
|
| 5 |
tags:
|
| 6 |
- russian
|
| 7 |
- classification
|
|
@@ -10,15 +10,16 @@ tags:
|
|
| 10 |
- emotion-recognition
|
| 11 |
- multiclass
|
| 12 |
widget:
|
| 13 |
-
- text:
|
| 14 |
-
- text:
|
| 15 |
-
- text:
|
| 16 |
-
- text:
|
| 17 |
-
- text:
|
| 18 |
-
- text:
|
|
|
|
|
|
|
| 19 |
datasets:
|
| 20 |
-
- Djacon/
|
| 21 |
-
|
| 22 |
---
|
| 23 |
|
| 24 |
# First - you should prepare few functions to talk to model
|
|
@@ -27,7 +28,7 @@ datasets:
|
|
| 27 |
import torch
|
| 28 |
from transformers import BertForSequenceClassification, AutoTokenizer
|
| 29 |
|
| 30 |
-
|
| 31 |
tokenizer = AutoTokenizer.from_pretrained('Djacon/rubert-tiny2-russian-emotion-detection')
|
| 32 |
model = BertForSequenceClassification.from_pretrained('Djacon/rubert-tiny2-russian-emotion-detection')
|
| 33 |
|
|
@@ -36,27 +37,28 @@ model = BertForSequenceClassification.from_pretrained('Djacon/rubert-tiny2-russi
|
|
| 36 |
def predict_emotion(text: str) -> str:
|
| 37 |
inputs = tokenizer(text, truncation=True, return_tensors='pt')
|
| 38 |
inputs = inputs.to(model.device)
|
| 39 |
-
|
| 40 |
outputs = model(**inputs)
|
| 41 |
-
|
| 42 |
-
pred = torch.nn.functional.
|
| 43 |
pred = pred.argmax(dim=1)
|
|
|
|
|
|
|
| 44 |
|
| 45 |
-
return LABELS[pred[0]]
|
| 46 |
|
| 47 |
# Probabilistic prediction of emotion in a text
|
| 48 |
@torch.no_grad()
|
| 49 |
-
def predict_emotions(text: str) ->
|
| 50 |
inputs = tokenizer(text, truncation=True, return_tensors='pt')
|
| 51 |
inputs = inputs.to(model.device)
|
| 52 |
|
| 53 |
outputs = model(**inputs)
|
| 54 |
|
| 55 |
-
pred = torch.nn.functional.
|
| 56 |
|
| 57 |
emotions_list = {}
|
| 58 |
for i in range(len(pred[0].tolist())):
|
| 59 |
-
emotions_list[
|
| 60 |
return emotions_list
|
| 61 |
```
|
| 62 |
|
|
@@ -68,8 +70,8 @@ not_simple_prediction = predict_emotions("Какой же сегодня пре
|
|
| 68 |
|
| 69 |
print(simple_prediction)
|
| 70 |
print(not_simple_prediction)
|
| 71 |
-
#
|
| 72 |
-
# {'
|
| 73 |
```
|
| 74 |
|
| 75 |
# Citations
|
|
|
|
| 1 |
---
|
|
|
|
| 2 |
license: mit
|
| 3 |
+
language:
|
| 4 |
+
- ru
|
| 5 |
tags:
|
| 6 |
- russian
|
| 7 |
- classification
|
|
|
|
| 10 |
- emotion-recognition
|
| 11 |
- multiclass
|
| 12 |
widget:
|
| 13 |
+
- text: Как дела?
|
| 14 |
+
- text: Дурак твой дед
|
| 15 |
+
- text: Только попробуй!!!
|
| 16 |
+
- text: Не хочу в школу(
|
| 17 |
+
- text: Сейчас ровно час дня
|
| 18 |
+
- text: >-
|
| 19 |
+
А ты уверен, что эти полоски снизу не врут? Точно уверен? Вот прям 100
|
| 20 |
+
процентов?
|
| 21 |
datasets:
|
| 22 |
+
- Djacon/ru_go_emotions
|
|
|
|
| 23 |
---
|
| 24 |
|
| 25 |
# First - you should prepare few functions to talk to model
|
|
|
|
| 28 |
import torch
|
| 29 |
from transformers import BertForSequenceClassification, AutoTokenizer
|
| 30 |
|
| 31 |
+
LABELS_RU = ['нейтрально', 'радость', 'грусть', 'гнев', 'интерес', 'удивление', 'отвращение', 'страх', 'вина', 'стыд']
|
| 32 |
tokenizer = AutoTokenizer.from_pretrained('Djacon/rubert-tiny2-russian-emotion-detection')
|
| 33 |
model = BertForSequenceClassification.from_pretrained('Djacon/rubert-tiny2-russian-emotion-detection')
|
| 34 |
|
|
|
|
| 37 |
def predict_emotion(text: str) -> str:
|
| 38 |
inputs = tokenizer(text, truncation=True, return_tensors='pt')
|
| 39 |
inputs = inputs.to(model.device)
|
| 40 |
+
|
| 41 |
outputs = model(**inputs)
|
| 42 |
+
|
| 43 |
+
pred = torch.nn.functional.sigmoid(outputs.logits)
|
| 44 |
pred = pred.argmax(dim=1)
|
| 45 |
+
|
| 46 |
+
return LABELS_RU[pred[0]]
|
| 47 |
|
|
|
|
| 48 |
|
| 49 |
# Probabilistic prediction of emotion in a text
|
| 50 |
@torch.no_grad()
|
| 51 |
+
def predict_emotions(text: str) -> dict:
|
| 52 |
inputs = tokenizer(text, truncation=True, return_tensors='pt')
|
| 53 |
inputs = inputs.to(model.device)
|
| 54 |
|
| 55 |
outputs = model(**inputs)
|
| 56 |
|
| 57 |
+
pred = torch.nn.functional.sigmoid(outputs.logits)
|
| 58 |
|
| 59 |
emotions_list = {}
|
| 60 |
for i in range(len(pred[0].tolist())):
|
| 61 |
+
emotions_list[LABELS_RU[i]] = round(pred[0].tolist()[i], 4)
|
| 62 |
return emotions_list
|
| 63 |
```
|
| 64 |
|
|
|
|
| 70 |
|
| 71 |
print(simple_prediction)
|
| 72 |
print(not_simple_prediction)
|
| 73 |
+
# радость
|
| 74 |
+
# {'нейтрально': 0.1985, 'радость': 0.7419, 'грусть': 0.0261, 'гнев': 0.0295, 'интерес': 0.1983, 'удивление': 0.4305, 'отвращение': 0.0082, 'страх': 0.008, 'вина': 0.0046, 'стыд': 0.0097}
|
| 75 |
```
|
| 76 |
|
| 77 |
# Citations
|