```python
from senta import Senta
my_senta = Senta()

# получить предварительно обученную модель, мы предоставляем три предварительно обученных модели, все из которых основаны на SKEP
print(my_senta.get_support_model()) # ["ernie_1.0_skep_large_ch", "ernie_2.0_skep_large_en", "roberta_skep_large_en"]
                                    # ernie_1.0_skep_large_ch, китайская предварительно обученная модель skep, основанная на ERNIE 1.0 large.
                                    # ernie_2.0_skep_large_en, английская предварительно обученная модель skep, основанная на ERNIE 2.0 large.
                                    # roberta_skep_large_en, английская предварительно обученная модель skep, основанная на RoBERTa large, которая используется в нашей статье.

# получить поддерживаемую задачу
print(my_senta.get_support_task()) # ["sentiment_classify", "aspect_sentiment_classify", "extraction"]

use_cuda = True # установить True или False

# прогнозировать разные задачи
my_senta.init_model(model_class="roberta_skep_large_en", task="sentiment_classify", use_cuda=use_cuda)
texts = ["a sometimes tedious film ."]
result = my_senta.predict(texts)
print(result)

my_senta.init_model(model_class="roberta_skep_large_en", task="aspect_sentiment_classify", use_cuda=use_cuda)
texts = ["I love the operating system and the preloaded software."]
aspects = ["operating system"]
result = my_senta.predict(texts, aspects)
print(result)

my_senta.init_model(model_class="roberta_skep_large_en", task="extraction", use_cuda=use_cuda)
texts = ["The JCC would be very pleased to welcome your organization as a corporate sponsor ."]
result = my_senta.predict(texts)
print(result)
```

```shell
# скачать код
git clone https://github.com/baidu/Senta.git

# загрузить предварительно обученную модель skep
cd ./Senta/model_files
sh download_roberta_skep_large_en.sh # скачать модель roberta_skep_large_en. Для других предварительно обученных моделей skep вы можете найти их в этой папке.
cd -

# скачать набор данных задачи
cd ./Senta/data/
sh download_en_data.sh # скачать англоязычный набор данных, используемый в нашей статье. Китайский набор данных можно найти в сценарии загрузки в этой папке.
cd - 
```

sh ./script/run_pretrain_roberta_skep_large_en.sh # предобученная модель roberta_skep_large_en, которая используется в нашей статье

sh ./script/run_train.sh ./config/roberta_skep_large_en.SST-2.cls.json # дообучение на SST-2
sh ./script/run_infer.sh ./config/roberta_skep_large_en.SST-2.infer.json # предсказание

sh ./script/run_train.sh ./config/roberta_skep_large_en.absa_laptops.cls.json # дообучение на ABSA (ноутбуки)
sh ./script/run_infer.sh ./config/roberta_skep_large_en.absa_laptops.infer.json # предсказание

sh ./script/run_train.sh ./config/roberta_skep_large_en.MPQA.orl.json # дообучение на MPQA 2.0
sh ./script/run_infer.sh ./config/roberta_skep_large_en.MPQA.infer.json # предсказание

@inproceedings{tian-etal-2020-skep,
    title = "{SKEP}: Sentiment Knowledge Enhanced Pre-training for Sentiment Analysis",
    author = "Tian, Hao  and
      Gao, Can  and
      Xiao, Xinyan  and
      Liu, Hao  and
      He, Bolei  and
      Wu, Hua  and
      Wang, Haifeng  and
      wu, feng",
    booktitle = "Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics",
    month = jul,
    year = "2020",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://www.aclweb.org/anthology/2020.acl-main.374",
    pages = "4067--4076",
    abstract = "Recently, sentiment analysis has seen remarkable advance with the help of pre-training approaches. However, sentiment knowledge, such as sentiment words and aspect-sentiment pairs, is ignored in the process of pre-training, despite the fact that they are widely used in traditional sentiment analysis approaches. In this paper, we introduce Sentiment Knowledge Enhanced Pre-training (SKEP) in order to learn a unified sentiment representation for multiple sentiment analysis tasks. With the help of automatically-mined knowledge, SKEP conducts sentiment masking and constructs three sentiment knowledge prediction objectives, so as to embed sentiment information at the word, polarity and aspect level into pre-trained sentiment representation. In particular, the prediction of aspect-sentiment pairs is converted into multi-label classification, aiming to capture the dependency between words in a pair. Experiments on three kinds of sentiment tasks show that SKEP significantly outperforms strong pre-training baseline, and achieves new state-of-the-art results on most of the test datasets. We release our code at https://github.com/baidu/Senta."
}