This is the cointegrated/rubert tiny model fine tuned for classification of sentiment for short Russian texts. The problem is formulated as multiclass classification: negative vs neutral vs positive . Usage The function below estimates the sentiment of the given text: Training We trained the model on the datasets collected by Smetanin. We have converted all training data into a 3 class format and have up and downsampled the training data to balance both the sources and the classes. The training code is available as a Colab notebook. The metrics on the balanced test set are the following: Source Macro F1 SentiRuEval2016 banks 0.83 SentiRuEval2016 tele 0.74 kaggle news 0.66 linis 0.50 mokoron 0.98 rureviews 0.72 rusentiment 0.67
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy