| # 繁體中文情緒分類: 負面(0)、正面(1) | |
| 依據ckiplab/albert預訓練模型微調。 | |
| # 使用範例: | |
| from transformers import AutoTokenizer, AutoModelForSequenceClassification | |
| tokenizer = AutoTokenizer.from_pretrained("Cheng-Lung/bert-sentiment") | |
| model = AutoModelForSequenceClassification.from_pretrained("Cheng-Lung/bert-sentiment") | |
| ## Pediction | |
| target_names=['Negative','Positive'] | |
| max_length = 200 # 最多字數 若超出模型訓練時的字數，以模型最大字數為依據 | |
| def get_sentiment_proba(text): | |
| # prepare our text into tokenized sequence | |
| inputs = tokenizer(text, padding=True, truncation=True, max_length=max_length, return_tensors="pt") | |
| # perform inference to our model | |
| outputs = model(**inputs) | |
| # get output probabilities by doing softmax | |
| probs = outputs[0].softmax(1) | |
| response = {'Negative': round(float(probs[0, 0]), 2), 'Positive': round(float(probs[0, 1]), 2)} | |
| # executing argmax function to get the candidate label | |
| #return probs.argmax() | |
| return response | |
| get_sentiment_proba('不喜歡這款產品') |