api.py

# demo_phobert_api.py
# -*- coding: utf-8 -*-

from fastapi import FastAPI
from pydantic import BaseModel
import torch
import re
import json
import emoji
from underthesea import word_tokenize
from transformers import AutoConfig, AutoTokenizer, AutoModelForSequenceClassification

# Khởi tạo FastAPI app
app = FastAPI(
    title="PhoBERT Emotion Classification API",
    description="API dự đoán cảm xúc của câu tiếng Việt sử dụng PhoBERT.",
    version="1.0"
)

###############################################################################
# TẢI MAPPING EMOJI - COPY Y NGUYÊN TỪ FILE TRAIN
###############################################################################
emoji_mapping = {
    "😀": "[joy]", "😃": "[joy]", "😄": "[joy]", "😁": "[joy]", "😆": "[joy]", "😅": "[joy]", "😂": "[joy]", "🤣": "[joy]",
    "🙂": "[love]", "🙃": "[love]", "😉": "[love]", "😊": "[love]", "😇": "[love]", "🥰": "[love]", "😍": "[love]",
    "🤩": "[love]", "😘": "[love]", "😗": "[love]", "☺": "[love]", "😚": "[love]", "😙": "[love]",
    "😋": "[satisfaction]", "😛": "[satisfaction]", "😜": "[satisfaction]", "🤪": "[satisfaction]", "😝": "[satisfaction]",
    "🤑": "[satisfaction]",
    "🤐": "[neutral]", "🤨": "[neutral]", "😐": "[neutral]", "😑": "[neutral]", "😶": "[neutral]",
    "😏": "[sarcasm]",
    "😒": "[disappointment]", "🙄": "[disappointment]", "😬": "[disappointment]",
    "😔": "[sadness]", "😪": "[sadness]", "😢": "[sadness]", "😭": "[sadness]", "😥": "[sadness]", "😓": "[sadness]",
    "😩": "[tiredness]", "😫": "[tiredness]", "🥱": "[tiredness]",
    "🤤": "[discomfort]", "🤢": "[discomfort]", "🤮": "[discomfort]", "🤧": "[discomfort]", "🥵": "[discomfort]",
    "🥶": "[discomfort]", "🥴": "[discomfort]", "😵": "[discomfort]", "🤯": "[discomfort]",
    "😕": "[confused]", "😟": "[confused]", "🙁": "[confused]", "☹": "[confused]",
    "😮": "[surprise]", "😯": "[surprise]", "😲": "[surprise]", "😳": "[surprise]", "🥺": "[pleading]",
    "😦": "[fear]", "😧": "[fear]", "😨": "[fear]", "😰": "[fear]", "😱": "[fear]",
    "😖": "[confusion]", "😣": "[confusion]", "😞": "[confusion]",
    "😤": "[anger]", "😡": "[anger]", "😠": "[anger]", "🤬": "[anger]", "😈": "[mischievous]", "👿": "[mischievous]"
}

###############################################################################
# HÀM XỬ LÝ (COPY TỪ FILE TRAIN)
###############################################################################
def replace_emojis(sentence, emoji_mapping):
    processed_sentence = []
    for char in sentence:
        if char in emoji_mapping:
            processed_sentence.append(emoji_mapping[char])
        elif not emoji.is_emoji(char):
            processed_sentence.append(char)
    return ''.join(processed_sentence)

def remove_profanity(sentence):
    profane_words = ["loz", "vloz", "vl", "dm", "đm", "clgt", "dmm", "cc", "vc", "đù mé", "vãi"]
    words = sentence.split()
    filtered = [w for w in words if w.lower() not in profane_words]
    return ' '.join(filtered)

def remove_special_characters(sentence):
    return re.sub(r"[\^\*@#&$%<>~{}|\\]", "", sentence)

def normalize_whitespace(sentence):
    return ' '.join(sentence.split())

def remove_repeated_characters(sentence):
    return re.sub(r"(.)\1{2,}", r"\1", sentence)

def replace_numbers(sentence):
    return re.sub(r"\d+", "[number]", sentence)

def tokenize_underthesea(sentence):
    tokens = word_tokenize(sentence)
    return " ".join(tokens)

# Nếu có abbreviations.json, load nó. Nếu không thì để rỗng.
try:
    with open("abbreviations.json", "r", encoding="utf-8") as f:
        abbreviations = json.load(f)
except Exception as e:
    abbreviations = {}

def preprocess_sentence(sentence):
    sentence = sentence.lower()
    sentence = replace_emojis(sentence, emoji_mapping)
    sentence = remove_profanity(sentence)
    sentence = remove_special_characters(sentence)
    sentence = normalize_whitespace(sentence)
    # Thay thế từ viết tắt nếu có trong abbreviations
    words = sentence.split()
    replaced = []
    for w in words:
        if w in abbreviations:
            replaced.append(" ".join(abbreviations[w]))
        else:
            replaced.append(w)
    sentence = " ".join(replaced)
    sentence = remove_repeated_characters(sentence)
    sentence = replace_numbers(sentence)
    sentence = tokenize_underthesea(sentence)
    return sentence

###############################################################################
# LOAD CHECKPOINT
###############################################################################
checkpoint_dir = "./checkpoint"  # Đường dẫn đến folder checkpoint
device = "cuda" if torch.cuda.is_available() else "cpu"

print("Loading config...")
config = AutoConfig.from_pretrained(checkpoint_dir)

# Mapping id to label theo thứ tự bạn cung cấp
custom_id2label = {
    0: 'Anger',
    1: 'Disgust',
    2: 'Enjoyment',
    3: 'Fear',
    4: 'Other',
    5: 'Sadness',
    6: 'Surprise'
}

if hasattr(config, "id2label") and config.id2label:
    if all(label.startswith("LABEL_") for label in config.id2label.values()):
        id2label = custom_id2label
    else:
        id2label = {int(k): v for k, v in config.id2label.items()}
else:
    id2label = custom_id2label

print("id2label loaded:", id2label)

print("Loading tokenizer...")
tokenizer = AutoTokenizer.from_pretrained(checkpoint_dir)

print("Loading model...")
model = AutoModelForSequenceClassification.from_pretrained(checkpoint_dir, config=config)
model.to(device)
model.eval()

###############################################################################
# HÀM PREDICT
###############################################################################
label2message = {
    'Anger': 'Hãy bình tĩnh và giải quyết vấn đề một cách bình thản.',
    'Disgust': 'Hãy tránh xa những thứ khiến bạn không thích.',
    'Enjoyment': 'Chúc mừng bạn có một ngày tuyệt vời!',
    'Fear': 'Hãy đối mặt với nỗi sợ để vượt qua chúng.',
    'Other': 'Cảm xúc của bạn hiện tại không được phân loại rõ ràng.',
    'Sadness': 'Hãy tìm kiếm sự hỗ trợ khi cần thiết.',
    'Surprise': 'Thật bất ngờ! Hãy tận hưởng khoảnh khắc này.'
}

def predict_text(text: str) -> str:
    text_proc = preprocess_sentence(text)
    inputs = tokenizer(
        [text_proc],
        padding=True,
        truncation=True,
        max_length=256,
        return_tensors="pt"
    ).to(device)

    with torch.no_grad():
        outputs = model(**inputs)
        pred_id = outputs.logits.argmax(dim=-1).item()

    if pred_id in id2label:
        label = id2label[pred_id]
        message = label2message.get(label, "")
        if message:
            return f"Dự đoán cảm xúc: {label}. {message}"
        else:
            return f"Dự đoán cảm xúc: {label}."
    else:
        return f"Nhãn không xác định (id={pred_id})"

###############################################################################
# ĐỊNH NGHĨA MODEL INPUT
###############################################################################
class InputText(BaseModel):
    text: str

###############################################################################
# API ENDPOINT
###############################################################################
@app.post("/predict")
def predict(input_text: InputText):
    """
    Nhận một câu tiếng Việt và trả về dự đoán cảm xúc.
    """
    result = predict_text(input_text.text)
    return {"result": result}

###############################################################################
# CHẠY API SERVER
###############################################################################
if __name__ == "__main__":
    import uvicorn
    uvicorn.run(app, host="0.0.0.0", port=8000)