train_eval.py

# coding: UTF-8
import numpy as np
import torch
import torch.nn as nn
import torch.nn.functional as F
from sklearn import metrics
import time
from utils import get_time_dif
from pytorch_pretrained.optimization import BertAdam
from AT import FGM
from RAT import RAT

# 权重初始化，默认xavier
def init_network(model, method='xavier', exclude='embedding', seed=123):
    for name, w in model.named_parameters():
        if exclude not in name:
            if len(w.size()) < 2:
                continue
            if 'weight' in name:
                if method == 'xavier':
                    nn.init.xavier_normal_(w)
                elif method == 'kaiming':
                    nn.init.kaiming_normal_(w)
                else:
                    nn.init.normal_(w)
            elif 'bias' in name:
                nn.init.constant_(w, 0)
            else:
                pass


def train(config, model, train_iter, dev_iter, test_iter):
    start_time = time.time()
    model.train()
    param_optimizer = list(model.named_parameters())
    no_decay = ['bias', 'LayerNorm.bias', 'LayerNorm.weight']
    optimizer_grouped_parameters = [
        {'params': [p for n, p in param_optimizer if not any(nd in n for nd in no_decay)], 'weight_decay': 0.01},
        {'params': [p for n, p in param_optimizer if any(nd in n for nd in no_decay)], 'weight_decay': 0.0}]
    # optimizer = torch.optim.Adam(model.parameters(), lr=config.learning_rate)
    optimizer = BertAdam(optimizer_grouped_parameters,
                         lr=config.learning_rate,
                         warmup=0.05,
                         t_total=len(train_iter) * config.num_epochs)
    total_batch = 0  # 记录进行到多少batch
    dev_best_acc = 0.0
    last_improve = 0  # 记录上次验证集loss下降的batch数
    flag = False  # 记录是否很久没有效果提升
    model.train()
    for epoch in range(config.num_epochs):
        print('Epoch [{}/{}]'.format(epoch + 1, config.num_epochs))
        for i, (trains, labels) in enumerate(train_iter):
            outputs = model(trains)
            model.zero_grad()
            loss = F.cross_entropy(outputs, labels)
            loss.backward()
            optimizer.step()
            if total_batch % 100 == 0:
                # 每多少轮输出在训练集和验证集上的效果
                true = labels.data.cpu()
                predic = torch.max(outputs.data, 1)[1].cpu()
                train_acc = metrics.accuracy_score(true, predic)
                dev_acc, dev_loss = evaluate(config, model, dev_iter)
                if dev_acc > dev_best_acc:
                    dev_best_acc = dev_acc
                    torch.save(model.state_dict(), config.save_path)
                    improve = '*'
                    last_improve = total_batch
                else:
                    improve = ''
                time_dif = get_time_dif(start_time)
                msg = 'Iter: {0:>6},  Train Loss: {1:>5.2},  Train Acc: {2:>6.2%},  Val Loss: {3:>5.2},  Val Acc: {4:>6.2%},  Time: {5} {6}'
                print(msg.format(total_batch, loss.item(), train_acc, dev_loss, dev_acc, time_dif, improve))
                model.train()
            total_batch += 1
            if total_batch - last_improve > config.require_improvement:
                # 验证集loss超过1000batch没下降，结束训练
                print("No optimization for a long time, auto-stopping...")
                flag = True
                break
        if flag:
            break
    test(config, model, test_iter)
    
    
def train_AT(config, model, train_iter, dev_iter, test_iter):
    fgm = FGM(model)
    print('AT')
    start_time = time.time()
    model.train()
    param_optimizer = list(model.named_parameters())
    no_decay = ['bias', 'LayerNorm.bias', 'LayerNorm.weight']
    optimizer_grouped_parameters = [
        {'params': [p for n, p in param_optimizer if not any(nd in n for nd in no_decay)], 'weight_decay': 0.01},
        {'params': [p for n, p in param_optimizer if any(nd in n for nd in no_decay)], 'weight_decay': 0.0}]
    # optimizer = torch.optim.Adam(model.parameters(), lr=config.learning_rate)
    optimizer = BertAdam(optimizer_grouped_parameters,
                         lr=config.learning_rate,
                         warmup=0.05,
                         t_total=len(train_iter) * config.num_epochs)
    total_batch = 0  # 记录进行到多少batch
    dev_best_acc = 0.0
    last_improve = 0  # 记录上次验证集loss下降的batch数
    flag = False  # 记录是否很久没有效果提升
    model.train()
    for epoch in range(config.num_epochs):
        print('Epoch [{}/{}]'.format(epoch + 1, config.num_epochs))
        for i, (trains, labels) in enumerate(train_iter):
            outputs = model(trains)
            model.zero_grad()
            loss = F.cross_entropy(outputs, labels)
            loss.backward(retain_graph=True)  # 保留计算图
            fgm.perturb() # 在embedding上添加对抗扰动
            
            output_at = model(trains)
            loss_at = F.cross_entropy(output_at, labels)   # 对抗训练的loss
#             loss_at *= 3
            fgm.restore() # 恢复embedding参数
            loss_at.backward()
            
            optimizer.step()
            loss += loss_at
            
            if total_batch % 100 == 0:
                # 每多少轮输出在训练集和验证集上的效果
                true = labels.data.cpu()
                predic = torch.max(outputs.data, 1)[1].cpu()
                train_acc = metrics.accuracy_score(true, predic)
                dev_acc, dev_loss = evaluate(config, model, dev_iter)
                if dev_acc > dev_best_acc:
                    dev_best_acc = dev_acc
                    torch.save(model.state_dict(), config.save_path)
                    improve = '*'
                    last_improve = total_batch
                else:
                    improve = ''
                time_dif = get_time_dif(start_time)
                msg = 'Iter: {0:>6},  Train Loss: {1:>5.2},  Train Acc: {2:>6.2%},  Val Loss: {3:>5.2},  Val Acc: {4:>6.2%},  Time: {5} {6}'
                print(msg.format(total_batch, loss.item(), train_acc, dev_loss, dev_acc, time_dif, improve))
                model.train()
            total_batch += 1
            if total_batch - last_improve > config.require_improvement:
                # 验证集loss超过1000batch没下降，结束训练
                print("No optimization for a long time, auto-stopping...")
                flag = True
                break
        if flag:
            break
    test(config, model, test_iter)
    
    
def train_RAT(config, model, train_iter, dev_iter, test_iter):
    rat = RAT(model)
    print('R-AT!')
    start_time = time.time()
    model.train()
    param_optimizer = list(model.named_parameters())
    no_decay = ['bias', 'LayerNorm.bias', 'LayerNorm.weight']
    optimizer_grouped_parameters = [
        {'params': [p for n, p in param_optimizer if not any(nd in n for nd in no_decay)], 'weight_decay': 0.01},
        {'params': [p for n, p in param_optimizer if any(nd in n for nd in no_decay)], 'weight_decay': 0.0}]
    # optimizer = torch.optim.Adam(model.parameters(), lr=config.learning_rate)
    optimizer = BertAdam(optimizer_grouped_parameters,
                         lr=config.learning_rate,
                         warmup=0.05,
                         t_total=len(train_iter) * config.num_epochs)
    total_batch = 0  # 记录进行到多少batch
    dev_best_acc= 0.0
    last_improve = 0  # 记录上次验证集loss下降的batch数
    flag = False  # 记录是否很久没有效果提升
    model.train()
    for epoch in range(config.num_epochs):
        print('Epoch [{}/{}]'.format(epoch + 1, config.num_epochs))
        for i, (trains, labels) in enumerate(train_iter):
            outputs = model(trains)
            model.zero_grad()
            loss = F.cross_entropy(outputs, labels)
            loss.backward(retain_graph=True)  # 保留计算图
            rat.perturb() # 在embedding上添加对抗扰动
            
            output_at1 = model(trains)
            output_at2 = model(trains)
            
            loss_at1 = F.cross_entropy(output_at1, labels)   # 对抗训练的loss1
            loss_at2 = F.cross_entropy(output_at2, labels)   # 对抗训练的loss1
            
            kl_loss = rat.compute_kl_loss(output_at1, output_at2)
            
            rat.restore() # 恢复embedding参数
            optimizer.zero_grad()
            loss = loss + 0.5*(loss_at1+loss_at2) +  0.5 * kl_loss
            
            loss.backward()
            
            optimizer.step()

            
            if total_batch % 100 == 0:
                # 每多少轮输出在训练集和验证集上的效果
                true = labels.data.cpu()
                predic = torch.max(outputs.data, 1)[1].cpu()
                train_acc = metrics.accuracy_score(true, predic)
                dev_acc, dev_loss = evaluate(config, model, dev_iter)
                if dev_acc > dev_best_acc:
                    dev_best_acc = dev_acc
                    torch.save(model.state_dict(), config.save_path)
                    improve = '*'
                    last_improve = total_batch
                else:
                    improve = ''
                time_dif = get_time_dif(start_time)
                msg = 'Iter: {0:>6},  Train Loss: {1:>5.2},  Train Acc: {2:>6.2%},  Val Loss: {3:>5.2},  Val Acc: {4:>6.2%},  Time: {5} {6}'
                print(msg.format(total_batch, loss.item(), train_acc, dev_loss, dev_acc, time_dif, improve))
                model.train()
            total_batch += 1
            if total_batch - last_improve > config.require_improvement:
                # 验证集loss超过1000batch没下降，结束训练
                print("No optimization for a long time, auto-stopping...")
                flag = True
                break
        if flag:
            break
    test(config, model, test_iter)
    
    
def compute_kl_loss(p, q):
    
    p_loss = F.kl_div(F.log_softmax(p, dim=-1), F.softmax(q, dim=-1), reduction='none')
    q_loss = F.kl_div(F.log_softmax(q, dim=-1), F.softmax(p, dim=-1), reduction='none')
    

    # You can choose whether to use function "sum" and "mean" depending on your task
    p_loss = p_loss.sum()
    q_loss = q_loss.sum()

    loss = (p_loss + q_loss) / 2
    return loss

def train_RDrop(config, model, train_iter, dev_iter, test_iter):
    print('R-Drop!')
    start_time = time.time()
    model.train()
    param_optimizer = list(model.named_parameters())
    no_decay = ['bias', 'LayerNorm.bias', 'LayerNorm.weight']
    optimizer_grouped_parameters = [
        {'params': [p for n, p in param_optimizer if not any(nd in n for nd in no_decay)], 'weight_decay': 0.01},
        {'params': [p for n, p in param_optimizer if any(nd in n for nd in no_decay)], 'weight_decay': 0.0}]
    # optimizer = torch.optim.Adam(model.parameters(), lr=config.learning_rate)
    optimizer = BertAdam(optimizer_grouped_parameters,
                         lr=config.learning_rate,
                         warmup=0.05,
                         t_total=len(train_iter) * config.num_epochs)
    total_batch = 0  # 记录进行到多少batch
    dev_best_acc = 0.0
    last_improve = 0  # 记录上次验证集loss下降的batch数
    flag = False  # 记录是否很久没有效果提升
    model.train()
    for epoch in range(config.num_epochs):
        print('Epoch [{}/{}]'.format(epoch + 1, config.num_epochs))
        for i, (trains, labels) in enumerate(train_iter):
            outputs1 = model(trains)
            outputs2 = model(trains)
            model.zero_grad()
            loss1 = F.cross_entropy(outputs1, labels)
            loss2 = F.cross_entropy(outputs2, labels)
            
            kl_loss = compute_kl_loss(outputs1, outputs2)

            loss = 0.5*(loss1+loss2) + 0.5 * kl_loss
    
            loss.backward()
            optimizer.step()
            if total_batch % 100 == 0:
                # 每多少轮输出在训练集和验证集上的效果
                true = labels.data.cpu()
                predic = torch.max(0.5*(outputs1+outputs2).data, 1)[1].cpu()
                train_acc = metrics.accuracy_score(true, predic)
                dev_acc, dev_loss = evaluate(config, model, dev_iter)
                if dev_acc > dev_best_acc:
                    dev_best_acc = dev_acc
                    torch.save(model.state_dict(), config.save_path)
                    improve = '*'
                    last_improve = total_batch
                else:
                    improve = ''
                time_dif = get_time_dif(start_time)
                msg = 'Iter: {0:>6},  Train Loss: {1:>5.2},  Train Acc: {2:>6.2%},  Val Loss: {3:>5.2},  Val Acc: {4:>6.2%},  Time: {5} {6}'
                print(msg.format(total_batch, loss.item(), train_acc, dev_loss, dev_acc, time_dif, improve))
                model.train()
            total_batch += 1
            if total_batch - last_improve > config.require_improvement:
                # 验证集loss超过1000batch没下降，结束训练
                print("No optimization for a long time, auto-stopping...")
                flag = True
                break
        if flag:
            break
    test(config, model, test_iter)
    
    
    
def test(config, model, test_iter):
    # test
    model.load_state_dict(torch.load(config.save_path))
    model.eval()
    start_time = time.time()
    test_acc, test_loss, test_report, test_confusion = evaluate(config, model, test_iter, test=True)
    msg = 'Test Loss: {0:>5.2},  Test Acc: {1:>6.2%}'
    print(msg.format(test_loss, test_acc))
    print("Precision, Recall and F1-Score...")
    print(test_report)
    print("Confusion Matrix...")
    print(test_confusion)
    time_dif = get_time_dif(start_time)
    print("Time usage:", time_dif)


def evaluate(config, model, data_iter, test=False):
    model.eval()
    loss_total = 0
    predict_all = np.array([], dtype=int)
    labels_all = np.array([], dtype=int)
    with torch.no_grad():
        for texts, labels in data_iter:
            outputs = model(texts)
            loss = F.cross_entropy(outputs, labels)
            loss_total += loss
            labels = labels.data.cpu().numpy()
            predic = torch.max(outputs.data, 1)[1].cpu().numpy()
            labels_all = np.append(labels_all, labels)
            predict_all = np.append(predict_all, predic)

    acc = metrics.accuracy_score(labels_all, predict_all)
    if test:
        report = metrics.classification_report(labels_all, predict_all, target_names=config.class_list, digits=4)
        confusion = metrics.confusion_matrix(labels_all, predict_all)
        return acc, loss_total / len(data_iter), report, confusion
    return acc, loss_total / len(data_iter)


def predict(config, model, data_iter):
#     model.load_state_dict(torch.load(config.save_path))
    model.eval()
    predict_all = np.array([], dtype=int)
    labels_all = np.array([], dtype=int)
    texts_all = np.array([], dtype=int)
    with torch.no_grad():
        for texts, labels in data_iter:
            outputs = model(texts)
            labels = labels.data.cpu().numpy()
            predic = torch.max(outputs.data, 1)[1].cpu().numpy()
            labels_all = np.append(labels_all, labels)
            predict_all = np.append(predict_all, predic)
            texts_all = np.append(texts_all, texts)
            for a,b in zip(outputs,predict_all):
                print(outputs)