train.py

import random
from data import ImageDetectionsField, TextField, RawField
from data import COCO, DataLoader
import evaluation
from evaluation import PTBTokenizer, Cider
#from models.transformer import Transformer, MemoryAugmentedEncoder, MeshedDecoder, ScaledDotProductAttentionMemory
from models.LSTNet import Transformer, TransformerEncoder, TransformerDecoderLayer, \
    ScaledDotProductAttention
import torch
from torch.optim import Adam
from torch.optim.lr_scheduler import LambdaLR
from torch.nn import NLLLoss
from tqdm import tqdm
from torch.utils.tensorboard import SummaryWriter
import argparse, os, pickle
import numpy as np
import itertools
import multiprocessing
from shutil import copyfile
import os
from models.augmentation import ImageAugmentation

# os.environ['CUDA_VISIBLE_DEVICES']='0'

random.seed(19890511)
torch.manual_seed(19890511)
np.random.seed(19890511)


def evaluate_loss(model, dataloader, loss_fn, text_field):


    # Validation loss
    model.eval()
    running_loss = .0
    with tqdm(desc='Epoch %d - validation' % e, unit='it', total=len(dataloader)) as pbar:
        with torch.no_grad():
            for it, (detections, captions) in enumerate(dataloader):
                detections, captions = detections.to(device), captions.to(device)
                out = model(detections, captions)
                captions = captions[:, 1:].contiguous()
                out = out[:, :-1].contiguous()
                loss = loss_fn(out.view(-1, len(text_field.vocab)), captions.view(-1))
                this_loss = loss.item()
                running_loss += this_loss

                pbar.set_postfix(loss=running_loss / (it + 1))
                pbar.update()

    val_loss = running_loss / len(dataloader)
    return val_loss


def evaluate_metrics(model, dataloader, text_field):
    import itertools
    model.eval()
    gen = {}
    gts = {}
    with tqdm(desc='Epoch %d - evaluation' % e, unit='it', total=len(dataloader)) as pbar:
        for it, (images, caps_gt) in enumerate(iter(dataloader)):
            images = images.to(device)
            with torch.no_grad():
                out, _ = model.beam_search(images, 20, text_field.vocab.stoi['<eos>'], 5, out_size=1)
            caps_gen = text_field.decode(out, join_words=False)
            for i, (gts_i, gen_i) in enumerate(zip(caps_gt, caps_gen)):
                gen_i = ' '.join([k for k, g in itertools.groupby(gen_i)])
                gen['%d_%d' % (it, i)] = [gen_i, ]
                gts['%d_%d' % (it, i)] = gts_i
            pbar.update()

    gts = evaluation.PTBTokenizer.tokenize(gts)
    gen = evaluation.PTBTokenizer.tokenize(gen)
    scores, _ = evaluation.compute_scores(gts, gen)
    return scores


def train_xe(model, dataloader, optim, text_field):
    # Training with cross-entropy
    model.train()
    scheduler.step()
    running_loss = .0
    with tqdm(desc='Epoch %d - train' % e, unit='it', total=len(dataloader)) as pbar:
        for it, (detections, captions) in enumerate(dataloader):
            detections=ia.randnChooseOne4(detections)
            detections, captions = detections.to(device), captions.to(device)
            out = model(detections, captions)
            optim.zero_grad()
            captions_gt = captions[:, 1:].contiguous()
            out = out[:, :-1].contiguous()
            loss = loss_fn(out.view(-1, len(text_field.vocab)), captions_gt.view(-1))
            loss.backward()

            optim.step()
            this_loss = loss.item()
            running_loss += this_loss

            pbar.set_postfix(loss=running_loss / (it + 1))
            pbar.update()
            #scheduler.step()

    loss = running_loss / len(dataloader)
    return loss


def train_scst(model, dataloader, optim, cider, text_field):
    # Training with self-critical
    tokenizer_pool = multiprocessing.Pool()
    running_reward = .0
    running_reward_baseline = .0
    model.train()
    scheduler.step()
    running_loss = .0
    seq_len = 20
    beam_size = 5

    with tqdm(desc='Epoch %d - train' % e, unit='it', total=len(dataloader)) as pbar:
        for it, (detections, caps_gt) in enumerate(dataloader):
            detections=ia.randnChooseOne4(detections)
            detections = detections.to(device)
            outs, log_probs = model.beam_search(detections, seq_len, text_field.vocab.stoi['<eos>'],
                                                beam_size, out_size=beam_size)
            optim.zero_grad()

            # Rewards
            caps_gen = text_field.decode(outs.view(-1, seq_len))
            caps_gt = list(itertools.chain(*([c, ] * beam_size for c in caps_gt)))
            caps_gen, caps_gt = tokenizer_pool.map(evaluation.PTBTokenizer.tokenize, [caps_gen, caps_gt])
            reward = cider.compute_score(caps_gt, caps_gen)[1].astype(np.float32)
            reward = torch.from_numpy(reward).to(device).view(detections.shape[0], beam_size)
            reward_baseline = torch.mean(reward, -1, keepdim=True)
            loss = -torch.mean(log_probs, -1) * (reward - reward_baseline)

            loss = loss.mean()
            loss.backward()
            optim.step()

            running_loss += loss.item()
            running_reward += reward.mean().item()
            running_reward_baseline += reward_baseline.mean().item()
            pbar.set_postfix(loss=running_loss / (it + 1), reward=running_reward / (it + 1),
                             reward_baseline=running_reward_baseline / (it + 1))
            pbar.update()

    loss = running_loss / len(dataloader)
    reward = running_reward / len(dataloader)
    reward_baseline = running_reward_baseline / len(dataloader)
    return loss, reward, reward_baseline


if __name__ == '__main__':
    device = torch.device('cuda')
    parser = argparse.ArgumentParser(description='Towards Local Visual Modeling for Image Captioning')
    parser.add_argument('--exp_name', type=str, default='LSTNet')
    parser.add_argument('--batch_size', type=int, default=50)
    parser.add_argument('--rl_batch_size', type=int, default=100)
    parser.add_argument('--workers', type=int, default=4)
    parser.add_argument('--m', type=int, default=40)
    parser.add_argument('--head', type=int, default=8)
    parser.add_argument('--warmup', type=int, default=10000)
    parser.add_argument('--resume_last', action='store_true')
    parser.add_argument('--resume_best', action='store_true')
    parser.add_argument('--features_path', type=str,default='/home/data/coco_grid_feats2.hdf5')
    parser.add_argument('--annotation_folder', type=str,default='/home/data/m2_annotations')
    parser.add_argument('--logs_folder', type=str, default='tensorboard_logs')
    args = parser.parse_args()
    print(args)

    print('Meshed-Memory Transformer Training')
    ia=ImageAugmentation()

    writer = SummaryWriter(log_dir=os.path.join(args.logs_folder, args.exp_name))

    # Pipeline for image regions
    image_field = ImageDetectionsField(detections_path=args.features_path, max_detections=49, load_in_tmp=False)

    # Pipeline for text
    text_field = TextField(init_token='<bos>', eos_token='<eos>', lower=True, tokenize='spacy',
                           remove_punctuation=True, nopoints=False)

    # Create the dataset
    dataset = COCO(image_field, text_field, 'coco/images/', args.annotation_folder, args.annotation_folder)
    train_dataset, val_dataset, test_dataset = dataset.splits

    # if not os.path.isfile('vocab_%s.pkl' % args.exp_name):
    #     print("Building vocabulary")
    #     text_field.build_vocab(train_dataset, val_dataset, min_freq=5)
    #     pickle.dump(text_field.vocab, open('vocab_%s.pkl' % args.exp_name, 'wb'))
    # else:
    #     text_field.vocab = pickle.load(open('vocab_%s.pkl' % args.exp_name, 'rb'))


    text_field.vocab = pickle.load(open('vocab_transformer_grid.pkl' , 'rb'))

    # Model and dataloaders
    #TransformerEncoder, TransformerDecoderLayer, ScaledDotProductAttention
    encoder = TransformerEncoder(3, 0, attention_module=ScaledDotProductAttention,
                                     attention_module_kwargs={'m': args.m})
    decoder = TransformerDecoderLayer(len(text_field.vocab), 54, 3, text_field.vocab.stoi['<pad>'])
    model = Transformer(text_field.vocab.stoi['<bos>'], encoder, decoder).to(device)

    dict_dataset_train = train_dataset.image_dictionary({'image': image_field, 'text': RawField()})
    ref_caps_train = list(train_dataset.text)
    cider_train = Cider(PTBTokenizer.tokenize(ref_caps_train))
    dict_dataset_val = val_dataset.image_dictionary({'image': image_field, 'text': RawField()})
    dict_dataset_test = test_dataset.image_dictionary({'image': image_field, 'text': RawField()})

    '''
    def lambda_lr(s):
        warm_up = args.warmup
        s += 1
        return (model.d_model ** -.5) * min(s ** -.5, s * warm_up ** -1.5)
    '''

    def lambda_lr(s):
        base_lr = 0.0001
        print("s:", s)
        if s <= 3:
            lr = base_lr * s / 4
        elif s <= 10:
            lr = base_lr
        elif s <= 20:
            lr = base_lr * 0.2
        elif s <= 25:
            lr = base_lr * 0.2 * 0.2
        elif s <=40:
            lr = 5e-6
        elif s <=50:
            lr = 5e-7
        elif s <=55:
            lr = 5e-7 *0.5
        else:
            lr = 5e-8
        # s += 1
        return lr


    # Initial conditions
    optim = Adam(model.parameters(), lr=1, betas=(0.9, 0.98))
    scheduler = LambdaLR(optim, lambda_lr)
    loss_fn = NLLLoss(ignore_index=text_field.vocab.stoi['<pad>'])
    use_rl = False
    best_cider = .0
    patience = 0
    start_epoch = 0

    if args.resume_last or args.resume_best:
        if args.resume_last:
            fname = 'saved_models/%s_last.pth' % args.exp_name
        else:
            fname = 'saved_models/%s_best.pth' % args.exp_name

        if os.path.exists(fname):
            data = torch.load(fname)
            torch.set_rng_state(data['torch_rng_state'])
            torch.cuda.set_rng_state(data['cuda_rng_state'])
            np.random.set_state(data['numpy_rng_state'])
            random.setstate(data['random_rng_state'])
            model.load_state_dict(data['state_dict'], strict=False)
            optim.load_state_dict(data['optimizer'])
            scheduler.load_state_dict(data['scheduler'])
            start_epoch = data['epoch'] + 1
            best_cider = data['best_cider']
            patience = data['patience']
            use_rl = data['use_rl']
            print('Resuming from epoch %d, validation loss %f, and best cider %f' % (
                data['epoch'], data['val_loss'], data['best_cider']))

    print("Training starts")
    for e in range(start_epoch, start_epoch + 100):
        dataloader_train = DataLoader(train_dataset, batch_size=args.batch_size, shuffle=True, num_workers=args.workers,
                                      drop_last=True)
        dataloader_val = DataLoader(val_dataset, batch_size=args.batch_size, shuffle=False, num_workers=args.workers)
        dict_dataloader_train = DataLoader(dict_dataset_train, batch_size=args. rl_batch_size // 5, shuffle=True,
                                           num_workers=args.workers)
        dict_dataloader_val = DataLoader(dict_dataset_val, batch_size=args. rl_batch_size // 5)
        dict_dataloader_test = DataLoader(dict_dataset_test, batch_size=args.rl_batch_size // 5)

        if not use_rl:
            train_loss = train_xe(model, dataloader_train, optim, text_field)
            writer.add_scalar('data/train_loss', train_loss, e)
        else:
            train_loss, reward, reward_baseline = train_scst(model, dict_dataloader_train, optim, cider_train, text_field)
            writer.add_scalar('data/train_loss', train_loss, e)
            writer.add_scalar('data/reward', reward, e)
            writer.add_scalar('data/reward_baseline', reward_baseline, e)

        # Validation loss
        val_loss = evaluate_loss(model, dataloader_val, loss_fn, text_field)
        writer.add_scalar('data/val_loss', val_loss, e)

        # Validation scores
        scores = evaluate_metrics(model, dict_dataloader_val, text_field)
        print("Validation scores", scores)
        val_cider = scores['CIDEr']
        writer.add_scalar('data/val_cider', val_cider, e)
        writer.add_scalar('data/val_bleu1', scores['BLEU'][0], e)
        writer.add_scalar('data/val_bleu4', scores['BLEU'][3], e)
        writer.add_scalar('data/val_meteor', scores['METEOR'], e)
        writer.add_scalar('data/val_rouge', scores['ROUGE'], e)

        # Test scores
        scores = evaluate_metrics(model, dict_dataloader_test, text_field)
        print("Test scores", scores)
        test_cider = scores['CIDEr']
        writer.add_scalar('data/test_cider', scores['CIDEr'], e)
        writer.add_scalar('data/test_bleu1', scores['BLEU'][0], e)
        writer.add_scalar('data/test_bleu4', scores['BLEU'][3], e)
        writer.add_scalar('data/test_meteor', scores['METEOR'], e)
        writer.add_scalar('data/test_rouge', scores['ROUGE'], e)

        # Prepare for next epoch
        best = False
        if val_cider >= best_cider:
            best_cider = val_cider
            patience = 0
            best = True
        else:
            patience += 1


        switch_to_rl = False
        exit_train = False

        if e == 25:
            if not use_rl:
                use_rl = True
                switch_to_rl = True
                patience = 0
                optim = Adam(model.parameters(), lr=1)
                scheduler = LambdaLR(optim, lambda_lr)
                for i in range(25):
                    scheduler.step()
                print("Switching to RL")
        if e== 60:
            print('patience reached.')
            exit_train = True

        # if patience == 5:
        #     if not use_rl:
        #         use_rl = True
        #         switch_to_rl = True
        #         patience = 0
        #         optim = Adam(model.parameters(), lr=5e-6)
        #         print("Switching to RL")
        #     else:
        #         print('patience reached.')
        #         exit_train = True

        if switch_to_rl and not best:
            data = torch.load('saved_models/%s_best.pth' % args.exp_name)
            torch.set_rng_state(data['torch_rng_state'])
            torch.cuda.set_rng_state(data['cuda_rng_state'])
            np.random.set_state(data['numpy_rng_state'])
            random.setstate(data['random_rng_state'])
            model.load_state_dict(data['state_dict'])
            print('Resuming from epoch %d, validation loss %f, and best cider %f' % (
                data['epoch'], data['val_loss'], data['best_cider']))

        torch.save({
            'torch_rng_state': torch.get_rng_state(),
            'cuda_rng_state': torch.cuda.get_rng_state(),
            'numpy_rng_state': np.random.get_state(),
            'random_rng_state': random.getstate(),
            'epoch': e,
            'val_loss': val_loss,
            'val_cider': val_cider,
            'state_dict': model.state_dict(),
            'optimizer': optim.state_dict(),
            'scheduler': scheduler.state_dict(),
            'patience': patience,
            'best_cider': best_cider,
            'use_rl': use_rl,
        }, 'saved_models/%s_last.pth' % args.exp_name)

        if best:
            copyfile('saved_models/%s_last.pth' % args.exp_name, 'saved_models/%s_best.pth' % args.exp_name)


        if exit_train:
            writer.close()
            break