def single_model_predict_tta():
assert len(model_name_list) == 1
model_name = model_name_list[0]
model = Net(model_name).to(device)
model_save_path = os.path.join(
config.dir_weight, '{}.bin'.format(model_name))
model.load_state_dict(torch.load(model_save_path))
transforms = tta.Compose([
tta.HorizontalFlip(),
# tta.Rotate90(angles=[0, 180]),
# tta.Scale(scales=[1, 2, 4]),
# tta.Multiply(factors=[0.9, 1, 1.1]),
tta.FiveCrops(224, 224)
])
tta_model = tta.ClassificationTTAWrapper(model, transforms)
pred_list = []
with torch.no_grad():
for batch_x, _ in tqdm(test_loader):
batch_x = batch_x.to(device)
probs = tta_model(batch_x)
probs = torch.max(torch.softmax(probs, dim=1), dim=1)
probs = probs[1].cpu().numpy()
pred_list += probs.tolist()
submission = pd.DataFrame({
"id": range(len(pred_list)),
"label": [int2label(x) for x in pred_list]
})
submission.to_csv(config.dir_csv_test, index=False, header=False)
def multi_model_predict_tta():
preds_dict = dict()
for model_name in model_name_list:
for fold_idx in range(5):
model = Net(model_name).to(device)
model_save_path = os.path.join(
config.dir_weight, '{}_fold{}.bin'.format(model_name, fold_idx))
model.load_state_dict(torch.load(model_save_path))
'/home/muyun99/data/dataset/AIyanxishe/Image_Classification/weight/resnet18_train_size_256_fold0.bin'
transforms = tta.Compose([
tta.Resize([int(config.size_test_image), int(config.size_test_image)]),
tta.HorizontalFlip(),
# tta.Rotate90(angles=[0, 180]),
# tta.Scale(scales=[1, 2, 4]),
# tta.Multiply(factors=[0.9, 1, 1.1]),
tta.FiveCrops(config.size_test_image, config.size_test_image)
])
tta_model = tta.ClassificationTTAWrapper(model, transforms)
pred_list = predict(tta_model)
submission = pd.DataFrame(pred_list)
submission.to_csv(
'{}/{}_fold{}_submission.csv'.format(config.dir_submission, config.save_model_name, fold_idx),
index=False,
header=False
)
preds_dict['{}_{}'.format(model_name, fold_idx)] = pred_list
pred_list = get_pred_list(preds_dict)
submission = pd.DataFrame(
{"id": range(len(pred_list)), "label": [int2label(x) for x in pred_list]})
submission.to_csv(config.dir_csv_test, index=False, header=False)
def main():
load_dotenv('cassava.env')
seed_everything(SEED)
root_path = os.getenv('ROOT_PATH')
train_csv_path = root_path + 'train.csv'
train_root_path = root_path + 'train_images'
num_classes = int(os.getenv('NUM_CLASSES', 5))
num_epoch = int(os.getenv('NUM_EPOCH', 10))
num_folds = int(os.getenv('NUM_FOLDS', 5))
batch_size = int(os.getenv('BATCH_SIZE'), 16)
grad_acc = int(os.getenv('GRAD_ACC', 8))
resize = os.getenv('RESIZE', 224)
normalize = A.Normalize(mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225])
train_transform = A.Compose([
A.HorizontalFlip(),
A.ShiftScaleRotate(p=1.0),
A.ColorJitter(brightness=0.1, contrast=0.2, saturation=0.2, hue=0.0, p=1.0, always_apply=False),
A.RandomResizedCrop(resize, resize, p=1.0, always_apply=True),
normalize,
ToTensorV2(p=1.0),
], p=1.0)
test_transform = A.Compose([
A.Resize(int(resize * 1.5), int(resize * 1.5)),
normalize,
ToTensorV2(p=1.0),
], p=1.0)
tta_transform = tta.Compose([
tta.FiveCrops(resize, resize),
])
criterion = MixedLabelLoss(nn.CrossEntropyLoss(reduction='none'))
augmentations = [snapmix, ]
df = pd.read_csv(train_csv_path)
folds = StratifiedKFold(n_splits=num_folds, shuffle=True, random_state=SEED).split(df['image_id'], df['label'])
for _fold, (train, test) in enumerate(folds):
train = df.iloc[train]
test = df.iloc[test]
scheduler = optim.lr_scheduler.CosineAnnealingLR
model = TimmNet('efficientnet_b3a', num_classes, criterion, learning_rate=1e-3, scheduler=scheduler,
n_epoch=num_epoch, eta_min=1e-6, augmentations=augmentations, tta_transform=tta_transform)
dm = DataFrameDataModule(train, train_root_path, test, batch_size=batch_size,
train_transform=train_transform, test_transform=test_transform)
mlf_logger = MLFlowLogger(
experiment_name='cassava',
tracking_uri='file:./cassava'
)
trainer = Trainer(gpus=-1, precision=32, deterministic=True, accumulate_grad_batches=grad_acc,
profiler='simple', val_check_interval=1.0, logger=mlf_logger, max_epochs=num_epoch)
trainer.fit(model, datamodule=dm)
def test_fivecrop_transform():
transform = tta.FiveCrops(crop_height=1, crop_width=1)
a = torch.arange(25).reshape(1, 1, 5, 5).float()
output = [0, 20, 24, 4, 12]
for i, p in enumerate(transform.params):
aug = transform.apply_aug_image(a, **{transform.pname: p})
assert aug.item() == output[i]
#
# def test_resize_transform():
# transform = tta.Resize(sizes=[(10, 10), (5, 5)], original_size=(5, 5))
# a = torch.arange(25).reshape(1, 1, 5, 5).float()
# for i, p in enumerate(transform.params):
# aug = transform.apply_aug_image(a, **{transform.pname: p})
# assert aug.item() == output[i]
for p in transform.params:
aug = transform.apply_aug_image(a, **{transform.pname: p})
deaug = transform.apply_deaug_mask(aug, **{transform.pname: p})
assert torch.allclose(a, deaug)
@pytest.mark.parametrize(
"transform",
[
tta.HorizontalFlip(),
tta.VerticalFlip(),
tta.Rotate90(angles=[0, 90, 180, 270]),
tta.Scale(scales=[1, 2, 4], interpolation="nearest"),
tta.Add(values=[-1, 0, 1, 2]),
tta.Multiply(factors=[-1, 0, 1, 2]),
tta.FiveCrops(crop_height=3, crop_width=5),
tta.Resize(sizes=[(4, 5), (8, 10), (2, 2)], interpolation="nearest")
],
)
def test_label_is_same(transform):
a = torch.arange(20).reshape(1, 1, 4, 5).float()
for p in transform.params:
aug = transform.apply_aug_image(a, **{transform.pname: p})
deaug = transform.apply_deaug_label(aug, **{transform.pname: p})
assert torch.allclose(aug, deaug)
def test_add_transform():
transform = tta.Add(values=[-1, 0, 1])
a = torch.arange(20).reshape(1, 1, 4, 5).float()
for p in transform.params:
import time
from progress.bar import Bar
import torchvision.transforms as transforms
from dataloader.EyeQ_loader import DatasetGenerator
from utils.trainer_unc_origin import train_step, validation_step, save_output
from utils.metric_origin import compute_metric
from refuge.cropDiscRegion import cropDiscRegion
import pandas as pd
from networks.efficient_origin import EfficientNet_b0, SENet154, Xception, ResNext50, DenseNet121_v0, ResNest, ResNest_unc
import ttach as tta
import natsort
from PIL import Image
import cv2
from eval import eval
transforms_tta = tta.Compose([tta.FiveCrops(224, 224)])
os.environ["CUDA_VISIBLE_DEVICES"] = "2"
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
np.random.seed(0)
# Setting parameters
parser = argparse.ArgumentParser(description='EyeQ_dense121')
parser.add_argument('--model_dir',
type=str,
default='./result_origin_unc_refuge/')
parser.add_argument('--pre_model', type=str, default=True)
parser.add_argument('--save_model', type=str, default='classification_results')
parser.add_argument('--crop_size', type=int, default=224)