return new_cmap
if __name__ == '__main__':
parse_args_and_merge_const()
random_state = np.random.RandomState(const.RANDOM_SEED)
if os.path.exists('models') is False:
os.makedirs('models')
df = pd.read_csv(const.base_path + const.USE_CSV)
inf_df = df
inf_dataset = DeepFashionCAPDataset(inf_df,
random_state=random_state,
mode=const.DATASET_PROC_METHOD_INF,
base_path=const.base_path)
inf_dataloader = torch.utils.data.DataLoader(
inf_dataset,
batch_size=1, #const.INF_BATCH_SIZE,
shuffle=False,
num_workers=6)
inf_step = len(inf_dataloader)
net = const.USE_NET(const.USE_IORN)
net = net.to(const.device)
net.load_state_dict(torch.load(const.INIT_MODEL), strict=False)
writer = SummaryWriter(const.INF_DIR)
inf_step = len(inf_dataloader)
random_state = np.random.RandomState(const.RANDOM_SEED)
# const.DATASET_PROC_METHOD_TRAIN = 'ELASTIC_ROTATION_BBOXRESIZE'
# const.DATASET_PROC_METHOD_TRAIN = 'ROTATION_BBOXRESIZE'
const.DATASET_PROC_METHOD_TRAIN = 'BBOXRESIZE'
const.USE_CSV = 'info.csv'
# const.USE_CSV = 'debug_info.csv'
if os.path.exists('models') is False:
os.makedirs('models')
print(const.base_path)
df = pd.read_csv(const.base_path + const.USE_CSV)
train_df = df[df['evaluation_status'] == 'val']
train_dataset = DeepFashionCAPDataset(train_df,
base_path=const.base_path,
random_state=random_state,
mode=const.DATASET_PROC_METHOD_TRAIN)
train_dataloader = torch.utils.data.DataLoader(train_dataset,
batch_size=const.BATCH_SIZE,
shuffle=True,
num_workers=4)
# val_df = df[df['evaluation_status'] == 'val']
# val_dataset = DeepFashionCAPDataset(val_df, mode=const.DATASET_PROC_METHOD_VAL)
# val_dataloader = torch.utils.data.DataLoader(val_dataset, batch_size=const.VAL_BATCH_SIZE, shuffle=False, num_workers=1)
# for i in range(100):
# train_dataset.plot_sample(i)
#
step = 0
for i, sample in enumerate(train_dataloader):
step += 1
# Initialize random generators with given seed
if const.RANDOM_SEED != None:
torch.manual_seed(const.RANDOM_SEED)
np.random.seed(const.RANDOM_SEED)
random.seed(const.RANDOM_SEED)
random_state = np.random.RandomState(const.RANDOM_SEED)
if os.path.exists('models') is False:
os.makedirs('models')
df = pd.read_csv(const.base_path + const.USE_CSV)
train_df = df[df['evaluation_status'] == 'train']
train_dataset = DeepFashionCAPDataset(train_df,
random_state=random_state,
mode=const.DATASET_PROC_METHOD_TRAIN,
base_path=const.base_path,
el_alpha=const.EL_ALPHA,
el_sigma=const.EL_SIGMA)
train_dataloader = torch.utils.data.DataLoader(train_dataset,
batch_size=const.BATCH_SIZE,
shuffle=True,
num_workers=8)
val_df = df[df['evaluation_status'] == 'val']
val_dataset = DeepFashionCAPDataset(val_df,
random_state=random_state,
mode=const.DATASET_PROC_METHOD_VAL,
base_path=const.base_path)
val_dataloader = torch.utils.data.DataLoader(val_dataset,
batch_size=const.VAL_BATCH_SIZE,
shuffle=False,
num_workers=8)
import pandas as pd
import torch
import torch.utils.data
from src import const
from src.utils import parse_args_and_merge_const
from tensorboardX import SummaryWriter
import os
if __name__ == '__main__':
parse_args_and_merge_const()
if os.path.exists('models') is False:
os.makedirs('models')
df = pd.read_csv(base_path + const.USE_CSV)
train_df = df[df['evaluation_status'] == 'train']
train_dataset = DeepFashionCAPDataset(train_df,
mode=const.DATASET_PROC_METHOD_TRAIN)
train_dataloader = torch.utils.data.DataLoader(train_dataset,
batch_size=const.BATCH_SIZE,
shuffle=True,
num_workers=4)
val_df = df[df['evaluation_status'] == 'test']
val_dataset = DeepFashionCAPDataset(val_df,
mode=const.DATASET_PROC_METHOD_VAL)
val_dataloader = torch.utils.data.DataLoader(
val_dataset,
batch_size=const.VAL_BATCH_SIZE,
shuffle=False,
num_workers=4)
val_step = len(val_dataloader)
net = const.USE_NET()
"{0}, (label: {1})".format(
classes[preds[idx].max(dim=0)[1].item()],
classes[int(labels[idx].item())]),
color=("green" if preds[idx].max(
dim=0)[1].item() == int(labels[idx].item()) else "red"))
return fig
if __name__ == '__main__':
parse_args_and_merge_const()
if os.path.exists('models') is False:
os.makedirs('models')
df = pd.read_csv(base_path + const.USE_CSV)
test_df = df[df['evaluation_status'] == 'test']
test_dataset = DeepFashionCAPDataset(test_df,
mode=const.DATASET_PROC_METHOD_TRAIN)
test_dataloader = torch.utils.data.DataLoader(test_dataset,
batch_size=const.BATCH_SIZE,
shuffle=True,
num_workers=4)
test_step = len(test_dataloader)
net = const.USE_NET()
net.load_state_dict(torch.load(const.save_model_path))
net = net.to(const.device)
writer = SummaryWriter(const.VAL_DIR)
step = 0
print("Start Evaluate")
net.eval()