Train from --data path/to/dataset.zip feature (#4185)

* Train from `--data path/to/dataset.zip` feature

* Update dataset_stats()

* cleanup

* cleanup2

train.py

@@ -35,7 +35,7 @@ from utils.datasets import create_dataloader
 from utils.general import labels_to_class_weights, increment_path, labels_to_image_weights, init_seeds, \
     strip_optimizer, get_latest_run, check_dataset, check_file, check_git_status, check_img_size, \
     check_requirements, print_mutation, set_logging, one_cycle, colorstr
-from utils.google_utils import attempt_download
+from utils.downloads import attempt_download
 from utils.loss import ComputeLoss
 from utils.plots import plot_labels, plot_evolution
 from utils.torch_utils import ModelEMA, select_device, intersect_dicts, torch_distributed_zero_first, de_parallel
@@ -78,9 +78,9 @@ def train(hyp,  # path/to/hyp.yaml or hyp dictionary
     plots = not evolve  # create plots
     cuda = device.type != 'cpu'
     init_seeds(1 + RANK)
-    with open(data, encoding='ascii', errors='ignore') as f:
-        data_dict = yaml.safe_load(f)
-
+    with torch_distributed_zero_first(RANK):
+        data_dict = check_dataset(data)  # check
+    train_path, val_path = data_dict['train'], data_dict['val']
     nc = 1 if single_cls else int(data_dict['nc'])  # number of classes
     names = ['item'] if single_cls and len(data_dict['names']) != 1 else data_dict['names']  # class names
     assert len(names) == nc, f'{len(names)} names found for nc={nc} dataset in {data}'  # check
@@ -106,9 +106,6 @@ def train(hyp,  # path/to/hyp.yaml or hyp dictionary
         LOGGER.info(f'Transferred {len(csd)}/{len(model.state_dict())} items from {weights}')  # report
     else:
         model = Model(cfg, ch=3, nc=nc, anchors=hyp.get('anchors')).to(device)  # create
-    with torch_distributed_zero_first(RANK):
-        check_dataset(data_dict)  # check
-    train_path, val_path = data_dict['train'], data_dict['val']
 
     # Freeze
     freeze = []  # parameter names to freeze (full or partial)