Override fixes and general updates (#129)

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Kalen Michael <kalenmike@gmail.com>

tests/test_model.py

@@ -6,16 +6,11 @@ from ultralytics import YOLO
 def test_model_init():
     model = YOLO("yolov8n.yaml")
     model.info()
-    try:
-        YOLO()
-    except Exception:
-        print("Successfully caught constructor assert!")
-    raise Exception("constructor error didn't occur")
 
 
 def test_model_forward():
     model = YOLO("yolov8n.yaml")
-    img = torch.rand(512 * 512 * 3).view(1, 3, 512, 512)
+    img = torch.rand(1, 3, 320, 320)
     model.forward(img)
     model(img)
 
@@ -23,24 +18,24 @@ def test_model_forward():
 def test_model_info():
     model = YOLO("yolov8n.yaml")
     model.info()
-    model = model.load("best.pt")
+    model = YOLO("yolov8n.pt")
     model.info(verbose=True)
 
 
 def test_model_fuse():
     model = YOLO("yolov8n.yaml")
     model.fuse()
-    model.load("best.pt")
+    model = YOLO("yolov8n.pt")
     model.fuse()
 
 
 def test_visualize_preds():
-    model = YOLO("best.pt")
+    model = YOLO("yolov8n.pt")
     model.predict(source="ultralytics/assets")
 
 
 def test_val():
-    model = YOLO("best.pt")
+    model = YOLO("yolov8n.pt")
     model.val(data="coco128.yaml", imgsz=32)
 
 
@@ -54,11 +49,11 @@ def test_model_resume():
 
 
 def test_model_train_pretrained():
-    model = YOLO("best.pt")
+    model = YOLO("yolov8n.pt")
     model.train(data="coco128.yaml", epochs=1, imgsz=32)
-    model = model.new("yolov8n.yaml")
+    model = YOLO("yolov8n.yaml")
     model.train(data="coco128.yaml", epochs=1, imgsz=32)
-    img = torch.rand(512 * 512 * 3).view(1, 3, 512, 512)
+    img = torch.rand(1, 3, 320, 320)
     model(img)
 
 
@@ -78,7 +73,6 @@ def test_exports():
     10          TensorFlow.js         tfjs       _web_model  False  False
     11           PaddlePaddle       paddle    _paddle_model   True   True
     """
-    from ultralytics import YOLO
     from ultralytics.yolo.engine.exporter import export_formats
 
     print(export_formats())

ultralytics/hub/utils.py

@@ -1,15 +1,15 @@
 import shutil
 import threading
 import time
-import uuid
 
 import requests
 
 from ultralytics.hub.config import HUB_API_ROOT
-from ultralytics.yolo.utils import LOGGER, RANK, SETTINGS, colorstr, emojis
+from ultralytics.yolo.utils import DEFAULT_CONFIG, LOGGER, RANK, SETTINGS, colorstr, emojis, yaml_load
 
 PREFIX = colorstr('Ultralytics: ')
 HELP_MSG = 'If this issue persists please visit https://github.com/ultralytics/hub/issues for assistance.'
+DEFAULT_CONFIG_DICT = yaml_load(DEFAULT_CONFIG)
 
 
 def check_dataset_disk_space(url='https://github.com/ultralytics/yolov5/releases/download/v1.0/coco128.zip', sf=2.0):
@@ -26,7 +26,7 @@ def check_dataset_disk_space(url='https://github.com/ultralytics/yolov5/releases
 
 
 def request_with_credentials(url: str) -> any:
-    """ Make a ajax request with cookies attached """
+    """ Make an ajax request with cookies attached """
     from google.colab import output  # noqa
     from IPython import display  # noqa
     display.display(
@@ -92,17 +92,18 @@ def smart_request(*args, retry=3, timeout=30, thread=True, code=-1, method="post
     retry_codes = (408, 500)  # retry only these codes
     methods = {'post': requests.post, 'get': requests.get}  # request methods
 
-    def fcn(*args, **kwargs):
-        t0 = time.time()
+    def func(*func_args, **func_kwargs):
+        r = None  # response
+        t0 = time.time()  # initial time for timer
         for i in range(retry + 1):
             if (time.time() - t0) > timeout:
                 break
-            r = methods[method](*args, **kwargs)  # i.e. post(url, data, json, files)
+            r = methods[method](*func_args, **func_kwargs)  # i.e. post(url, data, json, files)
             if r.status_code == 200:
                 break
             try:
                 m = r.json().get('message', 'No JSON message.')
-            except Exception:
+            except AttributeError:
                 m = 'Unable to read JSON.'
             if i == 0:
                 if r.status_code in retry_codes:
@@ -118,22 +119,25 @@ def smart_request(*args, retry=3, timeout=30, thread=True, code=-1, method="post
         return r
 
     if thread:
-        threading.Thread(target=fcn, args=args, kwargs=kwargs, daemon=True).start()
+        threading.Thread(target=func, args=args, kwargs=kwargs, daemon=True).start()
     else:
-        return fcn(*args, **kwargs)
+        return func(*args, **kwargs)
 
 
-def sync_analytics(cfg, enabled=False):
+def sync_analytics(cfg, all_keys=False, enabled=True):
     """
    Sync analytics data if enabled in the global settings
 
     Args:
         cfg (DictConfig): Configuration for the task and mode.
+        all_keys (bool): Sync all items, not just non-default values.
         enabled (bool): For debugging.
     """
     if SETTINGS['sync'] and RANK in {-1, 0} and enabled:
         cfg = dict(cfg)  # convert type from DictConfig to dict
-        cfg['uuid'] = uuid.getnode()  # add the device UUID to the configuration data
+        if not all_keys:
+            cfg = {k: v for k, v in cfg.items() if v != DEFAULT_CONFIG_DICT[k]}  # retain only non-default values
+        cfg['uuid'] = SETTINGS['uuid']  # add the device UUID to the configuration data
 
         # Send a request to the HUB API to sync the analytics data
-        smart_request(f'{HUB_API_ROOT}/analytics', data=cfg, headers=None, code=3, retry=0)
+        smart_request(f'{HUB_API_ROOT}/v1/usage/anonymous', data=cfg, headers=None, code=3, retry=0)

ultralytics/nn/tasks.py

@@ -12,8 +12,8 @@ from ultralytics.nn.modules import (C1, C2, C3, C3TR, SPP, SPPF, Bottleneck, Bot
                                     GhostBottleneck, GhostConv, Segment)
 from ultralytics.yolo.utils import LOGGER, colorstr, yaml_load
 from ultralytics.yolo.utils.checks import check_yaml
-from ultralytics.yolo.utils.torch_utils import (fuse_conv_and_bn, initialize_weights, intersect_state_dicts,
-                                                make_divisible, model_info, scale_img, time_sync)
+from ultralytics.yolo.utils.torch_utils import (fuse_conv_and_bn, initialize_weights, intersect_dicts, make_divisible,
+                                                model_info, scale_img, time_sync)
 
 
 class BaseModel(nn.Module):
@@ -150,7 +150,7 @@ class DetectionModel(BaseModel):
 
     def load(self, weights, verbose=True):
         csd = weights['model'].float().state_dict()  # checkpoint state_dict as FP32
-        csd = intersect_state_dicts(csd, self.state_dict())  # intersect
+        csd = intersect_dicts(csd, self.state_dict())  # intersect
         self.load_state_dict(csd, strict=False)  # load
         if verbose:
             LOGGER.info(f'Transferred {len(csd)}/{len(self.model.state_dict())} items from pretrained weights')
@@ -191,7 +191,7 @@ class ClassificationModel(BaseModel):
     def load(self, weights):
         model = weights["model"] if isinstance(weights, dict) else weights  # torchvision models are not dicts
         csd = model.float().state_dict()
-        csd = intersect_state_dicts(csd, self.state_dict())  # intersect
+        csd = intersect_dicts(csd, self.state_dict())  # intersect
         self.load_state_dict(csd, strict=False)  # load
 
     @staticmethod

ultralytics/yolo/configs/default.yaml

@@ -1,11 +1,11 @@
 # YOLO 🚀 by Ultralytics, GPL-3.0 license
 # Default training settings and hyperparameters for medium-augmentation COCO training
 
-task: "classify" # choices=['detect', 'segment', 'classify', 'init'] # init is a special case. Specify task to run.
+task: "detect" # choices=['detect', 'segment', 'classify', 'init'] # init is a special case. Specify task to run.
 mode: "train" # choices=['train', 'val', 'predict'] # mode to run task in.
 
 # Train settings -------------------------------------------------------------------------------------------------------
-model: null # i.e. yolov5s.pt, yolo.yaml. Path to model file
+model: null # i.e. yolov8n.pt, yolov8n.yaml. Path to model file
 data: null # i.e. coco128.yaml. Path to data file
 epochs: 100 # number of epochs to train for
 patience: 50  # TODO: epochs to wait for no observable improvement for early stopping of training

ultralytics/yolo/engine/exporter.py

@@ -137,8 +137,6 @@ class Exporter:
         """
         if overrides is None:
             overrides = {}
-        if 'batch_size' not in overrides:
-            overrides['batch_size'] = 1  # set default export batch size
         self.args = get_config(config, overrides)
         project = self.args.project or f"runs/{self.args.task}"
         name = self.args.name or "exp"  # hardcode mode as export doesn't require it
@@ -166,6 +164,8 @@ class Exporter:
             assert not self.args.dynamic, '--half not compatible with --dynamic, i.e. use either --half or --dynamic'
 
         # Checks
+        if self.args.batch_size == 16:
+            self.args.batch_size = 1  # TODO: resolve batch_size 16 default in config.yaml
         self.imgsz = check_imgsz(self.args.imgsz, stride=model.stride, min_dim=2)  # check image size
         if self.args.optimize:
             assert self.device.type == 'cpu', '--optimize not compatible with cuda devices, i.e. use --device cpu'

ultralytics/yolo/engine/model.py

@@ -1,6 +1,7 @@
-import torch
 from pathlib import Path
 
+import torch
+
 from ultralytics import yolo  # noqa
 from ultralytics.nn.tasks import ClassificationModel, DetectionModel, SegmentationModel, attempt_load_weights
 from ultralytics.yolo.configs import get_config

ultralytics/yolo/engine/trainer.py

@@ -107,6 +107,8 @@ class BaseTrainer:
         self.device = utils.torch_utils.select_device(self.args.device, self.batch_size)
         self.amp = self.device.type != 'cpu'
         self.scaler = amp.GradScaler(enabled=self.amp)
+        if self.device.type == 'cpu':
+            self.args.workers = 0  # faster CPU training as time dominated by inference, not dataloading
 
         # Model and Dataloaders.
         self.model = self.args.model

ultralytics/yolo/utils/__init__.py

@@ -6,6 +6,7 @@ import platform
 import sys
 import tempfile
 import threading
+import uuid
 from pathlib import Path
 
 import cv2
@@ -160,7 +161,7 @@ def get_user_config_dir(sub_dir='Ultralytics'):
         raise ValueError(f'Unsupported operating system: {os_name}')
 
     # GCP and AWS lambda fix, only /tmp is writeable
-    if not is_dir_writeable(path.parent):
+    if not is_dir_writeable(str(path.parent)):
         path = Path('/tmp') / sub_dir
 
     # Create the subdirectory if it does not exist
@@ -172,9 +173,9 @@ def get_user_config_dir(sub_dir='Ultralytics'):
 USER_CONFIG_DIR = get_user_config_dir()  # Ultralytics settings dir
 
 
-def emojis(str=''):
+def emojis(string=''):
     # Return platform-dependent emoji-safe version of string
-    return str.encode().decode('ascii', 'ignore') if platform.system() == 'Windows' else str
+    return string.encode().decode('ascii', 'ignore') if platform.system() == 'Windows' else string
 
 
 def colorstr(*input):
@@ -282,27 +283,39 @@ def yaml_load(file='data.yaml'):
     """
     with open(file, errors='ignore') as f:
         # Add YAML filename to dict and return
-        return {**yaml.safe_load(f), 'yaml_file': file}
+        return {**yaml.safe_load(f), 'yaml_file': str(file)}
 
 
 def get_settings(file=USER_CONFIG_DIR / 'settings.yaml'):
     """
-    Function that loads a global settings YAML, or creates it and populates it with default values if it does not exist.
+    Loads a global settings YAML file or creates one with default values if it does not exist.
 
-    If the datasets or weights directories are set to None, the current working directory will be used.
-    The 'sync' setting determines whether analytics will be synced to help with YOLO development.
+    Args:
+        file (Path): Path to the settings YAML file. Defaults to 'settings.yaml' in the USER_CONFIG_DIR.
+
+    Returns:
+        dict: Dictionary of settings key-value pairs.
     """
     from ultralytics.yolo.utils.torch_utils import torch_distributed_zero_first
 
-    with torch_distributed_zero_first(RANK):
-        if not file.exists():
-            settings = {
+    defaults = {
         'datasets_dir': None,  # default datasets directory. If None, current working directory is used.
         'weights_dir': None,  # default weights directory. If None, current working directory is used.
-                'sync': True}  # sync analytics to help with YOLO development
-            yaml_save(file, settings)
+        'runs_dir': None,  # default runs directory. If None, current working directory is used.
+        'sync': True,  # sync analytics to help with YOLO development
+        'uuid': uuid.getnode(),  # device UUID to align analytics
+        'yaml_file': str(file)}  # setting YAML file path
+
+    with torch_distributed_zero_first(RANK):
+        if not file.exists():
+            yaml_save(file, defaults)
+
+        settings = yaml_load(file)
+        if settings.keys() != defaults.keys():
+            settings = {**defaults, **settings}  # merge **defaults with **settings (prefer **settings)
+            yaml_save(file, settings)  # save updated defaults
 
-    return yaml_load(file)
+        return settings
 
 
 # Run below code on utils init -----------------------------------------------------------------------------------------

ultralytics/yolo/utils/callbacks/hub.py

@@ -48,7 +48,7 @@ def on_train_end(trainer):
         # Upload final model and metrics with exponential standoff
         LOGGER.info(f"{PREFIX}Training completed successfully ✅\n"
                     f"{PREFIX}Uploading final {session.model_id}")
-        session.upload_model(trainer.epoch, trainer.best, map=trainer.metrics['metrics/mAP50(B)'], final=True)
+        session.upload_model(trainer.epoch, trainer.best, map=trainer.metrics['metrics/mAP50-95(B)'], final=True)
         session.alive = False  # stop heartbeats
         LOGGER.info(f"{PREFIX}View model at https://hub.ultralytics.com/models/{session.model_id} 🚀")
 

ultralytics/yolo/utils/torch_utils.py

@@ -201,7 +201,7 @@ def copy_attr(a, b, include=(), exclude=()):
             setattr(a, k, v)
 
 
-def intersect_state_dicts(da, db, exclude=()):
+def intersect_dicts(da, db, exclude=()):
     # Dictionary intersection of matching keys and shapes, omitting 'exclude' keys, using da values
     return {k: v for k, v in da.items() if k in db and all(x not in k for x in exclude) and v.shape == db[k].shape}