ultralytics 8.0.171 new SAHI guide and callbacks fix (#4748)
Co-authored-by: chuzihang <49548797+Aria-Leo@users.noreply.github.com> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Sanghyun Choi <farewell518@gmail.com> Co-authored-by: Awsome <1579093407@qq.com> Co-authored-by: ConanQZ <49194386+ConanQZ@users.noreply.github.com>
This commit is contained in:
Glenn Jocher
GitHub
parent
aa9133bb88
commit
577d066fb2
15 changed files with 215 additions and 35 deletions
|
|
@ -1,6 +1,6 @@
|
|||
# Ultralytics YOLO 🚀, AGPL-3.0 license
|
||||
|
||||
__version__ = '8.0.170'
|
||||
__version__ = '8.0.171'
|
||||
|
||||
from ultralytics.models import RTDETR, SAM, YOLO
|
||||
from ultralytics.models.fastsam import FastSAM
|
||||
|
|
|
|||
|
|
@ -30,6 +30,12 @@ model.train(data="coco128.yaml", epochs=100) # train the model
|
|||
|
||||
Ultralytics supports many model architectures. Visit https://docs.ultralytics.com/models to view detailed information and usage. Any of these models can be used by loading their configs or pretrained checkpoints if available.
|
||||
|
||||
## Contributing New Models
|
||||
## Contribute New Models
|
||||
|
||||
If you've developed a new model architecture or have improvements for existing models that you'd like to contribute to the Ultralytics community, please submit your contribution in a new Pull Request. For more details, visit our [Contributing Guide](https://docs.ultralytics.com/help/contributing).
|
||||
Have you trained a new YOLO variant or achieved state-of-the-art performance with specific tuning? We'd love to showcase your work in our Models section! Contributions from the community in the form of new models, architectures, or optimizations are highly valued and can significantly enrich our repository.
|
||||
|
||||
By contributing to this section, you're helping us offer a wider array of model choices and configurations to the community. It's a fantastic way to share your knowledge and expertise while making the Ultralytics YOLO ecosystem even more versatile.
|
||||
|
||||
To get started, please consult our [Contributing Guide](https://docs.ultralytics.com/help/contributing) for step-by-step instructions on how to submit a Pull Request (PR) 🛠️. Your contributions are eagerly awaited!
|
||||
|
||||
Let's join hands to extend the range and capabilities of the Ultralytics YOLO models 🙏!
|
||||
|
|
|
|||
|
|
@ -121,11 +121,11 @@ class BasePredictor:
|
|||
im = np.ascontiguousarray(im) # contiguous
|
||||
im = torch.from_numpy(im)
|
||||
|
||||
img = im.to(self.device)
|
||||
img = img.half() if self.model.fp16 else img.float() # uint8 to fp16/32
|
||||
im = im.to(self.device)
|
||||
im = im.half() if self.model.fp16 else im.float() # uint8 to fp16/32
|
||||
if not_tensor:
|
||||
img /= 255 # 0 - 255 to 0.0 - 1.0
|
||||
return img
|
||||
im /= 255 # 0 - 255 to 0.0 - 1.0
|
||||
return im
|
||||
|
||||
def inference(self, im, *args, **kwargs):
|
||||
visualize = increment_path(self.save_dir / Path(self.batch[0][0]).stem,
|
||||
|
|
|
|||
|
|
@ -42,7 +42,7 @@ class ClassificationValidator(BaseValidator):
|
|||
"""Initialize confusion matrix, class names, and top-1 and top-5 accuracy."""
|
||||
self.names = model.names
|
||||
self.nc = len(model.names)
|
||||
self.confusion_matrix = ConfusionMatrix(nc=self.nc, task='classify')
|
||||
self.confusion_matrix = ConfusionMatrix(nc=self.nc, conf=self.args.conf, task='classify')
|
||||
self.pred = []
|
||||
self.targets = []
|
||||
|
||||
|
|
|
|||
|
|
@ -68,7 +68,7 @@ class DetectionValidator(BaseValidator):
|
|||
self.nc = len(model.names)
|
||||
self.metrics.names = self.names
|
||||
self.metrics.plot = self.args.plots
|
||||
self.confusion_matrix = ConfusionMatrix(nc=self.nc)
|
||||
self.confusion_matrix = ConfusionMatrix(nc=self.nc, conf=self.args.conf)
|
||||
self.seen = 0
|
||||
self.jdict = []
|
||||
self.stats = []
|
||||
|
|
|
|||
|
|
@ -83,6 +83,12 @@ yolo pose track source=... tracker=...
|
|||
|
||||
By default, trackers will use the configuration in `ultralytics/cfg/trackers`. We also support using a modified tracker config file. Please refer to the tracker config files in `ultralytics/cfg/trackers`.
|
||||
|
||||
## Contributing New Trackers
|
||||
## Contribute to Our Trackers Section
|
||||
|
||||
If you've developed a new tracker architecture or have improvements for existing trackers that you'd like to contribute to the Ultralytics community, please submit your contribution in a new Pull Request. For more details, visit our [Contributing Guide](https://docs.ultralytics.com/help/contributing).
|
||||
Are you proficient in multi-object tracking and have successfully implemented or adapted a tracking algorithm with Ultralytics YOLO? We invite you to contribute to our Trackers section! Your real-world applications and solutions could be invaluable for users working on tracking tasks.
|
||||
|
||||
By contributing to this section, you help expand the scope of tracking solutions available within the Ultralytics YOLO framework, adding another layer of functionality and utility for the community.
|
||||
|
||||
To initiate your contribution, please refer to our [Contributing Guide](https://docs.ultralytics.com/help/contributing) for comprehensive instructions on submitting a Pull Request (PR) 🛠️. We are excited to see what you bring to the table!
|
||||
|
||||
Together, let's enhance the tracking capabilities of the Ultralytics YOLO ecosystem 🙏!
|
||||
|
|
|
|||
|
|
@ -198,7 +198,8 @@ def add_integration_callbacks(instance):
|
|||
"""
|
||||
|
||||
# Load HUB callbacks
|
||||
from .hub import callbacks
|
||||
from .hub import callbacks as hub_cb
|
||||
callbacks_list = [hub_cb]
|
||||
|
||||
# Load training callbacks
|
||||
if 'Trainer' in instance.__class__.__name__:
|
||||
|
|
@ -210,13 +211,15 @@ def add_integration_callbacks(instance):
|
|||
from .raytune import callbacks as tune_cb
|
||||
from .tensorboard import callbacks as tb_cb
|
||||
from .wb import callbacks as wb_cb
|
||||
callbacks.update({**clear_cb, **comet_cb, **dvc_cb, **mlflow_cb, **neptune_cb, **tune_cb, **tb_cb, **wb_cb})
|
||||
callbacks_list.extend([clear_cb, comet_cb, dvc_cb, mlflow_cb, neptune_cb, tune_cb, tb_cb, wb_cb])
|
||||
|
||||
# Load export callbacks (patch to avoid CoreML protobuf error)
|
||||
if 'Exporter' in instance.__class__.__name__:
|
||||
from .tensorboard import callbacks as tb_cb
|
||||
callbacks.update(tb_cb)
|
||||
callbacks_list.append(tb_cb)
|
||||
|
||||
for k, v in callbacks.items():
|
||||
if v not in instance.callbacks[k]: # prevent duplicate callbacks addition
|
||||
instance.callbacks[k].append(v) # callback[name].append(func)
|
||||
# Add the callbacks to the callbacks dictionary
|
||||
for callbacks in callbacks_list:
|
||||
for k, v in callbacks.items():
|
||||
if v not in instance.callbacks[k]:
|
||||
instance.callbacks[k].append(v)
|
||||
|
|
|
|||
|
|
@ -189,7 +189,7 @@ class ConfusionMatrix:
|
|||
self.task = task
|
||||
self.matrix = np.zeros((nc + 1, nc + 1)) if self.task == 'detect' else np.zeros((nc, nc))
|
||||
self.nc = nc # number of classes
|
||||
self.conf = conf
|
||||
self.conf = 0.25 if conf is None else conf # argument may be None from default cfg
|
||||
self.iou_thres = iou_thres
|
||||
|
||||
def process_cls_preds(self, preds, targets):
|
||||
|
|
|
|||
|
|
@ -12,11 +12,10 @@ import torch
|
|||
from PIL import Image, ImageDraw, ImageFont
|
||||
from PIL import __version__ as pil_version
|
||||
|
||||
from ultralytics.utils import LOGGER, TryExcept, plt_settings, threaded
|
||||
from ultralytics.utils import LOGGER, TryExcept, ops, plt_settings, threaded
|
||||
|
||||
from .checks import check_font, check_version, is_ascii
|
||||
from .files import increment_path
|
||||
from .ops import clip_boxes, scale_image, xywh2xyxy, xyxy2xywh
|
||||
|
||||
|
||||
class Colors:
|
||||
|
|
@ -163,7 +162,7 @@ class Annotator:
|
|||
im_gpu = im_gpu * inv_alph_masks[-1] + mcs
|
||||
im_mask = (im_gpu * 255)
|
||||
im_mask_np = im_mask.byte().cpu().numpy()
|
||||
self.im[:] = im_mask_np if retina_masks else scale_image(im_mask_np, self.im.shape)
|
||||
self.im[:] = im_mask_np if retina_masks else ops.scale_image(im_mask_np, self.im.shape)
|
||||
if self.pil:
|
||||
# Convert im back to PIL and update draw
|
||||
self.fromarray(self.im)
|
||||
|
|
@ -268,8 +267,9 @@ def plot_labels(boxes, cls, names=(), save_dir=Path(''), on_plot=None):
|
|||
import pandas as pd
|
||||
import seaborn as sn
|
||||
|
||||
# Filter matplotlib>=3.7.2 warning
|
||||
# Filter matplotlib>=3.7.2 warning and Seaborn use_inf and is_categorical FutureWarnings
|
||||
warnings.filterwarnings('ignore', category=UserWarning, message='The figure layout has changed to tight')
|
||||
warnings.filterwarnings('ignore', category=FutureWarning)
|
||||
|
||||
# Plot dataset labels
|
||||
LOGGER.info(f"Plotting labels to {save_dir / 'labels.jpg'}... ")
|
||||
|
|
@ -285,8 +285,8 @@ def plot_labels(boxes, cls, names=(), save_dir=Path(''), on_plot=None):
|
|||
# Matplotlib labels
|
||||
ax = plt.subplots(2, 2, figsize=(8, 8), tight_layout=True)[1].ravel()
|
||||
y = ax[0].hist(cls, bins=np.linspace(0, nc, nc + 1) - 0.5, rwidth=0.8)
|
||||
with contextlib.suppress(Exception): # color histogram bars by class
|
||||
[y[2].patches[i].set_color([x / 255 for x in colors(i)]) for i in range(nc)] # known issue #3195
|
||||
for i in range(nc):
|
||||
y[2].patches[i].set_color([x / 255 for x in colors(i)])
|
||||
ax[0].set_ylabel('instances')
|
||||
if 0 < len(names) < 30:
|
||||
ax[0].set_xticks(range(len(names)))
|
||||
|
|
@ -298,7 +298,7 @@ def plot_labels(boxes, cls, names=(), save_dir=Path(''), on_plot=None):
|
|||
|
||||
# Rectangles
|
||||
boxes[:, 0:2] = 0.5 # center
|
||||
boxes = xywh2xyxy(boxes) * 1000
|
||||
boxes = ops.xywh2xyxy(boxes) * 1000
|
||||
img = Image.fromarray(np.ones((1000, 1000, 3), dtype=np.uint8) * 255)
|
||||
for cls, box in zip(cls[:500], boxes[:500]):
|
||||
ImageDraw.Draw(img).rectangle(box, width=1, outline=colors(cls)) # plot
|
||||
|
|
@ -348,12 +348,12 @@ def save_one_box(xyxy, im, file=Path('im.jpg'), gain=1.02, pad=10, square=False,
|
|||
|
||||
if not isinstance(xyxy, torch.Tensor): # may be list
|
||||
xyxy = torch.stack(xyxy)
|
||||
b = xyxy2xywh(xyxy.view(-1, 4)) # boxes
|
||||
b = ops.xyxy2xywh(xyxy.view(-1, 4)) # boxes
|
||||
if square:
|
||||
b[:, 2:] = b[:, 2:].max(1)[0].unsqueeze(1) # attempt rectangle to square
|
||||
b[:, 2:] = b[:, 2:] * gain + pad # box wh * gain + pad
|
||||
xyxy = xywh2xyxy(b).long()
|
||||
clip_boxes(xyxy, im.shape)
|
||||
xyxy = ops.xywh2xyxy(b).long()
|
||||
ops.clip_boxes(xyxy, im.shape)
|
||||
crop = im[int(xyxy[0, 1]):int(xyxy[0, 3]), int(xyxy[0, 0]):int(xyxy[0, 2]), ::(1 if BGR else -1)]
|
||||
if save:
|
||||
file.parent.mkdir(parents=True, exist_ok=True) # make directory
|
||||
|
|
@ -425,7 +425,7 @@ def plot_images(images,
|
|||
classes = cls[idx].astype('int')
|
||||
|
||||
if len(bboxes):
|
||||
boxes = xywh2xyxy(bboxes[idx, :4]).T
|
||||
boxes = ops.xywh2xyxy(bboxes[idx, :4]).T
|
||||
labels = bboxes.shape[1] == 4 # labels if no conf column
|
||||
conf = None if labels else bboxes[idx, 4] # check for confidence presence (label vs pred)
|
||||
|
||||
|
|
@ -554,7 +554,7 @@ def output_to_target(output, max_det=300):
|
|||
for i, o in enumerate(output):
|
||||
box, conf, cls = o[:max_det, :6].cpu().split((4, 1, 1), 1)
|
||||
j = torch.full((conf.shape[0], 1), i)
|
||||
targets.append(torch.cat((j, cls, xyxy2xywh(box), conf), 1))
|
||||
targets.append(torch.cat((j, cls, ops.xyxy2xywh(box), conf), 1))
|
||||
targets = torch.cat(targets, 0).numpy()
|
||||
return targets[:, 0], targets[:, 1], targets[:, 2:]
|
||||
|
||||
|
|
|
|||
|
|
@ -14,7 +14,7 @@ def select_candidates_in_gts(xy_centers, gt_bboxes, eps=1e-9):
|
|||
Select the positive anchor center in gt.
|
||||
|
||||
Args:
|
||||
xy_centers (Tensor): shape(h*w, 4)
|
||||
xy_centers (Tensor): shape(h*w, 2)
|
||||
gt_bboxes (Tensor): shape(b, n_boxes, 4)
|
||||
|
||||
Returns:
|
||||
|
|
@ -228,7 +228,7 @@ class TaskAlignedAssigner(nn.Module):
|
|||
target_gt_idx = target_gt_idx + batch_ind * self.n_max_boxes # (b, h*w)
|
||||
target_labels = gt_labels.long().flatten()[target_gt_idx] # (b, h*w)
|
||||
|
||||
# Assigned target boxes, (b, max_num_obj, 4) -> (b, h*w)
|
||||
# Assigned target boxes, (b, max_num_obj, 4) -> (b, h*w, 4)
|
||||
target_bboxes = gt_bboxes.view(-1, 4)[target_gt_idx]
|
||||
|
||||
# Assigned target scores
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue