Move loss calculation to head (#2874)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Laughing-q <1185102784@qq.com> Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> Co-authored-by: Laughing <61612323+Laughing-q@users.noreply.github.com>
This commit is contained in:
Ayush Chaurasia
GitHub
@ -2,8 +2,6 @@
|
||||
from copy import copy
|
||||
|
||||
import numpy as np
|
||||
import torch
|
||||
import torch.nn as nn
|
||||
|
||||
from ultralytics.nn.tasks import DetectionModel
|
||||
from ultralytics.yolo import v8
|
||||
@ -11,10 +9,7 @@ from ultralytics.yolo.data import build_dataloader, build_yolo_dataset
|
||||
from ultralytics.yolo.data.dataloaders.v5loader import create_dataloader
|
||||
from ultralytics.yolo.engine.trainer import BaseTrainer
|
||||
from ultralytics.yolo.utils import DEFAULT_CFG, LOGGER, RANK, colorstr
|
||||
from ultralytics.yolo.utils.loss import BboxLoss
|
||||
from ultralytics.yolo.utils.ops import xywh2xyxy
|
||||
from ultralytics.yolo.utils.plotting import plot_images, plot_labels, plot_results
|
||||
from ultralytics.yolo.utils.tal import TaskAlignedAssigner, dist2bbox, make_anchors
|
||||
from ultralytics.yolo.utils.torch_utils import de_parallel, torch_distributed_zero_first
|
||||
|
||||
|
||||
@ -91,12 +86,6 @@ class DetectionTrainer(BaseTrainer):
|
||||
self.loss_names = 'box_loss', 'cls_loss', 'dfl_loss'
|
||||
return v8.detect.DetectionValidator(self.test_loader, save_dir=self.save_dir, args=copy(self.args))
|
||||
|
||||
def criterion(self, preds, batch):
|
||||
"""Compute loss for YOLO prediction and ground-truth."""
|
||||
if not hasattr(self, 'compute_loss'):
|
||||
self.compute_loss = Loss(de_parallel(self.model))
|
||||
return self.compute_loss(preds, batch)
|
||||
|
||||
def label_loss_items(self, loss_items=None, prefix='train'):
|
||||
"""
|
||||
Returns a loss dict with labelled training loss items tensor
|
||||
@ -135,102 +124,6 @@ class DetectionTrainer(BaseTrainer):
|
||||
plot_labels(boxes, cls.squeeze(), names=self.data['names'], save_dir=self.save_dir, on_plot=self.on_plot)
|
||||
|
||||
|
||||
# Criterion class for computing training losses
|
||||
class Loss:
|
||||
|
||||
def __init__(self, model): # model must be de-paralleled
|
||||
|
||||
device = next(model.parameters()).device # get model device
|
||||
h = model.args # hyperparameters
|
||||
|
||||
m = model.model[-1] # Detect() module
|
||||
self.bce = nn.BCEWithLogitsLoss(reduction='none')
|
||||
self.hyp = h
|
||||
self.stride = m.stride # model strides
|
||||
self.nc = m.nc # number of classes
|
||||
self.no = m.no
|
||||
self.reg_max = m.reg_max
|
||||
self.device = device
|
||||
|
||||
self.use_dfl = m.reg_max > 1
|
||||
|
||||
self.assigner = TaskAlignedAssigner(topk=10, num_classes=self.nc, alpha=0.5, beta=6.0)
|
||||
self.bbox_loss = BboxLoss(m.reg_max - 1, use_dfl=self.use_dfl).to(device)
|
||||
self.proj = torch.arange(m.reg_max, dtype=torch.float, device=device)
|
||||
|
||||
def preprocess(self, targets, batch_size, scale_tensor):
|
||||
"""Preprocesses the target counts and matches with the input batch size to output a tensor."""
|
||||
if targets.shape[0] == 0:
|
||||
out = torch.zeros(batch_size, 0, 5, device=self.device)
|
||||
else:
|
||||
i = targets[:, 0] # image index
|
||||
_, counts = i.unique(return_counts=True)
|
||||
counts = counts.to(dtype=torch.int32)
|
||||
out = torch.zeros(batch_size, counts.max(), 5, device=self.device)
|
||||
for j in range(batch_size):
|
||||
matches = i == j
|
||||
n = matches.sum()
|
||||
if n:
|
||||
out[j, :n] = targets[matches, 1:]
|
||||
out[..., 1:5] = xywh2xyxy(out[..., 1:5].mul_(scale_tensor))
|
||||
return out
|
||||
|
||||
def bbox_decode(self, anchor_points, pred_dist):
|
||||
"""Decode predicted object bounding box coordinates from anchor points and distribution."""
|
||||
if self.use_dfl:
|
||||
b, a, c = pred_dist.shape # batch, anchors, channels
|
||||
pred_dist = pred_dist.view(b, a, 4, c // 4).softmax(3).matmul(self.proj.type(pred_dist.dtype))
|
||||
# pred_dist = pred_dist.view(b, a, c // 4, 4).transpose(2,3).softmax(3).matmul(self.proj.type(pred_dist.dtype))
|
||||
# pred_dist = (pred_dist.view(b, a, c // 4, 4).softmax(2) * self.proj.type(pred_dist.dtype).view(1, 1, -1, 1)).sum(2)
|
||||
return dist2bbox(pred_dist, anchor_points, xywh=False)
|
||||
|
||||
def __call__(self, preds, batch):
|
||||
"""Calculate the sum of the loss for box, cls and dfl multiplied by batch size."""
|
||||
loss = torch.zeros(3, device=self.device) # box, cls, dfl
|
||||
feats = preds[1] if isinstance(preds, tuple) else preds
|
||||
pred_distri, pred_scores = torch.cat([xi.view(feats[0].shape[0], self.no, -1) for xi in feats], 2).split(
|
||||
(self.reg_max * 4, self.nc), 1)
|
||||
|
||||
pred_scores = pred_scores.permute(0, 2, 1).contiguous()
|
||||
pred_distri = pred_distri.permute(0, 2, 1).contiguous()
|
||||
|
||||
dtype = pred_scores.dtype
|
||||
batch_size = pred_scores.shape[0]
|
||||
imgsz = torch.tensor(feats[0].shape[2:], device=self.device, dtype=dtype) * self.stride[0] # image size (h,w)
|
||||
anchor_points, stride_tensor = make_anchors(feats, self.stride, 0.5)
|
||||
|
||||
# targets
|
||||
targets = torch.cat((batch['batch_idx'].view(-1, 1), batch['cls'].view(-1, 1), batch['bboxes']), 1)
|
||||
targets = self.preprocess(targets.to(self.device), batch_size, scale_tensor=imgsz[[1, 0, 1, 0]])
|
||||
gt_labels, gt_bboxes = targets.split((1, 4), 2) # cls, xyxy
|
||||
mask_gt = gt_bboxes.sum(2, keepdim=True).gt_(0)
|
||||
|
||||
# pboxes
|
||||
pred_bboxes = self.bbox_decode(anchor_points, pred_distri) # xyxy, (b, h*w, 4)
|
||||
|
||||
_, target_bboxes, target_scores, fg_mask, _ = self.assigner(
|
||||
pred_scores.detach().sigmoid(), (pred_bboxes.detach() * stride_tensor).type(gt_bboxes.dtype),
|
||||
anchor_points * stride_tensor, gt_labels, gt_bboxes, mask_gt)
|
||||
|
||||
target_scores_sum = max(target_scores.sum(), 1)
|
||||
|
||||
# cls loss
|
||||
# loss[1] = self.varifocal_loss(pred_scores, target_scores, target_labels) / target_scores_sum # VFL way
|
||||
loss[1] = self.bce(pred_scores, target_scores.to(dtype)).sum() / target_scores_sum # BCE
|
||||
|
||||
# bbox loss
|
||||
if fg_mask.sum():
|
||||
target_bboxes /= stride_tensor
|
||||
loss[0], loss[2] = self.bbox_loss(pred_distri, pred_bboxes, anchor_points, target_bboxes, target_scores,
|
||||
target_scores_sum, fg_mask)
|
||||
|
||||
loss[0] *= self.hyp.box # box gain
|
||||
loss[1] *= self.hyp.cls # cls gain
|
||||
loss[2] *= self.hyp.dfl # dfl gain
|
||||
|
||||
return loss.sum() * batch_size, loss.detach() # loss(box, cls, dfl)
|
||||
|
||||
|
||||
def train(cfg=DEFAULT_CFG, use_python=False):
|
||||
"""Train and optimize YOLO model given training data and device."""
|
||||
model = cfg.model or 'yolov8n.pt'
|
||||
|
||||
Reference in New Issue
Block a user