updates

train.py

@@ -100,19 +100,23 @@ def train(
  else:
  cutoff = load_darknet_weights(model, weights + 'darknet53.conv.74')
 
- # Scheduler (reduce lr at epochs 218, 245, i.e. batches 400k, 450k)
+ # Scheduler https://github.com/ultralytics/yolov3/issues/238
  # lf = lambda x: 1 - x / epochs # linear ramp to zero
- # lf = lambda x: 10 ** (-2 * x / epochs) # exp ramp to lr0 * 1e-2
- lf = lambda x: 1 - 10 ** (hyp['lrf'] * (1 - x / epochs)) # inv exp ramp to lr0 * 1e-2
+ # lf = lambda x: 10 ** (hyp['lrf'] * x / epochs) # exp ramp to lr0 * hyp['lrf']
+ lf = lambda x: 1 - 10 ** (hyp['lrf'] * (1 - x / epochs)) # inv exp ramp to lr0 * hyp['lrf']
  scheduler = optim.lr_scheduler.LambdaLR(optimizer, lr_lambda=lf, last_epoch=start_epoch - 1)
- # scheduler = optim.lr_scheduler.MultiStepLR(optimizer,milestones=[218, 245],gamma=0.1,last_epoch=start_epoch - 1)
+ # scheduler = optim.lr_scheduler.MultiStepLR(optimizer, milestones=[218, 245], gamma=0.1, last_epoch=start_epoch - 1)
 
- # Plot lr schedule
+ # # Plot lr schedule
  # y = []
  # for _ in range(epochs):
  # scheduler.step()
  # y.append(optimizer.param_groups[0]['lr'])
- # plt.plot(y)
+ # plt.plot(y, label='LambdaLR')
+ # plt.xlabel('epoch')
+ # plt.xlabel('LR')
+ # plt.tight_layout()
+ # plt.savefig('LR.png', dpi=300)
 
  # Dataset
  dataset = LoadImagesAndLabels(train_path, img_size=img_size, augment=True)
@@ -165,8 +169,6 @@ def train(
  imgs = imgs.to(device)
  targets = targets.to(device)
  nt = len(targets)
- # if nt == 0: # if no targets continue
- # continue
 
  # Plot images with bounding boxes
  if epoch == 0 and i == 0: