Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Fix warmup accumulate #3722

Merged
merged 6 commits into from
Jun 28, 2021
Merged

Fix warmup accumulate #3722

Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
a93e484
gradient accumulation during warmup in train.py
yellowdolphin Jun 22, 2021
ce33f84
replace last_opt_step tracking by do_step(ni)
yellowdolphin Jun 23, 2021
b7c489f
add docstrings
yellowdolphin Jun 23, 2021
dd5a574
move down nw
yellowdolphin Jun 24, 2021
53aae7f
Update train.py
glenn-jocher Jun 28, 2021
9ac476e
revert math import move
glenn-jocher Jun 28, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 3 additions & 1 deletion train.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -270,6 +270,7 @@ def train(hyp, # path/to/hyp.yaml or hyp dictionary
t0 = time.time()
nw = max(round(hyp['warmup_epochs'] * nb), 1000) # number of warmup iterations, max(3 epochs, 1k iterations)
# nw = min(nw, (epochs - start_epoch) / 2 * nb) # limit warmup to < 1/2 of training
last_opt_step = -1
maps = np.zeros(nc) # mAP per class
results = (0, 0, 0, 0, 0, 0, 0) # P, R, mAP@.5, mAP@.5-.95, val_loss(box, obj, cls)
scheduler.last_epoch = start_epoch - 1 # do not move
Expand Down Expand Up @@ -344,12 +345,13 @@ def train(hyp, # path/to/hyp.yaml or hyp dictionary
scaler.scale(loss).backward()

# Optimize
if ni % accumulate == 0:
if ni - last_opt_step >= accumulate:
scaler.step(optimizer) # optimizer.step
scaler.update()
optimizer.zero_grad()
if ema:
ema.update(model)
last_opt_step = ni

# Print
if RANK in [-1, 0]:
Expand Down