Skip to content

Group tensorboard metrics #39

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

Merged
VictorSanh merged 12 commits into from
Aug 5, 2021
Merged

Group tensorboard metrics #39

Show file tree
Hide file tree
Changes from 6 commits
Commits
Show all changes
12 commits
Select commit Hold shift + click to select a range
bbafd50
Training groupings
VictorSanh Aug 4, 2021
c68cad4
validation grouping
VictorSanh Aug 4, 2021
0447b63
steps vs samples
VictorSanh Aug 4, 2021
f14e772
iteration time (speed -> samples or iterations per second)
VictorSanh Aug 4, 2021
cfb02c7
tensorboard group time (from `log_timers_to_tensorboard`)
VictorSanh Aug 4, 2021
9710112
comment on the writing condition
VictorSanh Aug 4, 2021
f828e79
Update megatron/global_vars.py
VictorSanh Aug 4, 2021
dda7458
Update megatron/training.py
VictorSanh Aug 4, 2021
73b7c61
Update megatron/training.py
VictorSanh Aug 4, 2021
c56c704
Update megatron/training.py
VictorSanh Aug 4, 2021
1dcf28a
Update megatron/training.py
VictorSanh Aug 4, 2021
7a43335
link bug fix issue on megatron-lm side
VictorSanh Aug 5, 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
2 changes: 1 addition & 1 deletion megatron/global_vars.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -244,7 +244,7 @@ def write(self, names, writer, iteration, normalizer=1.0, reset=False):
assert normalizer > 0.0
for name in names:
value = self.timers[name].elapsed(reset=reset) / normalizer
writer.add_scalar(name + '-time', value, iteration)
writer.add_scalar('time/' + name + '-time', value, iteration)
Comment thread
VictorSanh marked this conversation as resolved.
Outdated

def log(self, names, normalizer=1.0, reset=True):
"""Log a group of timers."""
Expand Down
48 changes: 26 additions & 22 deletions megatron/training.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -547,33 +547,35 @@ def add_to_logging(name):
# Tensorboard values.
if writer and (iteration % args.tensorboard_log_interval == 0 ) and \
is_last_rank():
writer.add_scalar('steps-vs-samples/y=steps,x=samples', iteration, args.consumed_train_samples)
writer.add_scalar('steps-vs-samples/y=samples,x=steps', args.consumed_train_samples, iteration)
if args.log_learning_rate_to_tensorboard:
writer.add_scalar('learning-rate', learning_rate, iteration)
writer.add_scalar('learning-rate vs samples', learning_rate,
writer.add_scalar('learning-rate/learning-rate', learning_rate, iteration)
writer.add_scalar('learning-rate/learning-rate vs samples', learning_rate,
args.consumed_train_samples)
if args.log_batch_size_to_tensorboard:
writer.add_scalar('batch-size', batch_size, iteration)
writer.add_scalar('batch-size vs samples', batch_size,
writer.add_scalar('batch-size/batch-size', batch_size, iteration)
writer.add_scalar('batch-size/batch-size vs samples', batch_size,
args.consumed_train_samples)
for key in loss_dict:
writer.add_scalar(key , loss_dict[key], iteration)
writer.add_scalar(key + ' vs samples', loss_dict[key],
writer.add_scalar(f"lm-loss-training/{key}", loss_dict[key], iteration)
writer.add_scalar(f"lm-loss-training/{key}" + ' vs samples', loss_dict[key],
args.consumed_train_samples)
if args.log_loss_scale_to_tensorboard:
writer.add_scalar('loss-scale', loss_scale, iteration)
writer.add_scalar('loss-scale vs samples', loss_scale,
writer.add_scalar('loss-scale/loss-scale', loss_scale, iteration)
writer.add_scalar('loss-scale/loss-scale vs samples', loss_scale,
args.consumed_train_samples)
if grad_norm is not None:
writer.add_scalar('grad-norm', grad_norm, iteration)
writer.add_scalar('grad-norm vs samples', grad_norm,
writer.add_scalar('grad-norm/grad-norm', grad_norm, iteration)
writer.add_scalar('grad-norm/grad-norm vs samples', grad_norm,
args.consumed_train_samples)
if num_zeros_in_grad is not None:
writer.add_scalar('num-zeros', num_zeros_in_grad, iteration)
writer.add_scalar('num-zeros vs samples', num_zeros_in_grad,
writer.add_scalar('num-zeros/num-zeros', num_zeros_in_grad, iteration)
writer.add_scalar('num-zeros/num-zeros vs samples', num_zeros_in_grad,
args.consumed_train_samples)
if params_norm is not None:
writer.add_scalar('params-norm', params_norm, iteration)
writer.add_scalar('params-norm vs samples', params_norm,
writer.add_scalar('params-norm/params-norm', params_norm, iteration)
writer.add_scalar('params-norm/params-norm vs samples', params_norm,
args.consumed_train_samples)
if args.log_timers_to_tensorboard:
timers.write(timers_to_log, writer, iteration,
Expand All @@ -582,10 +584,12 @@ def add_to_logging(name):
if iteration % args.log_interval == 0:
elapsed_time = timers('interval-time').elapsed()
elapsed_time_per_iteration = elapsed_time / total_iterations
if writer and torch.distributed.get_rank() == 0:
if writer and is_last_rank(): # Changed `torch.distributed.get_rank() == 0` to `is_last_rank()` as a writing condition
Comment thread
VictorSanh marked this conversation as resolved.
Outdated
if args.log_timers_to_tensorboard:
writer.add_scalar('iteration-time',
writer.add_scalar('iteration-time/iteration-time (s)',
elapsed_time_per_iteration, iteration)
writer.add_scalar('iteration-time/iteration-time (s) vs samples',
elapsed_time_per_iteration, args.consumed_train_samples)
log_string = ' iteration {:8d}/{:8d} |'.format(
iteration, args.train_iters)
log_string += ' consumed samples: {:12d} |'.format(
Expand Down Expand Up @@ -772,7 +776,7 @@ def evaluate(forward_step_func, data_iterator, model, verbose=False):
forward_backward_func = forward_backward_pipelining_without_interleaving
else:
forward_backward_func = forward_backward_no_pipelining

if args.deepspeed:
# DeepSpeed uses eval_batch() and already aggregates losses.
assert isinstance(model, list) and len(model) == 1
Expand All @@ -782,7 +786,7 @@ def evaluate(forward_step_func, data_iterator, model, verbose=False):
loss_dicts = forward_backward_func(
forward_step_func, data_iterator, model, optimizer=None,
timers=None, forward_only=True)

if mpu.is_pipeline_last_stage(ignore_virtual=True):
# Reduce across processes.
for loss_dict in loss_dicts:
Expand Down Expand Up @@ -816,16 +820,16 @@ def evaluate_and_print_results(prefix, forward_step_func,
ppl = math.exp(min(20, total_loss_dict[key].item()))
string += '{} PPL: {:.6E} | '.format(key, ppl)
if writer and is_last_rank():
writer.add_scalar('{} validation'.format(key),
writer.add_scalar('lm-loss-validation/{} validation'.format(key),
Comment thread
VictorSanh marked this conversation as resolved.
Outdated
total_loss_dict[key].item(),
iteration)
writer.add_scalar('{} validation vs samples'.format(key),
writer.add_scalar('lm-loss-validation/{} validation vs samples'.format(key),
Comment thread
VictorSanh marked this conversation as resolved.
Outdated
total_loss_dict[key].item(),
args.consumed_train_samples)
if args.log_validation_ppl_to_tensorboard:
writer.add_scalar('{} validation ppl'.format(key), ppl,
writer.add_scalar('lm-loss-validation/{} validation ppl'.format(key), ppl,
Comment thread
VictorSanh marked this conversation as resolved.
Outdated
iteration)
writer.add_scalar('{} validation ppl vs samples'.format(key),
writer.add_scalar('lm-loss-validation/{} validation ppl vs samples'.format(key),
Comment thread
VictorSanh marked this conversation as resolved.
Outdated
ppl, args.consumed_train_samples)

length = len(string) + 1
Expand Down