I can do training, but somehow the validation inside the training not work

[saikanov]

i got his from csv file in the trainig result

and usually there is a delay between epoch to validation, so validation doesnt occur in my case

could someone please help me? i cant tell if my model good enough or not.

[saikanov]

Update: the validation is running, but the results.csv val cls etc still zero

is this a normal thing?

[JulioZhao97]

@saikanov Hello! This is set by design, you can change code here (lines I comment out)

DocLayout-YOLO/doclayout_yolo/engine/validator.py

Lines 172 to 208 in 32a8ec2

	# HACK: don't record loss!!!
	self.run_callbacks("on_val_start")
	dt = (
	Profile(device=self.device),
	# Profile(device=self.device),
	Profile(device=self.device),
	Profile(device=self.device),
	)
	bar = TQDM(self.dataloader, desc=self.get_desc(), total=len(self.dataloader))
	self.init_metrics(de_parallel(model))
	self.jdict = [] # empty before each val
	for batch_i, batch in enumerate(bar):
	self.run_callbacks("on_val_batch_start")
	self.batch_i = batch_i
	# Preprocess
	with dt[0]:
	batch = self.preprocess(batch)
	# Inference
	with dt[1]:
	preds = model(batch["img"], augment=augment)
	# Loss
	# with dt[2]:
	# if self.training:
	# self.loss += model.loss(batch, preds)[1]
	# Postprocess
	with dt[2]:
	preds = self.postprocess(preds, conf=self.args.conf)
	self.update_metrics(preds, batch)
	if self.args.plots and batch_i < 3:
	self.plot_val_samples(batch, batch_i)
	self.plot_predictions(batch, preds, batch_i)
	self.run_callbacks("on_val_batch_end")

[saikanov]

@saikanov Hello! This is set by design, you can change code here (lines I comment out)

DocLayout-YOLO/doclayout_yolo/engine/validator.py

Lines 172 to 208 in 32a8ec2

HACK: don't record loss!!!

self.run_callbacks("on_val_start")
dt = (
Profile(device=self.device),
# Profile(device=self.device),
Profile(device=self.device),
Profile(device=self.device),
)
bar = TQDM(self.dataloader, desc=self.get_desc(), total=len(self.dataloader))
self.init_metrics(de_parallel(model))
self.jdict = [] # empty before each val
for batch_i, batch in enumerate(bar):
self.run_callbacks("on_val_batch_start")
self.batch_i = batch_i
# Preprocess
with dt[0]:
batch = self.preprocess(batch)

 # Inference 
 with dt[1]: 
     preds = model(batch["img"], augment=augment) 

 # Loss 
 # with dt[2]: 
 #     if self.training: 
 #         self.loss += model.loss(batch, preds)[1] 
  
 # Postprocess 
 with dt[2]: 
     preds = self.postprocess(preds, conf=self.args.conf) 

 self.update_metrics(preds, batch) 
 if self.args.plots and batch_i < 3: 
     self.plot_val_samples(batch, batch_i) 
     self.plot_predictions(batch, preds, batch_i) 

 self.run_callbacks("on_val_batch_end")

alright thanks i will try it