mirrors
/
yolov5
mirror of https://github.com/ultralytics/yolov5.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

refactored the code

pull/13483/head
Bala-Vignesh-Reddy 2025-03-22 23:25:11 +05:30
parent 42486238a9
commit 6695bac0cd
5 changed files with 14 additions and 19 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
classify/train.py
View File

@ -23,6 +23,7 @@ from datetime import datetime
from pathlib import Path
import torch
CHECK_PYTORCH_18 = torch.__version__.startswith("1.8")
import torch.distributed as dist
import torch.hub as hub
import torch.optim.lr_scheduler as lr_scheduler
@ -200,8 +201,7 @@ def train(opt, device):
criterion = smartCrossEntropyLoss(label_smoothing=opt.label_smoothing) # loss function
best_fitness = 0.0
# adding a check to torch version
scaler = None
if torch.__version__.startswith("1.8"):
if CHECK_PYTORCH_18:
scaler = torch.cuda.amp.GradScaler(enabled=cuda)
else:
scaler = torch.amp.GradScaler("cuda", enabled=cuda)
@ -225,8 +225,7 @@ def train(opt, device):
images, labels = images.to(device, non_blocking=True), labels.to(device)
# Forward
amp_autocast = None
if torch.__version__.startswith("1.8"):
if CHECK_PYTORCH_18:
amp_autocast = torch.cuda.amp.autocast(enabled=device.type != "cpu")
else:
amp_autocast = torch.amp.autocast("cuda", enabled=device.type != "cpu")

4
classify/val.py
View File

@ -26,6 +26,7 @@ import sys
from pathlib import Path
import torch
CHECK_PYTORCH_18 = torch.__version__.startswith("1.8")
from tqdm import tqdm
FILE = Path(__file__).resolve()
@ -110,8 +111,7 @@ def run(
bar = tqdm(dataloader, desc, n, not training, bar_format=TQDM_BAR_FORMAT, position=0)
# checking the version
amp_autocast = None
if torch.__version__.startswith("1.8"):
if CHECK_PYTORCH_18:
amp_autocast = torch.cuda.amp.autocast(enabled=device.type != "cpu")
else:
amp_autocast = torch.amp.autocast("cuda", enabled=device.type != "cpu")

8
models/common.py
View File

@ -20,7 +20,7 @@ import requests
import torch
import torch.nn as nn
from PIL import Image
CHECK_PYTORCH_18 = torch.__version__.startswith("1.8")
# Import 'ultralytics' package or install if missing
try:
import ultralytics
@ -863,8 +863,7 @@ class AutoShape(nn.Module):
p = next(self.model.parameters()) if self.pt else torch.empty(1, device=self.model.device) # param
autocast = self.amp and (p.device.type != "cpu") # Automatic Mixed Precision (AMP) inference
if isinstance(ims, torch.Tensor): # torch
amp_autocast = None
if torch.__version__.startswith("1.8"):
if CHECK_PYTORCH_18:
amp_autocast = torch.cuda.amp.autocast(enabled=autocast)
else:
amp_autocast = torch.amp.autocast("cuda", enabled=autocast)
@ -895,8 +894,7 @@ class AutoShape(nn.Module):
x = np.ascontiguousarray(np.array(x).transpose((0, 3, 1, 2))) # stack and BHWC to BCHW
x = torch.from_numpy(x).to(p.device).type_as(p) / 255 # uint8 to fp16/32
amp_autocast = None
if torch.__version__.startswith("1.8"):
if CHECK_PYTORCH_18:
amp_autocast = torch.cuda.amp.autocast(enabled=autocast)
else:
amp_autocast = torch.amp.autocast("cuda", enabled=autocast)

7
segment/train.py
View File

@ -39,6 +39,7 @@ ROOT = FILE.parents[1] # YOLOv5 root directory
if str(ROOT) not in sys.path:
sys.path.append(str(ROOT)) # add ROOT to PATH
ROOT = Path(os.path.relpath(ROOT, Path.cwd())) # relative
CHECK_PYTORCH_18 = torch.__version__.startswith("1.8")
import segment.val as validate # for end-of-epoch mAP
from models.experimental import attempt_load
@ -321,8 +322,7 @@ def train(hyp, opt, device, callbacks):
maps = np.zeros(nc) # mAP per class
results = (0, 0, 0, 0, 0, 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
scaler = None
if torch.__version__.startswith("1.8"):
if CHECK_PYTORCH_18:
scaler = torch.cuda.amp.GradScaler(enabled=amp)
else:
scaler = torch.amp.GradScaler("cuda", enabled=amp)
@ -383,8 +383,7 @@ def train(hyp, opt, device, callbacks):
if sf != 1:
ns = [math.ceil(x * sf / gs) * gs for x in imgs.shape[2:]] # new shape (stretched to gs-multiple)
imgs = nn.functional.interpolate(imgs, size=ns, mode="bilinear", align_corners=False)
amp_autocast = None
if torch.__version__.startswith("1.8"):
if CHECK_PYTORCH_18:
amp_autocast = torch.cuda.amp.autocast(enabled=amp)
else:
amp_autocast = torch.amp.autocast("cuda", enabled=amp)

7
train.py
View File

@ -43,6 +43,7 @@ ROOT = FILE.parents[0] # YOLOv5 root directory
if str(ROOT) not in sys.path:
sys.path.append(str(ROOT)) # add ROOT to PATH
ROOT = Path(os.path.relpath(ROOT, Path.cwd())) # relative
CHECK_PYTORCH_18 = torch.__version__.startswith("1.8")
import val as validate # for end-of-epoch mAP
from models.experimental import attempt_load
@ -353,8 +354,7 @@ def train(hyp, opt, device, callbacks):
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
scaler = None
if torch.__version__.startswith("1.8"):
if CHECK_PYTORCH_18:
scaler = torch.cuda.amp.GradScaler(enabled=amp)
else:
scaler = torch.amp.GradScaler("cuda", enabled=amp)
@ -415,8 +415,7 @@ def train(hyp, opt, device, callbacks):
# Forward
# with Autocast:
amp_autocast = None
if torch.__version__.startswith("1.8"):
if CHECK_PYTORCH_18:
amp_autocast = torch.cuda.amp.autocast(enabled=amp)
else:
amp_autocast = torch.amp.autocast("cuda", enabled=amp)