apoorva/greenhouse
2
1
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

YOLOv5 v5.0 release compatibility update for YOLOv3 (#1737)

Browse Source 
* YOLOv5 v5.0 release compatibility update

* Update README

* Update README

* Conv act LeakyReLU(0.1)

* update plots_study()

* update speeds
This commit is contained in:
Glenn Jocher
2021-04-12 18:00:47 +02:00
committed by GitHub
parent 5d8f03020c
commit 8eb4cde090
37 changed files with 1791 additions and 908 deletions
Download Patch File Download Diff File Expand all files Collapse all files
models/common.py
+123 -40
View File
@@ -1,16 +1,21 @@
# This file contains modules common to various models
# YOLOv3 common modules
import math
from copy import copy
from pathlib import Path
import numpy as np
import pandas as pd
import requests
import torch
import torch.nn as nn
from PIL import Image, ImageDraw
from PIL import Image
from torch.cuda import amp
from utils.datasets import letterbox
from utils.general import non_max_suppression, make_divisible, scale_coords, xyxy2xywh
from utils.plots import color_list
from utils.general import non_max_suppression, make_divisible, scale_coords, increment_path, xyxy2xywh
from utils.plots import color_list, plot_one_box
from utils.torch_utils import time_synchronized
def autopad(k, p=None): # kernel, padding
@@ -40,6 +45,52 @@ class Conv(nn.Module):
return self.act(self.conv(x))
class TransformerLayer(nn.Module):
# Transformer layer https://arxiv.org/abs/2010.11929 (LayerNorm layers removed for better performance)
def __init__(self, c, num_heads):
super().__init__()
self.q = nn.Linear(c, c, bias=False)
self.k = nn.Linear(c, c, bias=False)
self.v = nn.Linear(c, c, bias=False)
self.ma = nn.MultiheadAttention(embed_dim=c, num_heads=num_heads)
self.fc1 = nn.Linear(c, c, bias=False)
self.fc2 = nn.Linear(c, c, bias=False)
def forward(self, x):
x = self.ma(self.q(x), self.k(x), self.v(x))[0] + x
x = self.fc2(self.fc1(x)) + x
return x
class TransformerBlock(nn.Module):
# Vision Transformer https://arxiv.org/abs/2010.11929
def __init__(self, c1, c2, num_heads, num_layers):
super().__init__()
self.conv = None
if c1 != c2:
self.conv = Conv(c1, c2)
self.linear = nn.Linear(c2, c2) # learnable position embedding
self.tr = nn.Sequential(*[TransformerLayer(c2, num_heads) for _ in range(num_layers)])
self.c2 = c2
def forward(self, x):
if self.conv is not None:
x = self.conv(x)
b, _, w, h = x.shape
p = x.flatten(2)
p = p.unsqueeze(0)
p = p.transpose(0, 3)
p = p.squeeze(3)
e = self.linear(p)
x = p + e
x = self.tr(x)
x = x.unsqueeze(3)
x = x.transpose(0, 3)
x = x.reshape(b, self.c2, w, h)
return x
class Bottleneck(nn.Module):
# Standard bottleneck
def __init__(self, c1, c2, shortcut=True, g=1, e=0.5): # ch_in, ch_out, shortcut, groups, expansion
@@ -87,6 +138,14 @@ class C3(nn.Module):
return self.cv3(torch.cat((self.m(self.cv1(x)), self.cv2(x)), dim=1))
class C3TR(C3):
# C3 module with TransformerBlock()
def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5):
super().__init__(c1, c2, n, shortcut, g, e)
c_ = int(c2 * e)
self.m = TransformerBlock(c_, c_, 4, n)
class SPP(nn.Module):
# Spatial pyramid pooling layer used in YOLOv3-SPP
def __init__(self, c1, c2, k=(5, 9, 13)):
@@ -166,7 +225,6 @@ class NMS(nn.Module):
class autoShape(nn.Module):
# input-robust model wrapper for passing cv2/np/PIL/torch inputs. Includes preprocessing, inference and NMS
img_size = 640 # inference size (pixels)
conf = 0.25 # NMS confidence threshold
iou = 0.45 # NMS IoU threshold
classes = None # (optional list) filter by class
@@ -179,27 +237,33 @@ class autoShape(nn.Module):
print('autoShape already enabled, skipping... ') # model already converted to model.autoshape()
return self
@torch.no_grad()
def forward(self, imgs, size=640, augment=False, profile=False):
# Inference from various sources. For height=720, width=1280, RGB images example inputs are:
# Inference from various sources. For height=640, width=1280, RGB images example inputs are:
# filename: imgs = 'data/samples/zidane.jpg'
# URI: = 'https://github.com/ultralytics/yolov5/releases/download/v1.0/zidane.jpg'
# OpenCV: = cv2.imread('image.jpg')[:,:,::-1] # HWC BGR to RGB x(720,1280,3)
# PIL: = Image.open('image.jpg') # HWC x(720,1280,3)
# numpy: = np.zeros((720,1280,3)) # HWC
# torch: = torch.zeros(16,3,720,1280) # BCHW
# OpenCV: = cv2.imread('image.jpg')[:,:,::-1] # HWC BGR to RGB x(640,1280,3)
# PIL: = Image.open('image.jpg') # HWC x(640,1280,3)
# numpy: = np.zeros((640,1280,3)) # HWC
# torch: = torch.zeros(16,3,320,640) # BCHW (scaled to size=640, 0-1 values)
# multiple: = [Image.open('image1.jpg'), Image.open('image2.jpg'), ...] # list of images
t = [time_synchronized()]
p = next(self.model.parameters()) # for device and type
if isinstance(imgs, torch.Tensor): # torch
return self.model(imgs.to(p.device).type_as(p), augment, profile) # inference
with amp.autocast(enabled=p.device.type != 'cpu'):
return self.model(imgs.to(p.device).type_as(p), augment, profile) # inference
# Pre-process
n, imgs = (len(imgs), imgs) if isinstance(imgs, list) else (1, [imgs]) # number of images, list of images
shape0, shape1 = [], [] # image and inference shapes
shape0, shape1, files = [], [], [] # image and inference shapes, filenames
for i, im in enumerate(imgs):
f = f'image{i}' # filename
if isinstance(im, str): # filename or uri
im = Image.open(requests.get(im, stream=True).raw if im.startswith('http') else im) # open
im = np.array(im) # to numpy
im, f = np.asarray(Image.open(requests.get(im, stream=True).raw if im.startswith('http') else im)), im
elif isinstance(im, Image.Image): # PIL Image
im, f = np.asarray(im), getattr(im, 'filename', f) or f
files.append(Path(f).with_suffix('.jpg').name)
if im.shape[0] < 5: # image in CHW
im = im.transpose((1, 2, 0)) # reverse dataloader .transpose(2, 0, 1)
im = im[:, :, :3] if im.ndim == 3 else np.tile(im[:, :, None], 3) # enforce 3ch input
@@ -213,82 +277,101 @@ class autoShape(nn.Module):
x = np.stack(x, 0) if n > 1 else x[0][None] # stack
x = np.ascontiguousarray(x.transpose((0, 3, 1, 2))) # BHWC to BCHW
x = torch.from_numpy(x).to(p.device).type_as(p) / 255. # uint8 to fp16/32
t.append(time_synchronized())
# Inference
with torch.no_grad():
with amp.autocast(enabled=p.device.type != 'cpu'):
# Inference
y = self.model(x, augment, profile)[0] # forward
y = non_max_suppression(y, conf_thres=self.conf, iou_thres=self.iou, classes=self.classes) # NMS
t.append(time_synchronized())
# Post-process
for i in range(n):
scale_coords(shape1, y[i][:, :4], shape0[i])
# Post-process
y = non_max_suppression(y, conf_thres=self.conf, iou_thres=self.iou, classes=self.classes) # NMS
for i in range(n):
scale_coords(shape1, y[i][:, :4], shape0[i])
return Detections(imgs, y, self.names)
t.append(time_synchronized())
return Detections(imgs, y, files, t, self.names, x.shape)
class Detections:
# detections class for YOLOv5 inference results
def __init__(self, imgs, pred, names=None):
# detections class for YOLOv3 inference results
def __init__(self, imgs, pred, files, times=None, names=None, shape=None):
super(Detections, self).__init__()
d = pred[0].device # device
gn = [torch.tensor([*[im.shape[i] for i in [1, 0, 1, 0]], 1., 1.], device=d) for im in imgs] # normalizations
self.imgs = imgs # list of images as numpy arrays
self.pred = pred # list of tensors pred[0] = (xyxy, conf, cls)
self.names = names # class names
self.files = files # image filenames
self.xyxy = pred # xyxy pixels
self.xywh = [xyxy2xywh(x) for x in pred] # xywh pixels
self.xyxyn = [x / g for x, g in zip(self.xyxy, gn)] # xyxy normalized
self.xywhn = [x / g for x, g in zip(self.xywh, gn)] # xywh normalized
self.n = len(self.pred)
self.n = len(self.pred) # number of images (batch size)
self.t = tuple((times[i + 1] - times[i]) * 1000 / self.n for i in range(3)) # timestamps (ms)
self.s = shape # inference BCHW shape
def display(self, pprint=False, show=False, save=False, render=False):
def display(self, pprint=False, show=False, save=False, render=False, save_dir=''):
colors = color_list()
for i, (img, pred) in enumerate(zip(self.imgs, self.pred)):
str = f'Image {i + 1}/{len(self.pred)}: {img.shape[0]}x{img.shape[1]} '
str = f'image {i + 1}/{len(self.pred)}: {img.shape[0]}x{img.shape[1]} '
if pred is not None:
for c in pred[:, -1].unique():
n = (pred[:, -1] == c).sum() # detections per class
str += f'{n} {self.names[int(c)]}s, ' # add to string
str += f"{n} {self.names[int(c)]}{'s' * (n > 1)}, " # add to string
if show or save or render:
img = Image.fromarray(img.astype(np.uint8)) if isinstance(img, np.ndarray) else img # from np
for *box, conf, cls in pred: # xyxy, confidence, class
# str += '%s %.2f, ' % (names[int(cls)], conf) # label
ImageDraw.Draw(img).rectangle(box, width=4, outline=colors[int(cls) % 10]) # plot
label = f'{self.names[int(cls)]} {conf:.2f}'
plot_one_box(box, img, label=label, color=colors[int(cls) % 10])
img = Image.fromarray(img.astype(np.uint8)) if isinstance(img, np.ndarray) else img # from np
if pprint:
print(str)
print(str.rstrip(', '))
if show:
img.show(f'Image {i}') # show
img.show(self.files[i]) # show
if save:
f = f'results{i}.jpg'
str += f"saved to '{f}'"
img.save(f) # save
f = self.files[i]
img.save(Path(save_dir) / f) # save
print(f"{'Saved' * (i == 0)} {f}", end=',' if i < self.n - 1 else f' to {save_dir}\n')
if render:
self.imgs[i] = np.asarray(img)
def print(self):
self.display(pprint=True) # print results
print(f'Speed: %.1fms pre-process, %.1fms inference, %.1fms NMS per image at shape {tuple(self.s)}' % self.t)
def show(self):
self.display(show=True) # show results
def save(self):
self.display(save=True) # save results
def save(self, save_dir='runs/hub/exp'):
save_dir = increment_path(save_dir, exist_ok=save_dir != 'runs/hub/exp') # increment save_dir
Path(save_dir).mkdir(parents=True, exist_ok=True)
self.display(save=True, save_dir=save_dir) # save results
def render(self):
self.display(render=True) # render results
return self.imgs
def __len__(self):
return self.n
def pandas(self):
# return detections as pandas DataFrames, i.e. print(results.pandas().xyxy[0])
new = copy(self) # return copy
ca = 'xmin', 'ymin', 'xmax', 'ymax', 'confidence', 'class', 'name' # xyxy columns
cb = 'xcenter', 'ycenter', 'width', 'height', 'confidence', 'class', 'name' # xywh columns
for k, c in zip(['xyxy', 'xyxyn', 'xywh', 'xywhn'], [ca, ca, cb, cb]):
a = [[x[:5] + [int(x[5]), self.names[int(x[5])]] for x in x.tolist()] for x in getattr(self, k)] # update
setattr(new, k, [pd.DataFrame(x, columns=c) for x in a])
return new
def tolist(self):
# return a list of Detections objects, i.e. 'for result in results.tolist():'
x = [Detections([self.imgs[i]], [self.pred[i]], self.names) for i in range(self.n)]
x = [Detections([self.imgs[i]], [self.pred[i]], self.names, self.s) for i in range(self.n)]
for d in x:
for k in ['imgs', 'pred', 'xyxy', 'xyxyn', 'xywh', 'xywhn']:
setattr(d, k, getattr(d, k)[0]) # pop out of list
return x
def __len__(self):
return self.n
class Classify(nn.Module):
# Classification head, i.e. x(b,c1,20,20) to x(b,c2)
models/experimental.py
+5 -4
View File
@@ -1,4 +1,4 @@
# This file contains experimental modules
# YOLOv3 experimental modules
import numpy as np
import torch
@@ -58,7 +58,7 @@ class GhostConv(nn.Module):
class GhostBottleneck(nn.Module):
# Ghost Bottleneck https://github.com/huawei-noah/ghostnet
def __init__(self, c1, c2, k, s):
def __init__(self, c1, c2, k=3, s=1): # ch_in, ch_out, kernel, stride
super(GhostBottleneck, self).__init__()
c_ = c2 // 2
self.conv = nn.Sequential(GhostConv(c1, c_, 1, 1), # pw
@@ -115,11 +115,12 @@ def attempt_load(weights, map_location=None):
model = Ensemble()
for w in weights if isinstance(weights, list) else [weights]:
attempt_download(w)
model.append(torch.load(w, map_location=map_location)['model'].float().fuse().eval()) # load FP32 model
ckpt = torch.load(w, map_location=map_location) # load
model.append(ckpt['ema' if ckpt.get('ema') else 'model'].float().fuse().eval()) # FP32 model
# Compatibility updates
for m in model.modules():
if type(m) in [nn.Hardswish, nn.LeakyReLU, nn.ReLU, nn.ReLU6]:
if type(m) in [nn.Hardswish, nn.LeakyReLU, nn.ReLU, nn.ReLU6, nn.SiLU]:
m.inplace = True # pytorch 1.7.0 compatibility
elif type(m) is Conv:
m._non_persistent_buffers_set = set() # pytorch 1.6.0 compatibility
models/export.py
+13 -6
View File
@@ -1,4 +1,4 @@
"""Exports a YOLOv5 *.pt model to ONNX and TorchScript formats
"""Exports a YOLOv3 *.pt model to ONNX and TorchScript formats
Usage:
$ export PYTHONPATH="$PWD" && python models/export.py --weights ./weights/yolov3.pt --img 640 --batch 1
@@ -17,12 +17,16 @@ import models
from models.experimental import attempt_load
from utils.activations import Hardswish, SiLU
from utils.general import set_logging, check_img_size
from utils.torch_utils import select_device
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument('--weights', type=str, default='./yolov3.pt', help='weights path') # from yolov3/models/
parser.add_argument('--img-size', nargs='+', type=int, default=[640, 640], help='image size') # height, width
parser.add_argument('--batch-size', type=int, default=1, help='batch size')
parser.add_argument('--dynamic', action='store_true', help='dynamic ONNX axes')
parser.add_argument('--grid', action='store_true', help='export Detect() layer grid')
parser.add_argument('--device', default='cpu', help='cuda device, i.e. 0 or 0,1,2,3 or cpu')
opt = parser.parse_args()
opt.img_size *= 2 if len(opt.img_size) == 1 else 1 # expand
print(opt)
@@ -30,7 +34,8 @@ if __name__ == '__main__':
t = time.time()
# Load PyTorch model
model = attempt_load(opt.weights, map_location=torch.device('cpu')) # load FP32 model
device = select_device(opt.device)
model = attempt_load(opt.weights, map_location=device) # load FP32 model
labels = model.names
# Checks
@@ -38,7 +43,7 @@ if __name__ == '__main__':
opt.img_size = [check_img_size(x, gs) for x in opt.img_size] # verify img_size are gs-multiples
# Input
img = torch.zeros(opt.batch_size, 3, *opt.img_size) # image size(1,3,320,192) iDetection
img = torch.zeros(opt.batch_size, 3, *opt.img_size).to(device) # image size(1,3,320,192) iDetection
# Update model
for k, m in model.named_modules():
@@ -50,14 +55,14 @@ if __name__ == '__main__':
m.act = SiLU()
# elif isinstance(m, models.yolo.Detect):
# m.forward = m.forward_export # assign forward (optional)
model.model[-1].export = True # set Detect() layer export=True
model.model[-1].export = not opt.grid # set Detect() layer grid export
y = model(img) # dry run
# TorchScript export
try:
print('\nStarting TorchScript export with torch %s...' % torch.__version__)
f = opt.weights.replace('.pt', '.torchscript.pt') # filename
ts = torch.jit.trace(model, img)
ts = torch.jit.trace(model, img, strict=False)
ts.save(f)
print('TorchScript export success, saved as %s' % f)
except Exception as e:
@@ -70,7 +75,9 @@ if __name__ == '__main__':
print('\nStarting ONNX export with onnx %s...' % onnx.__version__)
f = opt.weights.replace('.pt', '.onnx') # filename
torch.onnx.export(model, img, f, verbose=False, opset_version=12, input_names=['images'],
output_names=['classes', 'boxes'] if y is None else ['output'])
output_names=['classes', 'boxes'] if y is None else ['output'],
dynamic_axes={'images': {0: 'batch', 2: 'height', 3: 'width'}, # size(1,3,640,640)
'output': {0: 'batch', 2: 'y', 3: 'x'}} if opt.dynamic else None)
# Checks
onnx_model = onnx.load(f) # load onnx model
models/yolo.py
+24 -33
View File
@@ -1,14 +1,15 @@
# YOLOv3 YOLO-specific modules
import argparse
import logging
import sys
from copy import deepcopy
from pathlib import Path
sys.path.append('./') # to run '$ python *.py' files in subdirectories
logger = logging.getLogger(__name__)
from models.common import *
from models.experimental import MixConv2d, CrossConv
from models.experimental import *
from utils.autoanchor import check_anchor_order
from utils.general import make_divisible, check_file, set_logging
from utils.torch_utils import time_synchronized, fuse_conv_and_bn, model_info, scale_img, initialize_weights, \
@@ -50,7 +51,7 @@ class Detect(nn.Module):
self.grid[i] = self._make_grid(nx, ny).to(x[i].device)
y = x[i].sigmoid()
y[..., 0:2] = (y[..., 0:2] * 2. - 0.5 + self.grid[i].to(x[i].device)) * self.stride[i] # xy
y[..., 0:2] = (y[..., 0:2] * 2. - 0.5 + self.grid[i]) * self.stride[i] # xy
y[..., 2:4] = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i] # wh
z.append(y.view(bs, -1, self.no))
@@ -63,7 +64,7 @@ class Detect(nn.Module):
class Model(nn.Module):
def __init__(self, cfg='yolov3.yaml', ch=3, nc=None): # model, input channels, number of classes
def __init__(self, cfg='yolov3.yaml', ch=3, nc=None, anchors=None): # model, input channels, number of classes
super(Model, self).__init__()
if isinstance(cfg, dict):
self.yaml = cfg # model dict
@@ -71,13 +72,16 @@ class Model(nn.Module):
import yaml # for torch hub
self.yaml_file = Path(cfg).name
with open(cfg) as f:
self.yaml = yaml.load(f, Loader=yaml.FullLoader) # model dict
self.yaml = yaml.load(f, Loader=yaml.SafeLoader) # model dict
# Define model
ch = self.yaml['ch'] = self.yaml.get('ch', ch) # input channels
if nc and nc != self.yaml['nc']:
logger.info('Overriding model.yaml nc=%g with nc=%g' % (self.yaml['nc'], nc))
logger.info(f"Overriding model.yaml nc={self.yaml['nc']} with nc={nc}")
self.yaml['nc'] = nc # override yaml value
if anchors:
logger.info(f'Overriding model.yaml anchors with anchors={anchors}')
self.yaml['anchors'] = round(anchors) # override yaml value
self.model, self.save = parse_model(deepcopy(self.yaml), ch=[ch]) # model, savelist
self.names = [str(i) for i in range(self.yaml['nc'])] # default names
# print([x.shape for x in self.forward(torch.zeros(1, ch, 64, 64))])
@@ -107,7 +111,7 @@ class Model(nn.Module):
for si, fi in zip(s, f):
xi = scale_img(x.flip(fi) if fi else x, si, gs=int(self.stride.max()))
yi = self.forward_once(xi)[0] # forward
# cv2.imwrite('img%g.jpg' % s, 255 * xi[0].numpy().transpose((1, 2, 0))[:, :, ::-1]) # save
# cv2.imwrite(f'img_{si}.jpg', 255 * xi[0].cpu().numpy().transpose((1, 2, 0))[:, :, ::-1]) # save
yi[..., :4] /= si # de-scale
if fi == 2:
yi[..., 1] = img_size[0] - yi[..., 1] # de-flip ud
@@ -210,45 +214,30 @@ def parse_model(d, ch): # model_dict, input_channels(3)
pass
n = max(round(n * gd), 1) if n > 1 else n # depth gain
if m in [Conv, Bottleneck, SPP, DWConv, MixConv2d, Focus, CrossConv, BottleneckCSP, C3]:
if m in [Conv, GhostConv, Bottleneck, GhostBottleneck, SPP, DWConv, MixConv2d, Focus, CrossConv, BottleneckCSP,
C3, C3TR]:
c1, c2 = ch[f], args[0]
# Normal
# if i > 0 and args[0] != no: # channel expansion factor
# ex = 1.75 # exponential (default 2.0)
# e = math.log(c2 / ch[1]) / math.log(2)
# c2 = int(ch[1] * ex ** e)
# if m != Focus:
c2 = make_divisible(c2 * gw, 8) if c2 != no else c2
# Experimental
# if i > 0 and args[0] != no: # channel expansion factor
# ex = 1 + gw # exponential (default 2.0)
# ch1 = 32 # ch[1]
# e = math.log(c2 / ch1) / math.log(2) # level 1-n
# c2 = int(ch1 * ex ** e)
# if m != Focus:
# c2 = make_divisible(c2, 8) if c2 != no else c2
if c2 != no: # if not output
c2 = make_divisible(c2 * gw, 8)
args = [c1, c2, *args[1:]]
if m in [BottleneckCSP, C3]:
args.insert(2, n)
if m in [BottleneckCSP, C3, C3TR]:
args.insert(2, n) # number of repeats
n = 1
elif m is nn.BatchNorm2d:
args = [ch[f]]
elif m is Concat:
c2 = sum([ch[x if x < 0 else x + 1] for x in f])
c2 = sum([ch[x] for x in f])
elif m is Detect:
args.append([ch[x + 1] for x in f])
args.append([ch[x] for x in f])
if isinstance(args[1], int): # number of anchors
args[1] = [list(range(args[1] * 2))] * len(f)
elif m is Contract:
c2 = ch[f if f < 0 else f + 1] * args[0] ** 2
c2 = ch[f] * args[0] ** 2
elif m is Expand:
c2 = ch[f if f < 0 else f + 1] // args[0] ** 2
c2 = ch[f] // args[0] ** 2
else:
c2 = ch[f if f < 0 else f + 1]
c2 = ch[f]
m_ = nn.Sequential(*[m(*args) for _ in range(n)]) if n > 1 else m(*args) # module
t = str(m)[8:-2].replace('__main__.', '') # module type
@@ -257,6 +246,8 @@ def parse_model(d, ch): # model_dict, input_channels(3)
logger.info('%3s%18s%3s%10.0f %-40s%-30s' % (i, f, n, np, t, args)) # print
save.extend(x % i for x in ([f] if isinstance(f, int) else f) if x != -1) # append to savelist
layers.append(m_)
if i == 0:
ch = []
ch.append(c2)
return nn.Sequential(*layers), sorted(save)