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

yolo with runs.zip file exp 14 is the best weight

Browse Source
This commit is contained in:
Apoorva Gupta
2023-02-21 21:43:51 +05:30
parent 34abb2b0dd
commit dbe80aca78
43 changed files with 3670 additions and 5669 deletions
Download Patch File Download Diff File Expand all files Collapse all files
yolov3/models/common.py
+221 -495
View File
@@ -3,17 +3,12 @@
Common modules
"""
import ast
import contextlib
import json
import math
import platform
import warnings
import zipfile
from collections import OrderedDict, namedtuple
from copy import copy
from pathlib import Path
from urllib.parse import urlparse
import cv2
import numpy as np
@@ -21,37 +16,30 @@ import pandas as pd
import requests
import torch
import torch.nn as nn
from IPython.display import display
from PIL import Image
from torch.cuda import amp
from utils import TryExcept
from utils.dataloaders import exif_transpose, letterbox
from utils.general import (LOGGER, ROOT, Profile, check_requirements, check_suffix, check_version, colorstr,
increment_path, is_notebook, make_divisible, non_max_suppression, scale_boxes, xywh2xyxy,
xyxy2xywh, yaml_load)
from utils.datasets import exif_transpose, letterbox
from utils.general import (LOGGER, check_requirements, check_suffix, colorstr, increment_path, make_divisible,
non_max_suppression, scale_coords, xywh2xyxy, xyxy2xywh)
from utils.plots import Annotator, colors, save_one_box
from utils.torch_utils import copy_attr, smart_inference_mode
from utils.torch_utils import time_sync
def autopad(k, p=None, d=1): # kernel, padding, dilation
# Pad to 'same' shape outputs
if d > 1:
k = d * (k - 1) + 1 if isinstance(k, int) else [d * (x - 1) + 1 for x in k] # actual kernel-size
def autopad(k, p=None): # kernel, padding
# Pad to 'same'
if p is None:
p = k // 2 if isinstance(k, int) else [x // 2 for x in k] # auto-pad
return p
class Conv(nn.Module):
# Standard convolution with args(ch_in, ch_out, kernel, stride, padding, groups, dilation, activation)
default_act = nn.SiLU() # default activation
def __init__(self, c1, c2, k=1, s=1, p=None, g=1, d=1, act=True):
# Standard convolution
def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True): # ch_in, ch_out, kernel, stride, padding, groups
super().__init__()
self.conv = nn.Conv2d(c1, c2, k, s, autopad(k, p, d), groups=g, dilation=d, bias=False)
self.conv = nn.Conv2d(c1, c2, k, s, autopad(k, p), groups=g, bias=False)
self.bn = nn.BatchNorm2d(c2)
self.act = self.default_act if act is True else act if isinstance(act, nn.Module) else nn.Identity()
self.act = nn.SiLU() if act is True else (act if isinstance(act, nn.Module) else nn.Identity())
def forward(self, x):
return self.act(self.bn(self.conv(x)))
@@ -61,15 +49,9 @@ class Conv(nn.Module):
class DWConv(Conv):
# Depth-wise convolution
def __init__(self, c1, c2, k=1, s=1, d=1, act=True): # ch_in, ch_out, kernel, stride, dilation, activation
super().__init__(c1, c2, k, s, g=math.gcd(c1, c2), d=d, act=act)
class DWConvTranspose2d(nn.ConvTranspose2d):
# Depth-wise transpose convolution
def __init__(self, c1, c2, k=1, s=1, p1=0, p2=0): # ch_in, ch_out, kernel, stride, padding, padding_out
super().__init__(c1, c2, k, s, p1, p2, groups=math.gcd(c1, c2))
# Depth-wise convolution class
def __init__(self, c1, c2, k=1, s=1, act=True): # ch_in, ch_out, kernel, stride, padding, groups
super().__init__(c1, c2, k, s, g=math.gcd(c1, c2), act=act)
class TransformerLayer(nn.Module):
@@ -104,8 +86,8 @@ class TransformerBlock(nn.Module):
if self.conv is not None:
x = self.conv(x)
b, _, w, h = x.shape
p = x.flatten(2).permute(2, 0, 1)
return self.tr(p + self.linear(p)).permute(1, 2, 0).reshape(b, self.c2, w, h)
p = x.flatten(2).unsqueeze(0).transpose(0, 3).squeeze(3)
return self.tr(p + self.linear(p)).unsqueeze(3).transpose(0, 3).reshape(b, self.c2, w, h)
class Bottleneck(nn.Module):
@@ -137,21 +119,7 @@ class BottleneckCSP(nn.Module):
def forward(self, x):
y1 = self.cv3(self.m(self.cv1(x)))
y2 = self.cv2(x)
return self.cv4(self.act(self.bn(torch.cat((y1, y2), 1))))
class CrossConv(nn.Module):
# Cross Convolution Downsample
def __init__(self, c1, c2, k=3, s=1, g=1, e=1.0, shortcut=False):
# ch_in, ch_out, kernel, stride, groups, expansion, shortcut
super().__init__()
c_ = int(c2 * e) # hidden channels
self.cv1 = Conv(c1, c_, (1, k), (1, s))
self.cv2 = Conv(c_, c2, (k, 1), (s, 1), g=g)
self.add = shortcut and c1 == c2
def forward(self, x):
return x + self.cv2(self.cv1(x)) if self.add else self.cv2(self.cv1(x))
return self.cv4(self.act(self.bn(torch.cat((y1, y2), dim=1))))
class C3(nn.Module):
@@ -161,19 +129,12 @@ class C3(nn.Module):
c_ = int(c2 * e) # hidden channels
self.cv1 = Conv(c1, c_, 1, 1)
self.cv2 = Conv(c1, c_, 1, 1)
self.cv3 = Conv(2 * c_, c2, 1) # optional act=FReLU(c2)
self.cv3 = Conv(2 * c_, c2, 1) # act=FReLU(c2)
self.m = nn.Sequential(*(Bottleneck(c_, c_, shortcut, g, e=1.0) for _ in range(n)))
# self.m = nn.Sequential(*[CrossConv(c_, c_, 3, 1, g, 1.0, shortcut) for _ in range(n)])
def forward(self, x):
return self.cv3(torch.cat((self.m(self.cv1(x)), self.cv2(x)), 1))
class C3x(C3):
# C3 module with cross-convolutions
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 = nn.Sequential(*(CrossConv(c_, c_, 3, 1, g, 1.0, shortcut) for _ in range(n)))
return self.cv3(torch.cat((self.m(self.cv1(x)), self.cv2(x)), dim=1))
class C3TR(C3):
@@ -217,7 +178,7 @@ class SPP(nn.Module):
class SPPF(nn.Module):
# Spatial Pyramid Pooling - Fast (SPPF) layer for YOLOv3 by Glenn Jocher
# Spatial Pyramid Pooling - Fast (SPPF) layer for by Glenn Jocher
def __init__(self, c1, c2, k=5): # equivalent to SPP(k=(5, 9, 13))
super().__init__()
c_ = c1 // 2 # hidden channels
@@ -231,18 +192,18 @@ class SPPF(nn.Module):
warnings.simplefilter('ignore') # suppress torch 1.9.0 max_pool2d() warning
y1 = self.m(x)
y2 = self.m(y1)
return self.cv2(torch.cat((x, y1, y2, self.m(y2)), 1))
return self.cv2(torch.cat([x, y1, y2, self.m(y2)], 1))
class Focus(nn.Module):
# Focus wh information into c-space
def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True): # ch_in, ch_out, kernel, stride, padding, groups
super().__init__()
self.conv = Conv(c1 * 4, c2, k, s, p, g, act=act)
self.conv = Conv(c1 * 4, c2, k, s, p, g, act)
# self.contract = Contract(gain=2)
def forward(self, x): # x(b,c,w,h) -> y(b,4c,w/2,h/2)
return self.conv(torch.cat((x[..., ::2, ::2], x[..., 1::2, ::2], x[..., ::2, 1::2], x[..., 1::2, 1::2]), 1))
return self.conv(torch.cat([x[..., ::2, ::2], x[..., 1::2, ::2], x[..., ::2, 1::2], x[..., 1::2, 1::2]], 1))
# return self.conv(self.contract(x))
@@ -251,12 +212,12 @@ class GhostConv(nn.Module):
def __init__(self, c1, c2, k=1, s=1, g=1, act=True): # ch_in, ch_out, kernel, stride, groups
super().__init__()
c_ = c2 // 2 # hidden channels
self.cv1 = Conv(c1, c_, k, s, None, g, act=act)
self.cv2 = Conv(c_, c_, 5, 1, None, c_, act=act)
self.cv1 = Conv(c1, c_, k, s, None, g, act)
self.cv2 = Conv(c_, c_, 5, 1, None, c_, act)
def forward(self, x):
y = self.cv1(x)
return torch.cat((y, self.cv2(y)), 1)
return torch.cat([y, self.cv2(y)], 1)
class GhostBottleneck(nn.Module):
@@ -264,12 +225,11 @@ class GhostBottleneck(nn.Module):
def __init__(self, c1, c2, k=3, s=1): # ch_in, ch_out, kernel, stride
super().__init__()
c_ = c2 // 2
self.conv = nn.Sequential(
GhostConv(c1, c_, 1, 1), # pw
DWConv(c_, c_, k, s, act=False) if s == 2 else nn.Identity(), # dw
GhostConv(c_, c2, 1, 1, act=False)) # pw-linear
self.shortcut = nn.Sequential(DWConv(c1, c1, k, s, act=False), Conv(c1, c2, 1, 1,
act=False)) if s == 2 else nn.Identity()
self.conv = nn.Sequential(GhostConv(c1, c_, 1, 1), # pw
DWConv(c_, c_, k, s, act=False) if s == 2 else nn.Identity(), # dw
GhostConv(c_, c2, 1, 1, act=False)) # pw-linear
self.shortcut = nn.Sequential(DWConv(c1, c1, k, s, act=False),
Conv(c1, c2, 1, 1, act=False)) if s == 2 else nn.Identity()
def forward(self, x):
return self.conv(x) + self.shortcut(x)
@@ -314,350 +274,159 @@ class Concat(nn.Module):
class DetectMultiBackend(nn.Module):
# YOLOv3 MultiBackend class for python inference on various backends
def __init__(self, weights='yolov5s.pt', device=torch.device('cpu'), dnn=False, data=None, fp16=False, fuse=True):
# MultiBackend class for python inference on various backends
def __init__(self, weights='yolov3.pt', device=None, dnn=True):
# Usage:
# PyTorch: weights = *.pt
# TorchScript: *.torchscript
# ONNX Runtime: *.onnx
# ONNX OpenCV DNN: *.onnx --dnn
# OpenVINO: *_openvino_model
# CoreML: *.mlmodel
# TensorRT: *.engine
# TensorFlow SavedModel: *_saved_model
# TensorFlow GraphDef: *.pb
# TensorFlow Lite: *.tflite
# TensorFlow Edge TPU: *_edgetpu.tflite
# PaddlePaddle: *_paddle_model
from models.experimental import attempt_download, attempt_load # scoped to avoid circular import
# PyTorch: weights = *.pt
# TorchScript: *.torchscript.pt
# CoreML: *.mlmodel
# TensorFlow: *_saved_model
# TensorFlow: *.pb
# TensorFlow Lite: *.tflite
# ONNX Runtime: *.onnx
# OpenCV DNN: *.onnx with dnn=True
super().__init__()
w = str(weights[0] if isinstance(weights, list) else weights)
pt, jit, onnx, xml, engine, coreml, saved_model, pb, tflite, edgetpu, tfjs, paddle, triton = self._model_type(w)
fp16 &= pt or jit or onnx or engine # FP16
nhwc = coreml or saved_model or pb or tflite or edgetpu # BHWC formats (vs torch BCWH)
stride = 32 # default stride
cuda = torch.cuda.is_available() and device.type != 'cpu' # use CUDA
if not (pt or triton):
w = attempt_download(w) # download if not local
suffix, suffixes = Path(w).suffix.lower(), ['.pt', '.onnx', '.tflite', '.pb', '', '.mlmodel']
check_suffix(w, suffixes) # check weights have acceptable suffix
pt, onnx, tflite, pb, saved_model, coreml = (suffix == x for x in suffixes) # backend booleans
jit = pt and 'torchscript' in w.lower()
stride, names = 64, [f'class{i}' for i in range(1000)] # assign defaults
if pt: # PyTorch
model = attempt_load(weights if isinstance(weights, list) else w, device=device, inplace=True, fuse=fuse)
stride = max(int(model.stride.max()), 32) # model stride
names = model.module.names if hasattr(model, 'module') else model.names # get class names
model.half() if fp16 else model.float()
self.model = model # explicitly assign for to(), cpu(), cuda(), half()
elif jit: # TorchScript
if jit: # TorchScript
LOGGER.info(f'Loading {w} for TorchScript inference...')
extra_files = {'config.txt': ''} # model metadata
model = torch.jit.load(w, _extra_files=extra_files, map_location=device)
model.half() if fp16 else model.float()
if extra_files['config.txt']: # load metadata dict
d = json.loads(extra_files['config.txt'],
object_hook=lambda d: {int(k) if k.isdigit() else k: v
for k, v in d.items()})
model = torch.jit.load(w, _extra_files=extra_files)
if extra_files['config.txt']:
d = json.loads(extra_files['config.txt']) # extra_files dict
stride, names = int(d['stride']), d['names']
elif pt: # PyTorch
from models.experimental import attempt_load # scoped to avoid circular import
model = torch.jit.load(w) if 'torchscript' in w else attempt_load(weights, map_location=device)
stride = int(model.stride.max()) # model stride
names = model.module.names if hasattr(model, 'module') else model.names # get class names
elif coreml: # CoreML *.mlmodel
import coremltools as ct
model = ct.models.MLModel(w)
elif dnn: # ONNX OpenCV DNN
LOGGER.info(f'Loading {w} for ONNX OpenCV DNN inference...')
check_requirements('opencv-python>=4.5.4')
check_requirements(('opencv-python>=4.5.4',))
net = cv2.dnn.readNetFromONNX(w)
elif onnx: # ONNX Runtime
LOGGER.info(f'Loading {w} for ONNX Runtime inference...')
cuda = torch.cuda.is_available()
check_requirements(('onnx', 'onnxruntime-gpu' if cuda else 'onnxruntime'))
import onnxruntime
providers = ['CUDAExecutionProvider', 'CPUExecutionProvider'] if cuda else ['CPUExecutionProvider']
session = onnxruntime.InferenceSession(w, providers=providers)
output_names = [x.name for x in session.get_outputs()]
meta = session.get_modelmeta().custom_metadata_map # metadata
if 'stride' in meta:
stride, names = int(meta['stride']), eval(meta['names'])
elif xml: # OpenVINO
LOGGER.info(f'Loading {w} for OpenVINO inference...')
check_requirements('openvino') # requires openvino-dev: https://pypi.org/project/openvino-dev/
from openvino.runtime import Core, Layout, get_batch
ie = Core()
if not Path(w).is_file(): # if not *.xml
w = next(Path(w).glob('*.xml')) # get *.xml file from *_openvino_model dir
network = ie.read_model(model=w, weights=Path(w).with_suffix('.bin'))
if network.get_parameters()[0].get_layout().empty:
network.get_parameters()[0].set_layout(Layout('NCHW'))
batch_dim = get_batch(network)
if batch_dim.is_static:
batch_size = batch_dim.get_length()
executable_network = ie.compile_model(network, device_name='CPU') # device_name="MYRIAD" for Intel NCS2
stride, names = self._load_metadata(Path(w).with_suffix('.yaml')) # load metadata
elif engine: # TensorRT
LOGGER.info(f'Loading {w} for TensorRT inference...')
import tensorrt as trt # https://developer.nvidia.com/nvidia-tensorrt-download
check_version(trt.__version__, '7.0.0', hard=True) # require tensorrt>=7.0.0
if device.type == 'cpu':
device = torch.device('cuda:0')
Binding = namedtuple('Binding', ('name', 'dtype', 'shape', 'data', 'ptr'))
logger = trt.Logger(trt.Logger.INFO)
with open(w, 'rb') as f, trt.Runtime(logger) as runtime:
model = runtime.deserialize_cuda_engine(f.read())
context = model.create_execution_context()
bindings = OrderedDict()
output_names = []
fp16 = False # default updated below
dynamic = False
for i in range(model.num_bindings):
name = model.get_binding_name(i)
dtype = trt.nptype(model.get_binding_dtype(i))
if model.binding_is_input(i):
if -1 in tuple(model.get_binding_shape(i)): # dynamic
dynamic = True
context.set_binding_shape(i, tuple(model.get_profile_shape(0, i)[2]))
if dtype == np.float16:
fp16 = True
else: # output
output_names.append(name)
shape = tuple(context.get_binding_shape(i))
im = torch.from_numpy(np.empty(shape, dtype=dtype)).to(device)
bindings[name] = Binding(name, dtype, shape, im, int(im.data_ptr()))
binding_addrs = OrderedDict((n, d.ptr) for n, d in bindings.items())
batch_size = bindings['images'].shape[0] # if dynamic, this is instead max batch size
elif coreml: # CoreML
LOGGER.info(f'Loading {w} for CoreML inference...')
import coremltools as ct
model = ct.models.MLModel(w)
elif saved_model: # TF SavedModel
LOGGER.info(f'Loading {w} for TensorFlow SavedModel inference...')
import tensorflow as tf
keras = False # assume TF1 saved_model
model = tf.keras.models.load_model(w) if keras else tf.saved_model.load(w)
elif pb: # GraphDef https://www.tensorflow.org/guide/migrate#a_graphpb_or_graphpbtxt
LOGGER.info(f'Loading {w} for TensorFlow GraphDef inference...')
else: # TensorFlow model (TFLite, pb, saved_model)
import tensorflow as tf
if pb: # https://www.tensorflow.org/guide/migrate#a_graphpb_or_graphpbtxt
def wrap_frozen_graph(gd, inputs, outputs):
x = tf.compat.v1.wrap_function(lambda: tf.compat.v1.import_graph_def(gd, name=""), []) # wrapped
return x.prune(tf.nest.map_structure(x.graph.as_graph_element, inputs),
tf.nest.map_structure(x.graph.as_graph_element, outputs))
def wrap_frozen_graph(gd, inputs, outputs):
x = tf.compat.v1.wrap_function(lambda: tf.compat.v1.import_graph_def(gd, name=''), []) # wrapped
ge = x.graph.as_graph_element
return x.prune(tf.nest.map_structure(ge, inputs), tf.nest.map_structure(ge, outputs))
def gd_outputs(gd):
name_list, input_list = [], []
for node in gd.node: # tensorflow.core.framework.node_def_pb2.NodeDef
name_list.append(node.name)
input_list.extend(node.input)
return sorted(f'{x}:0' for x in list(set(name_list) - set(input_list)) if not x.startswith('NoOp'))
gd = tf.Graph().as_graph_def() # TF GraphDef
with open(w, 'rb') as f:
gd.ParseFromString(f.read())
frozen_func = wrap_frozen_graph(gd, inputs='x:0', outputs=gd_outputs(gd))
elif tflite or edgetpu: # https://www.tensorflow.org/lite/guide/python#install_tensorflow_lite_for_python
try: # https://coral.ai/docs/edgetpu/tflite-python/#update-existing-tf-lite-code-for-the-edge-tpu
from tflite_runtime.interpreter import Interpreter, load_delegate
except ImportError:
import tensorflow as tf
Interpreter, load_delegate = tf.lite.Interpreter, tf.lite.experimental.load_delegate,
if edgetpu: # TF Edge TPU https://coral.ai/software/#edgetpu-runtime
LOGGER.info(f'Loading {w} for TensorFlow Lite Edge TPU inference...')
delegate = {
'Linux': 'libedgetpu.so.1',
'Darwin': 'libedgetpu.1.dylib',
'Windows': 'edgetpu.dll'}[platform.system()]
interpreter = Interpreter(model_path=w, experimental_delegates=[load_delegate(delegate)])
else: # TFLite
LOGGER.info(f'Loading {w} for TensorFlow Lite inference...')
interpreter = Interpreter(model_path=w) # load TFLite model
interpreter.allocate_tensors() # allocate
input_details = interpreter.get_input_details() # inputs
output_details = interpreter.get_output_details() # outputs
# load metadata
with contextlib.suppress(zipfile.BadZipFile):
with zipfile.ZipFile(w, 'r') as model:
meta_file = model.namelist()[0]
meta = ast.literal_eval(model.read(meta_file).decode('utf-8'))
stride, names = int(meta['stride']), meta['names']
elif tfjs: # TF.js
raise NotImplementedError('ERROR: YOLOv3 TF.js inference is not supported')
elif paddle: # PaddlePaddle
LOGGER.info(f'Loading {w} for PaddlePaddle inference...')
check_requirements('paddlepaddle-gpu' if cuda else 'paddlepaddle')
import paddle.inference as pdi
if not Path(w).is_file(): # if not *.pdmodel
w = next(Path(w).rglob('*.pdmodel')) # get *.pdmodel file from *_paddle_model dir
weights = Path(w).with_suffix('.pdiparams')
config = pdi.Config(str(w), str(weights))
if cuda:
config.enable_use_gpu(memory_pool_init_size_mb=2048, device_id=0)
predictor = pdi.create_predictor(config)
input_handle = predictor.get_input_handle(predictor.get_input_names()[0])
output_names = predictor.get_output_names()
elif triton: # NVIDIA Triton Inference Server
LOGGER.info(f'Using {w} as Triton Inference Server...')
check_requirements('tritonclient[all]')
from utils.triton import TritonRemoteModel
model = TritonRemoteModel(url=w)
nhwc = model.runtime.startswith('tensorflow')
else:
raise NotImplementedError(f'ERROR: {w} is not a supported format')
# class names
if 'names' not in locals():
names = yaml_load(data)['names'] if data else {i: f'class{i}' for i in range(999)}
if names[0] == 'n01440764' and len(names) == 1000: # ImageNet
names = yaml_load(ROOT / 'data/ImageNet.yaml')['names'] # human-readable names
LOGGER.info(f'Loading {w} for TensorFlow *.pb inference...')
graph_def = tf.Graph().as_graph_def()
graph_def.ParseFromString(open(w, 'rb').read())
frozen_func = wrap_frozen_graph(gd=graph_def, inputs="x:0", outputs="Identity:0")
elif saved_model:
LOGGER.info(f'Loading {w} for TensorFlow saved_model inference...')
model = tf.keras.models.load_model(w)
elif tflite: # https://www.tensorflow.org/lite/guide/python#install_tensorflow_lite_for_python
if 'edgetpu' in w.lower():
LOGGER.info(f'Loading {w} for TensorFlow Edge TPU inference...')
import tflite_runtime.interpreter as tfli
delegate = {'Linux': 'libedgetpu.so.1', # install https://coral.ai/software/#edgetpu-runtime
'Darwin': 'libedgetpu.1.dylib',
'Windows': 'edgetpu.dll'}[platform.system()]
interpreter = tfli.Interpreter(model_path=w, experimental_delegates=[tfli.load_delegate(delegate)])
else:
LOGGER.info(f'Loading {w} for TensorFlow Lite inference...')
interpreter = tf.lite.Interpreter(model_path=w) # load TFLite model
interpreter.allocate_tensors() # allocate
input_details = interpreter.get_input_details() # inputs
output_details = interpreter.get_output_details() # outputs
self.__dict__.update(locals()) # assign all variables to self
def forward(self, im, augment=False, visualize=False):
# YOLOv3 MultiBackend inference
def forward(self, im, augment=False, visualize=False, val=False):
# MultiBackend inference
b, ch, h, w = im.shape # batch, channel, height, width
if self.fp16 and im.dtype != torch.float16:
im = im.half() # to FP16
if self.nhwc:
im = im.permute(0, 2, 3, 1) # torch BCHW to numpy BHWC shape(1,320,192,3)
if self.pt: # PyTorch
y = self.model(im, augment=augment, visualize=visualize) if augment or visualize else self.model(im)
elif self.jit: # TorchScript
y = self.model(im)
elif self.dnn: # ONNX OpenCV DNN
im = im.cpu().numpy() # torch to numpy
self.net.setInput(im)
y = self.net.forward()
elif self.onnx: # ONNX Runtime
im = im.cpu().numpy() # torch to numpy
y = self.session.run(self.output_names, {self.session.get_inputs()[0].name: im})
elif self.xml: # OpenVINO
im = im.cpu().numpy() # FP32
y = list(self.executable_network([im]).values())
elif self.engine: # TensorRT
if self.dynamic and im.shape != self.bindings['images'].shape:
i = self.model.get_binding_index('images')
self.context.set_binding_shape(i, im.shape) # reshape if dynamic
self.bindings['images'] = self.bindings['images']._replace(shape=im.shape)
for name in self.output_names:
i = self.model.get_binding_index(name)
self.bindings[name].data.resize_(tuple(self.context.get_binding_shape(i)))
s = self.bindings['images'].shape
assert im.shape == s, f"input size {im.shape} {'>' if self.dynamic else 'not equal to'} max model size {s}"
self.binding_addrs['images'] = int(im.data_ptr())
self.context.execute_v2(list(self.binding_addrs.values()))
y = [self.bindings[x].data for x in sorted(self.output_names)]
elif self.coreml: # CoreML
im = im.cpu().numpy()
y = self.model(im) if self.jit else self.model(im, augment=augment, visualize=visualize)
return y if val else y[0]
elif self.coreml: # CoreML *.mlmodel
im = im.permute(0, 2, 3, 1).cpu().numpy() # torch BCHW to numpy BHWC shape(1,320,192,3)
im = Image.fromarray((im[0] * 255).astype('uint8'))
# im = im.resize((192, 320), Image.ANTIALIAS)
y = self.model.predict({'image': im}) # coordinates are xywh normalized
if 'confidence' in y:
box = xywh2xyxy(y['coordinates'] * [[w, h, w, h]]) # xyxy pixels
conf, cls = y['confidence'].max(1), y['confidence'].argmax(1).astype(np.float)
y = np.concatenate((box, conf.reshape(-1, 1), cls.reshape(-1, 1)), 1)
else:
y = list(reversed(y.values())) # reversed for segmentation models (pred, proto)
elif self.paddle: # PaddlePaddle
im = im.cpu().numpy().astype(np.float32)
self.input_handle.copy_from_cpu(im)
self.predictor.run()
y = [self.predictor.get_output_handle(x).copy_to_cpu() for x in self.output_names]
elif self.triton: # NVIDIA Triton Inference Server
y = self.model(im)
else: # TensorFlow (SavedModel, GraphDef, Lite, Edge TPU)
im = im.cpu().numpy()
if self.saved_model: # SavedModel
y = self.model(im, training=False) if self.keras else self.model(im)
elif self.pb: # GraphDef
y = self.frozen_func(x=self.tf.constant(im))
else: # Lite or Edge TPU
input = self.input_details[0]
box = xywh2xyxy(y['coordinates'] * [[w, h, w, h]]) # xyxy pixels
conf, cls = y['confidence'].max(1), y['confidence'].argmax(1).astype(np.float)
y = np.concatenate((box, conf.reshape(-1, 1), cls.reshape(-1, 1)), 1)
elif self.onnx: # ONNX
im = im.cpu().numpy() # torch to numpy
if self.dnn: # ONNX OpenCV DNN
self.net.setInput(im)
y = self.net.forward()
else: # ONNX Runtime
y = self.session.run([self.session.get_outputs()[0].name], {self.session.get_inputs()[0].name: im})[0]
else: # TensorFlow model (TFLite, pb, saved_model)
im = im.permute(0, 2, 3, 1).cpu().numpy() # torch BCHW to numpy BHWC shape(1,320,192,3)
if self.pb:
y = self.frozen_func(x=self.tf.constant(im)).numpy()
elif self.saved_model:
y = self.model(im, training=False).numpy()
elif self.tflite:
input, output = self.input_details[0], self.output_details[0]
int8 = input['dtype'] == np.uint8 # is TFLite quantized uint8 model
if int8:
scale, zero_point = input['quantization']
im = (im / scale + zero_point).astype(np.uint8) # de-scale
self.interpreter.set_tensor(input['index'], im)
self.interpreter.invoke()
y = []
for output in self.output_details:
x = self.interpreter.get_tensor(output['index'])
if int8:
scale, zero_point = output['quantization']
x = (x.astype(np.float32) - zero_point) * scale # re-scale
y.append(x)
y = [x if isinstance(x, np.ndarray) else x.numpy() for x in y]
y[0][..., :4] *= [w, h, w, h] # xywh normalized to pixels
if isinstance(y, (list, tuple)):
return self.from_numpy(y[0]) if len(y) == 1 else [self.from_numpy(x) for x in y]
else:
return self.from_numpy(y)
def from_numpy(self, x):
return torch.from_numpy(x).to(self.device) if isinstance(x, np.ndarray) else x
def warmup(self, imgsz=(1, 3, 640, 640)):
# Warmup model by running inference once
warmup_types = self.pt, self.jit, self.onnx, self.engine, self.saved_model, self.pb, self.triton
if any(warmup_types) and (self.device.type != 'cpu' or self.triton):
im = torch.empty(*imgsz, dtype=torch.half if self.fp16 else torch.float, device=self.device) # input
for _ in range(2 if self.jit else 1): #
self.forward(im) # warmup
@staticmethod
def _model_type(p='path/to/model.pt'):
# Return model type from model path, i.e. path='path/to/model.onnx' -> type=onnx
# types = [pt, jit, onnx, xml, engine, coreml, saved_model, pb, tflite, edgetpu, tfjs, paddle]
from export import export_formats
from utils.downloads import is_url
sf = list(export_formats().Suffix) # export suffixes
if not is_url(p, check=False):
check_suffix(p, sf) # checks
url = urlparse(p) # if url may be Triton inference server
types = [s in Path(p).name for s in sf]
types[8] &= not types[9] # tflite &= not edgetpu
triton = not any(types) and all([any(s in url.scheme for s in ['http', 'grpc']), url.netloc])
return types + [triton]
@staticmethod
def _load_metadata(f=Path('path/to/meta.yaml')):
# Load metadata from meta.yaml if it exists
if f.exists():
d = yaml_load(f)
return d['stride'], d['names'] # assign stride, names
return None, None
y = self.interpreter.get_tensor(output['index'])
if int8:
scale, zero_point = output['quantization']
y = (y.astype(np.float32) - zero_point) * scale # re-scale
y[..., 0] *= w # x
y[..., 1] *= h # y
y[..., 2] *= w # w
y[..., 3] *= h # h
y = torch.tensor(y)
return (y, []) if val else y
class AutoShape(nn.Module):
# YOLOv3 input-robust model wrapper for passing cv2/np/PIL/torch inputs. Includes preprocessing, inference and NMS
# input-robust model wrapper for passing cv2/np/PIL/torch inputs. Includes preprocessing, inference and NMS
conf = 0.25 # NMS confidence threshold
iou = 0.45 # NMS IoU threshold
agnostic = False # NMS class-agnostic
multi_label = False # NMS multiple labels per box
classes = None # (optional list) filter by class, i.e. = [0, 15, 16] for COCO persons, cats and dogs
multi_label = False # NMS multiple labels per box
max_det = 1000 # maximum number of detections per image
amp = False # Automatic Mixed Precision (AMP) inference
def __init__(self, model, verbose=True):
def __init__(self, model):
super().__init__()
if verbose:
LOGGER.info('Adding AutoShape... ')
copy_attr(self, model, include=('yaml', 'nc', 'hyp', 'names', 'stride', 'abc'), exclude=()) # copy attributes
self.dmb = isinstance(model, DetectMultiBackend) # DetectMultiBackend() instance
self.pt = not self.dmb or model.pt # PyTorch model
self.model = model.eval()
if self.pt:
m = self.model.model.model[-1] if self.dmb else self.model.model[-1] # Detect()
m.inplace = False # Detect.inplace=False for safe multithread inference
m.export = True # do not output loss values
def autoshape(self):
LOGGER.info('AutoShape already enabled, skipping... ') # model already converted to model.autoshape()
return self
def _apply(self, fn):
# Apply to(), cpu(), cuda(), half() to model tensors that are not parameters or registered buffers
self = super()._apply(fn)
if self.pt:
m = self.model.model.model[-1] if self.dmb else self.model.model[-1] # Detect()
m.stride = fn(m.stride)
m.grid = list(map(fn, m.grid))
if isinstance(m.anchor_grid, list):
m.anchor_grid = list(map(fn, m.anchor_grid))
m = self.model.model[-1] # Detect()
m.stride = fn(m.stride)
m.grid = list(map(fn, m.grid))
if isinstance(m.anchor_grid, list):
m.anchor_grid = list(map(fn, m.anchor_grid))
return self
@smart_inference_mode()
def forward(self, ims, size=640, augment=False, profile=False):
# Inference from various sources. For size(height=640, width=1280), RGB images example inputs are:
# file: ims = 'data/images/zidane.jpg' # str or PosixPath
@torch.no_grad()
def forward(self, imgs, size=640, augment=False, profile=False):
# Inference from various sources. For height=640, width=1280, RGB images example inputs are:
# file: imgs = 'data/images/zidane.jpg' # str or PosixPath
# URI: = 'https://ultralytics.com/images/zidane.jpg'
# OpenCV: = cv2.imread('image.jpg')[:,:,::-1] # HWC BGR to RGB x(640,1280,3)
# PIL: = Image.open('image.jpg') or ImageGrab.grab() # HWC x(640,1280,3)
@@ -665,139 +434,129 @@ class AutoShape(nn.Module):
# 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
dt = (Profile(), Profile(), Profile())
with dt[0]:
if isinstance(size, int): # expand
size = (size, size)
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
with amp.autocast(autocast):
return self.model(ims.to(p.device).type_as(p), augment=augment) # inference
t = [time_sync()]
p = next(self.model.parameters()) # for device and type
if isinstance(imgs, torch.Tensor): # torch
with amp.autocast(enabled=p.device.type != 'cpu'):
return self.model(imgs.to(p.device).type_as(p), augment, profile) # inference
# Pre-process
n, ims = (len(ims), list(ims)) if isinstance(ims, (list, tuple)) else (1, [ims]) # number, list of images
shape0, shape1, files = [], [], [] # image and inference shapes, filenames
for i, im in enumerate(ims):
f = f'image{i}' # filename
if isinstance(im, (str, Path)): # filename or uri
im, f = Image.open(requests.get(im, stream=True).raw if str(im).startswith('http') else im), im
im = np.asarray(exif_transpose(im))
elif isinstance(im, Image.Image): # PIL Image
im, f = np.asarray(exif_transpose(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 cv2.cvtColor(im, cv2.COLOR_GRAY2BGR) # enforce 3ch input
s = im.shape[:2] # HWC
shape0.append(s) # image shape
g = max(size) / max(s) # gain
shape1.append([int(y * g) for y in s])
ims[i] = im if im.data.contiguous else np.ascontiguousarray(im) # update
shape1 = [make_divisible(x, self.stride) for x in np.array(shape1).max(0)] # inf shape
x = [letterbox(im, shape1, auto=False)[0] for im in ims] # pad
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
# Pre-process
n, imgs = (len(imgs), imgs) if isinstance(imgs, list) else (1, [imgs]) # number of images, list of images
shape0, shape1, files = [], [], [] # image and inference shapes, filenames
for i, im in enumerate(imgs):
f = f'image{i}' # filename
if isinstance(im, (str, Path)): # filename or uri
im, f = Image.open(requests.get(im, stream=True).raw if str(im).startswith('http') else im), im
im = np.asarray(exif_transpose(im))
elif isinstance(im, Image.Image): # PIL Image
im, f = np.asarray(exif_transpose(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
s = im.shape[:2] # HWC
shape0.append(s) # image shape
g = (size / max(s)) # gain
shape1.append([y * g for y in s])
imgs[i] = im if im.data.contiguous else np.ascontiguousarray(im) # update
shape1 = [make_divisible(x, int(self.stride.max())) for x in np.stack(shape1, 0).max(0)] # inference shape
x = [letterbox(im, new_shape=shape1, auto=False)[0] for im in imgs] # pad
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_sync())
with amp.autocast(autocast):
with amp.autocast(enabled=p.device.type != 'cpu'):
# Inference
with dt[1]:
y = self.model(x, augment=augment) # forward
y = self.model(x, augment, profile)[0] # forward
t.append(time_sync())
# Post-process
with dt[2]:
y = non_max_suppression(y if self.dmb else y[0],
self.conf,
self.iou,
self.classes,
self.agnostic,
self.multi_label,
max_det=self.max_det) # NMS
for i in range(n):
scale_boxes(shape1, y[i][:, :4], shape0[i])
y = non_max_suppression(y, self.conf, iou_thres=self.iou, classes=self.classes,
multi_label=self.multi_label, max_det=self.max_det) # NMS
for i in range(n):
scale_coords(shape1, y[i][:, :4], shape0[i])
return Detections(ims, y, files, dt, self.names, x.shape)
t.append(time_sync())
return Detections(imgs, y, files, t, self.names, x.shape)
class Detections:
# YOLOv3 detections class for inference results
def __init__(self, ims, pred, files, times=(0, 0, 0), names=None, shape=None):
# detections class for inference results
def __init__(self, imgs, pred, files, times=None, names=None, shape=None):
super().__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 ims] # normalizations
self.ims = ims # list of images as numpy arrays
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.times = times # profiling times
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) # number of images (batch size)
self.t = tuple(x.t / self.n * 1E3 for x in times) # timestamps (ms)
self.s = tuple(shape) # inference BCHW shape
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 _run(self, pprint=False, show=False, save=False, crop=False, render=False, labels=True, save_dir=Path('')):
s, crops = '', []
for i, (im, pred) in enumerate(zip(self.ims, self.pred)):
s += f'\nimage {i + 1}/{len(self.pred)}: {im.shape[0]}x{im.shape[1]} ' # string
def display(self, pprint=False, show=False, save=False, crop=False, render=False, save_dir=Path('')):
crops = []
for i, (im, pred) in enumerate(zip(self.imgs, self.pred)):
s = f'image {i + 1}/{len(self.pred)}: {im.shape[0]}x{im.shape[1]} ' # string
if pred.shape[0]:
for c in pred[:, -1].unique():
n = (pred[:, -1] == c).sum() # detections per class
s += f"{n} {self.names[int(c)]}{'s' * (n > 1)}, " # add to string
s = s.rstrip(', ')
if show or save or render or crop:
annotator = Annotator(im, example=str(self.names))
for *box, conf, cls in reversed(pred): # xyxy, confidence, class
label = f'{self.names[int(cls)]} {conf:.2f}'
if crop:
file = save_dir / 'crops' / self.names[int(cls)] / self.files[i] if save else None
crops.append({
'box': box,
'conf': conf,
'cls': cls,
'label': label,
'im': save_one_box(box, im, file=file, save=save)})
crops.append({'box': box, 'conf': conf, 'cls': cls, 'label': label,
'im': save_one_box(box, im, file=file, save=save)})
else: # all others
annotator.box_label(box, label if labels else '', color=colors(cls))
annotator.box_label(box, label, color=colors(cls))
im = annotator.im
else:
s += '(no detections)'
im = Image.fromarray(im.astype(np.uint8)) if isinstance(im, np.ndarray) else im # from np
if pprint:
LOGGER.info(s.rstrip(', '))
if show:
display(im) if is_notebook() else im.show(self.files[i])
im.show(self.files[i]) # show
if save:
f = self.files[i]
im.save(save_dir / f) # save
if i == self.n - 1:
LOGGER.info(f"Saved {self.n} image{'s' * (self.n > 1)} to {colorstr('bold', save_dir)}")
if render:
self.ims[i] = np.asarray(im)
if pprint:
s = s.lstrip('\n')
return f'{s}\nSpeed: %.1fms pre-process, %.1fms inference, %.1fms NMS per image at shape {self.s}' % self.t
self.imgs[i] = np.asarray(im)
if crop:
if save:
LOGGER.info(f'Saved results to {save_dir}\n')
return crops
@TryExcept('Showing images is not supported in this environment')
def show(self, labels=True):
self._run(show=True, labels=labels) # show results
def print(self):
self.display(pprint=True) # print results
LOGGER.info(f'Speed: %.1fms pre-process, %.1fms inference, %.1fms NMS per image at shape {tuple(self.s)}' %
self.t)
def save(self, labels=True, save_dir='runs/detect/exp', exist_ok=False):
save_dir = increment_path(save_dir, exist_ok, mkdir=True) # increment save_dir
self._run(save=True, labels=labels, save_dir=save_dir) # save results
def show(self):
self.display(show=True) # show results
def crop(self, save=True, save_dir='runs/detect/exp', exist_ok=False):
save_dir = increment_path(save_dir, exist_ok, mkdir=True) if save else None
return self._run(crop=True, save=save, save_dir=save_dir) # crop results
def save(self, save_dir='runs/detect/exp'):
save_dir = increment_path(save_dir, exist_ok=save_dir != 'runs/detect/exp', mkdir=True) # increment save_dir
self.display(save=True, save_dir=save_dir) # save results
def render(self, labels=True):
self._run(render=True, labels=labels) # render results
return self.ims
def crop(self, save=True, save_dir='runs/detect/exp'):
save_dir = increment_path(save_dir, exist_ok=save_dir != 'runs/detect/exp', mkdir=True) if save else None
return self.display(crop=True, save=save, save_dir=save_dir) # crop results
def render(self):
self.display(render=True) # render results
return self.imgs
def pandas(self):
# return detections as pandas DataFrames, i.e. print(results.pandas().xyxy[0])
@@ -811,57 +570,24 @@ class Detections:
def tolist(self):
# return a list of Detections objects, i.e. 'for result in results.tolist():'
r = range(self.n) # iterable
x = [Detections([self.ims[i]], [self.pred[i]], [self.files[i]], self.times, self.names, self.s) for i in r]
# for d in x:
# for k in ['ims', 'pred', 'xyxy', 'xyxyn', 'xywh', 'xywhn']:
# setattr(d, k, getattr(d, k)[0]) # pop out of list
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 print(self):
LOGGER.info(self.__str__())
def __len__(self): # override len(results)
def __len__(self):
return self.n
def __str__(self): # override print(results)
return self._run(pprint=True) # print results
def __repr__(self):
return f'YOLOv3 {self.__class__} instance\n' + self.__str__()
class Proto(nn.Module):
# YOLOv3 mask Proto module for segmentation models
def __init__(self, c1, c_=256, c2=32): # ch_in, number of protos, number of masks
super().__init__()
self.cv1 = Conv(c1, c_, k=3)
self.upsample = nn.Upsample(scale_factor=2, mode='nearest')
self.cv2 = Conv(c_, c_, k=3)
self.cv3 = Conv(c_, c2)
def forward(self, x):
return self.cv3(self.cv2(self.upsample(self.cv1(x))))
class Classify(nn.Module):
# YOLOv3 classification head, i.e. x(b,c1,20,20) to x(b,c2)
def __init__(self,
c1,
c2,
k=1,
s=1,
p=None,
g=1,
dropout_p=0.0): # ch_in, ch_out, kernel, stride, padding, groups, dropout probability
# Classification head, i.e. x(b,c1,20,20) to x(b,c2)
def __init__(self, c1, c2, k=1, s=1, p=None, g=1): # ch_in, ch_out, kernel, stride, padding, groups
super().__init__()
c_ = 1280 # efficientnet_b0 size
self.conv = Conv(c1, c_, k, s, autopad(k, p), g)
self.pool = nn.AdaptiveAvgPool2d(1) # to x(b,c_,1,1)
self.drop = nn.Dropout(p=dropout_p, inplace=True)
self.linear = nn.Linear(c_, c2) # to x(b,c2)
self.aap = nn.AdaptiveAvgPool2d(1) # to x(b,c1,1,1)
self.conv = nn.Conv2d(c1, c2, k, s, autopad(k, p), groups=g) # to x(b,c2,1,1)
self.flat = nn.Flatten()
def forward(self, x):
if isinstance(x, list):
x = torch.cat(x, 1)
return self.linear(self.drop(self.pool(self.conv(x)).flatten(1)))
z = torch.cat([self.aap(y) for y in (x if isinstance(x, list) else [x])], 1) # cat if list
return self.flat(self.conv(z)) # flatten to x(b,c2)
yolov3/models/experimental.py
+39 -29
View File
@@ -8,9 +8,24 @@ import numpy as np
import torch
import torch.nn as nn
from models.common import Conv
from utils.downloads import attempt_download
class CrossConv(nn.Module):
# Cross Convolution Downsample
def __init__(self, c1, c2, k=3, s=1, g=1, e=1.0, shortcut=False):
# ch_in, ch_out, kernel, stride, groups, expansion, shortcut
super().__init__()
c_ = int(c2 * e) # hidden channels
self.cv1 = Conv(c1, c_, (1, k), (1, s))
self.cv2 = Conv(c_, c2, (k, 1), (s, 1), g=g)
self.add = shortcut and c1 == c2
def forward(self, x):
return x + self.cv2(self.cv1(x)) if self.add else self.cv2(self.cv1(x))
class Sum(nn.Module):
# Weighted sum of 2 or more layers https://arxiv.org/abs/1911.09070
def __init__(self, n, weight=False): # n: number of inputs
@@ -48,8 +63,8 @@ class MixConv2d(nn.Module):
a[0] = 1
c_ = np.linalg.lstsq(a, b, rcond=None)[0].round() # solve for equal weight indices, ax = b
self.m = nn.ModuleList([
nn.Conv2d(c1, int(c_), k, s, k // 2, groups=math.gcd(c1, int(c_)), bias=False) for k, c_ in zip(k, c_)])
self.m = nn.ModuleList(
[nn.Conv2d(c1, int(c_), k, s, k // 2, groups=math.gcd(c1, int(c_)), bias=False) for k, c_ in zip(k, c_)])
self.bn = nn.BatchNorm2d(c2)
self.act = nn.SiLU()
@@ -63,49 +78,44 @@ class Ensemble(nn.ModuleList):
super().__init__()
def forward(self, x, augment=False, profile=False, visualize=False):
y = [module(x, augment, profile, visualize)[0] for module in self]
y = []
for module in self:
y.append(module(x, augment, profile, visualize)[0])
# y = torch.stack(y).max(0)[0] # max ensemble
# y = torch.stack(y).mean(0) # mean ensemble
y = torch.cat(y, 1) # nms ensemble
return y, None # inference, train output
def attempt_load(weights, device=None, inplace=True, fuse=True):
# Loads an ensemble of models weights=[a,b,c] or a single model weights=[a] or weights=a
def attempt_load(weights, map_location=None, inplace=True, fuse=True):
from models.yolo import Detect, Model
# Loads an ensemble of models weights=[a,b,c] or a single model weights=[a] or weights=a
model = Ensemble()
for w in weights if isinstance(weights, list) else [weights]:
ckpt = torch.load(attempt_download(w), map_location='cpu') # load
ckpt = (ckpt.get('ema') or ckpt['model']).to(device).float() # FP32 model
ckpt = torch.load(attempt_download(w), map_location=map_location) # load
ckpt = (ckpt['ema'] or ckpt['model']).float() # FP32 model
model.append(ckpt.fuse().eval() if fuse else ckpt.eval()) # fused or un-fused model in eval mode
# Model compatibility updates
if not hasattr(ckpt, 'stride'):
ckpt.stride = torch.tensor([32.])
if hasattr(ckpt, 'names') and isinstance(ckpt.names, (list, tuple)):
ckpt.names = dict(enumerate(ckpt.names)) # convert to dict
model.append(ckpt.fuse().eval() if fuse and hasattr(ckpt, 'fuse') else ckpt.eval()) # model in eval mode
# Module compatibility updates
# Compatibility updates
for m in model.modules():
t = type(m)
if t in (nn.Hardswish, nn.LeakyReLU, nn.ReLU, nn.ReLU6, nn.SiLU, Detect, Model):
m.inplace = inplace # torch 1.7.0 compatibility
if t is Detect and not isinstance(m.anchor_grid, list):
delattr(m, 'anchor_grid')
setattr(m, 'anchor_grid', [torch.zeros(1)] * m.nl)
if t is Detect:
if not isinstance(m.anchor_grid, list): # new Detect Layer compatibility
delattr(m, 'anchor_grid')
setattr(m, 'anchor_grid', [torch.zeros(1)] * m.nl)
elif t is Conv:
m._non_persistent_buffers_set = set() # torch 1.6.0 compatibility
elif t is nn.Upsample and not hasattr(m, 'recompute_scale_factor'):
m.recompute_scale_factor = None # torch 1.11.0 compatibility
# Return model
if len(model) == 1:
return model[-1]
# Return detection ensemble
print(f'Ensemble created with {weights}\n')
for k in 'names', 'nc', 'yaml':
setattr(model, k, getattr(model[0], k))
model.stride = model[torch.argmax(torch.tensor([m.stride.max() for m in model])).int()].stride # max stride
assert all(model[0].nc == m.nc for m in model), f'Models have different class counts: {[m.nc for m in model]}'
return model
return model[-1] # return model
else:
print(f'Ensemble created with {weights}\n')
for k in ['names']:
setattr(model, k, getattr(model[-1], k))
model.stride = model[torch.argmax(torch.tensor([m.stride.max() for m in model])).int()].stride # max stride
return model # return ensemble
yolov3/models/tf.py
+237 -217
View File
@@ -1,22 +1,25 @@
# YOLOv3 🚀 by Ultralytics, GPL-3.0 license
"""
TensorFlow, Keras and TFLite versions of YOLOv3
TensorFlow, Keras and TFLite versions of
Authored by https://github.com/zldrobit in PR https://github.com/ultralytics/yolov5/pull/1127
Usage:
$ python models/tf.py --weights yolov5s.pt
$ python models/tf.py --weights yolov3.pt
Export:
$ python export.py --weights yolov5s.pt --include saved_model pb tflite tfjs
$ python path/to/export.py --weights yolov3.pt --include saved_model pb tflite tfjs
"""
import argparse
import logging
import sys
from copy import deepcopy
from pathlib import Path
from packaging import version
FILE = Path(__file__).resolve()
ROOT = FILE.parents[1] # YOLOv3 root directory
ROOT = FILE.parents[1] # root directory
if str(ROOT) not in sys.path:
sys.path.append(str(ROOT)) # add ROOT to PATH
# ROOT = ROOT.relative_to(Path.cwd()) # relative
@@ -25,15 +28,21 @@ import numpy as np
import tensorflow as tf
import torch
import torch.nn as nn
from keras import backend
from keras.engine.base_layer import Layer
from keras.engine.input_spec import InputSpec
from keras.utils import conv_utils
from tensorflow import keras
from models.common import (C3, SPP, SPPF, Bottleneck, BottleneckCSP, C3x, Concat, Conv, CrossConv, DWConv,
DWConvTranspose2d, Focus, autopad)
from models.experimental import MixConv2d, attempt_load
from models.yolo import Detect, Segment
from models.common import C3, SPP, SPPF, Bottleneck, BottleneckCSP, Concat, Conv, DWConv, Focus, autopad
from models.experimental import CrossConv, MixConv2d, attempt_load
from models.yolo import Detect
from utils.activations import SiLU
from utils.general import LOGGER, make_divisible, print_args
# isort: off
from tensorflow.python.util.tf_export import keras_export
class TFBN(keras.layers.Layer):
# TensorFlow BatchNormalization wrapper
@@ -50,14 +59,33 @@ class TFBN(keras.layers.Layer):
return self.bn(inputs)
class TFMaxPool2d(keras.layers.Layer):
# TensorFlow MAX Pooling
def __init__(self, k, s, p, w=None):
super().__init__()
self.pool = keras.layers.MaxPool2D(pool_size=k, strides=s, padding='valid')
def call(self, inputs):
return self.pool(inputs)
class TFZeroPad2d(keras.layers.Layer):
# TensorFlow MAX Pooling
def __init__(self, p, w=None):
super().__init__()
if version.parse(tf.__version__) < version.parse('2.11.0'):
self.zero_pad = ZeroPadding2D(padding=p)
else:
self.zero_pad = keras.layers.ZeroPadding2D(padding=((p[0], p[1]), (p[2], p[3])))
def call(self, inputs):
return self.zero_pad(inputs)
class TFPad(keras.layers.Layer):
# Pad inputs in spatial dimensions 1 and 2
def __init__(self, pad):
super().__init__()
if isinstance(pad, int):
self.pad = tf.constant([[0, 0], [pad, pad], [pad, pad], [0, 0]])
else: # tuple/list
self.pad = tf.constant([[0, 0], [pad[0], pad[0]], [pad[1], pad[1]], [0, 0]])
self.pad = tf.constant([[0, 0], [pad, pad], [pad, pad], [0, 0]])
def call(self, inputs):
return tf.pad(inputs, self.pad, mode='constant', constant_values=0)
@@ -69,69 +97,31 @@ class TFConv(keras.layers.Layer):
# ch_in, ch_out, weights, kernel, stride, padding, groups
super().__init__()
assert g == 1, "TF v2.2 Conv2D does not support 'groups' argument"
assert isinstance(k, int), "Convolution with multiple kernels are not allowed."
# TensorFlow convolution padding is inconsistent with PyTorch (e.g. k=3 s=2 'SAME' padding)
# see https://stackoverflow.com/questions/52975843/comparing-conv2d-with-padding-between-tensorflow-and-pytorch
conv = keras.layers.Conv2D(
filters=c2,
kernel_size=k,
strides=s,
padding='SAME' if s == 1 else 'VALID',
use_bias=not hasattr(w, 'bn'),
c2, k, s, 'SAME' if s == 1 else 'VALID', use_bias=False if hasattr(w, 'bn') else True,
kernel_initializer=keras.initializers.Constant(w.conv.weight.permute(2, 3, 1, 0).numpy()),
bias_initializer='zeros' if hasattr(w, 'bn') else keras.initializers.Constant(w.conv.bias.numpy()))
self.conv = conv if s == 1 else keras.Sequential([TFPad(autopad(k, p)), conv])
self.bn = TFBN(w.bn) if hasattr(w, 'bn') else tf.identity
self.act = activations(w.act) if act else tf.identity
# activations
if isinstance(w.act, nn.LeakyReLU):
self.act = (lambda x: keras.activations.relu(x, alpha=0.1)) if act else tf.identity
elif isinstance(w.act, nn.Hardswish):
self.act = (lambda x: x * tf.nn.relu6(x + 3) * 0.166666667) if act else tf.identity
elif isinstance(w.act, (nn.SiLU, SiLU)):
self.act = (lambda x: keras.activations.swish(x)) if act else tf.identity
else:
raise Exception(f'no matching TensorFlow activation found for {w.act}')
def call(self, inputs):
return self.act(self.bn(self.conv(inputs)))
class TFDWConv(keras.layers.Layer):
# Depthwise convolution
def __init__(self, c1, c2, k=1, s=1, p=None, act=True, w=None):
# ch_in, ch_out, weights, kernel, stride, padding, groups
super().__init__()
assert c2 % c1 == 0, f'TFDWConv() output={c2} must be a multiple of input={c1} channels'
conv = keras.layers.DepthwiseConv2D(
kernel_size=k,
depth_multiplier=c2 // c1,
strides=s,
padding='SAME' if s == 1 else 'VALID',
use_bias=not hasattr(w, 'bn'),
depthwise_initializer=keras.initializers.Constant(w.conv.weight.permute(2, 3, 1, 0).numpy()),
bias_initializer='zeros' if hasattr(w, 'bn') else keras.initializers.Constant(w.conv.bias.numpy()))
self.conv = conv if s == 1 else keras.Sequential([TFPad(autopad(k, p)), conv])
self.bn = TFBN(w.bn) if hasattr(w, 'bn') else tf.identity
self.act = activations(w.act) if act else tf.identity
def call(self, inputs):
return self.act(self.bn(self.conv(inputs)))
class TFDWConvTranspose2d(keras.layers.Layer):
# Depthwise ConvTranspose2d
def __init__(self, c1, c2, k=1, s=1, p1=0, p2=0, w=None):
# ch_in, ch_out, weights, kernel, stride, padding, groups
super().__init__()
assert c1 == c2, f'TFDWConv() output={c2} must be equal to input={c1} channels'
assert k == 4 and p1 == 1, 'TFDWConv() only valid for k=4 and p1=1'
weight, bias = w.weight.permute(2, 3, 1, 0).numpy(), w.bias.numpy()
self.c1 = c1
self.conv = [
keras.layers.Conv2DTranspose(filters=1,
kernel_size=k,
strides=s,
padding='VALID',
output_padding=p2,
use_bias=True,
kernel_initializer=keras.initializers.Constant(weight[..., i:i + 1]),
bias_initializer=keras.initializers.Constant(bias[i])) for i in range(c1)]
def call(self, inputs):
return tf.concat([m(x) for m, x in zip(self.conv, tf.split(inputs, self.c1, 3))], 3)[:, 1:-1, 1:-1]
class TFFocus(keras.layers.Layer):
# Focus wh information into c-space
def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True, w=None):
@@ -141,8 +131,10 @@ class TFFocus(keras.layers.Layer):
def call(self, inputs): # x(b,w,h,c) -> y(b,w/2,h/2,4c)
# inputs = inputs / 255 # normalize 0-255 to 0-1
inputs = [inputs[:, ::2, ::2, :], inputs[:, 1::2, ::2, :], inputs[:, ::2, 1::2, :], inputs[:, 1::2, 1::2, :]]
return self.conv(tf.concat(inputs, 3))
return self.conv(tf.concat([inputs[:, ::2, ::2, :],
inputs[:, 1::2, ::2, :],
inputs[:, ::2, 1::2, :],
inputs[:, 1::2, 1::2, :]], 3))
class TFBottleneck(keras.layers.Layer):
@@ -158,32 +150,15 @@ class TFBottleneck(keras.layers.Layer):
return inputs + self.cv2(self.cv1(inputs)) if self.add else self.cv2(self.cv1(inputs))
class TFCrossConv(keras.layers.Layer):
# Cross Convolution
def __init__(self, c1, c2, k=3, s=1, g=1, e=1.0, shortcut=False, w=None):
super().__init__()
c_ = int(c2 * e) # hidden channels
self.cv1 = TFConv(c1, c_, (1, k), (1, s), w=w.cv1)
self.cv2 = TFConv(c_, c2, (k, 1), (s, 1), g=g, w=w.cv2)
self.add = shortcut and c1 == c2
def call(self, inputs):
return inputs + self.cv2(self.cv1(inputs)) if self.add else self.cv2(self.cv1(inputs))
class TFConv2d(keras.layers.Layer):
# Substitution for PyTorch nn.Conv2D
def __init__(self, c1, c2, k, s=1, g=1, bias=True, w=None):
super().__init__()
assert g == 1, "TF v2.2 Conv2D does not support 'groups' argument"
self.conv = keras.layers.Conv2D(filters=c2,
kernel_size=k,
strides=s,
padding='VALID',
use_bias=bias,
kernel_initializer=keras.initializers.Constant(
w.weight.permute(2, 3, 1, 0).numpy()),
bias_initializer=keras.initializers.Constant(w.bias.numpy()) if bias else None)
self.conv = keras.layers.Conv2D(
c2, k, s, 'VALID', use_bias=bias,
kernel_initializer=keras.initializers.Constant(w.weight.permute(2, 3, 1, 0).numpy()),
bias_initializer=keras.initializers.Constant(w.bias.numpy()) if bias else None, )
def call(self, inputs):
return self.conv(inputs)
@@ -200,7 +175,7 @@ class TFBottleneckCSP(keras.layers.Layer):
self.cv3 = TFConv2d(c_, c_, 1, 1, bias=False, w=w.cv3)
self.cv4 = TFConv(2 * c_, c2, 1, 1, w=w.cv4)
self.bn = TFBN(w.bn)
self.act = lambda x: keras.activations.swish(x)
self.act = lambda x: keras.activations.relu(x, alpha=0.1)
self.m = keras.Sequential([TFBottleneck(c_, c_, shortcut, g, e=1.0, w=w.m[j]) for j in range(n)])
def call(self, inputs):
@@ -224,22 +199,6 @@ class TFC3(keras.layers.Layer):
return self.cv3(tf.concat((self.m(self.cv1(inputs)), self.cv2(inputs)), axis=3))
class TFC3x(keras.layers.Layer):
# 3 module with cross-convolutions
def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5, w=None):
# ch_in, ch_out, number, shortcut, groups, expansion
super().__init__()
c_ = int(c2 * e) # hidden channels
self.cv1 = TFConv(c1, c_, 1, 1, w=w.cv1)
self.cv2 = TFConv(c1, c_, 1, 1, w=w.cv2)
self.cv3 = TFConv(2 * c_, c2, 1, 1, w=w.cv3)
self.m = keras.Sequential([
TFCrossConv(c_, c_, k=3, s=1, g=g, e=1.0, shortcut=shortcut, w=w.m[j]) for j in range(n)])
def call(self, inputs):
return self.cv3(tf.concat((self.m(self.cv1(inputs)), self.cv2(inputs)), axis=3))
class TFSPP(keras.layers.Layer):
# Spatial pyramid pooling layer used in YOLOv3-SPP
def __init__(self, c1, c2, k=(5, 9, 13), w=None):
@@ -271,7 +230,6 @@ class TFSPPF(keras.layers.Layer):
class TFDetect(keras.layers.Layer):
# TF YOLOv3 Detect layer
def __init__(self, nc=80, anchors=(), ch=(), imgsz=(640, 640), w=None): # detection layer
super().__init__()
self.stride = tf.convert_to_tensor(w.stride.numpy(), dtype=tf.float32)
@@ -281,7 +239,8 @@ class TFDetect(keras.layers.Layer):
self.na = len(anchors[0]) // 2 # number of anchors
self.grid = [tf.zeros(1)] * self.nl # init grid
self.anchors = tf.convert_to_tensor(w.anchors.numpy(), dtype=tf.float32)
self.anchor_grid = tf.reshape(self.anchors * tf.reshape(self.stride, [self.nl, 1, 1]), [self.nl, 1, -1, 1, 2])
self.anchor_grid = tf.reshape(self.anchors * tf.reshape(self.stride, [self.nl, 1, 1]),
[self.nl, 1, -1, 1, 2])
self.m = [TFConv2d(x, self.no * self.na, 1, w=w.m[i]) for i, x in enumerate(ch)]
self.training = False # set to False after building model
self.imgsz = imgsz
@@ -296,21 +255,19 @@ class TFDetect(keras.layers.Layer):
x.append(self.m[i](inputs[i]))
# x(bs,20,20,255) to x(bs,3,20,20,85)
ny, nx = self.imgsz[0] // self.stride[i], self.imgsz[1] // self.stride[i]
x[i] = tf.reshape(x[i], [-1, ny * nx, self.na, self.no])
x[i] = tf.transpose(tf.reshape(x[i], [-1, ny * nx, self.na, self.no]), [0, 2, 1, 3])
if not self.training: # inference
y = x[i]
grid = tf.transpose(self.grid[i], [0, 2, 1, 3]) - 0.5
anchor_grid = tf.transpose(self.anchor_grid[i], [0, 2, 1, 3]) * 4
xy = (tf.sigmoid(y[..., 0:2]) * 2 + grid) * self.stride[i] # xy
wh = tf.sigmoid(y[..., 2:4]) ** 2 * anchor_grid
y = tf.sigmoid(x[i])
xy = (y[..., 0:2] * 2 - 0.5 + self.grid[i]) * self.stride[i] # xy
wh = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i]
# Normalize xywh to 0-1 to reduce calibration error
xy /= tf.constant([[self.imgsz[1], self.imgsz[0]]], dtype=tf.float32)
wh /= tf.constant([[self.imgsz[1], self.imgsz[0]]], dtype=tf.float32)
y = tf.concat([xy, wh, tf.sigmoid(y[..., 4:5 + self.nc]), y[..., 5 + self.nc:]], -1)
z.append(tf.reshape(y, [-1, self.na * ny * nx, self.no]))
y = tf.concat([xy, wh, y[..., 4:]], -1)
z.append(tf.reshape(y, [-1, 3 * ny * nx, self.no]))
return tf.transpose(x, [0, 2, 1, 3]) if self.training else (tf.concat(z, 1),)
return x if self.training else (tf.concat(z, 1), x)
@staticmethod
def _make_grid(nx=20, ny=20):
@@ -320,44 +277,11 @@ class TFDetect(keras.layers.Layer):
return tf.cast(tf.reshape(tf.stack([xv, yv], 2), [1, 1, ny * nx, 2]), dtype=tf.float32)
class TFSegment(TFDetect):
# YOLOv3 Segment head for segmentation models
def __init__(self, nc=80, anchors=(), nm=32, npr=256, ch=(), imgsz=(640, 640), w=None):
super().__init__(nc, anchors, ch, imgsz, w)
self.nm = nm # number of masks
self.npr = npr # number of protos
self.no = 5 + nc + self.nm # number of outputs per anchor
self.m = [TFConv2d(x, self.no * self.na, 1, w=w.m[i]) for i, x in enumerate(ch)] # output conv
self.proto = TFProto(ch[0], self.npr, self.nm, w=w.proto) # protos
self.detect = TFDetect.call
def call(self, x):
p = self.proto(x[0])
# p = TFUpsample(None, scale_factor=4, mode='nearest')(self.proto(x[0])) # (optional) full-size protos
p = tf.transpose(p, [0, 3, 1, 2]) # from shape(1,160,160,32) to shape(1,32,160,160)
x = self.detect(self, x)
return (x, p) if self.training else (x[0], p)
class TFProto(keras.layers.Layer):
def __init__(self, c1, c_=256, c2=32, w=None):
super().__init__()
self.cv1 = TFConv(c1, c_, k=3, w=w.cv1)
self.upsample = TFUpsample(None, scale_factor=2, mode='nearest')
self.cv2 = TFConv(c_, c_, k=3, w=w.cv2)
self.cv3 = TFConv(c_, c2, w=w.cv3)
def call(self, inputs):
return self.cv3(self.cv2(self.upsample(self.cv1(inputs))))
class TFUpsample(keras.layers.Layer):
# TF version of torch.nn.Upsample()
def __init__(self, size, scale_factor, mode, w=None): # warning: all arguments needed including 'w'
super().__init__()
assert scale_factor % 2 == 0, 'scale_factor must be multiple of 2'
self.upsample = lambda x: tf.image.resize(x, (x.shape[1] * scale_factor, x.shape[2] * scale_factor), mode)
assert scale_factor == 2, "scale_factor must be 2"
self.upsample = lambda x: tf.image.resize(x, (x.shape[1] * 2, x.shape[2] * 2), method=mode)
# self.upsample = keras.layers.UpSampling2D(size=scale_factor, interpolation=mode)
# with default arguments: align_corners=False, half_pixel_centers=False
# self.upsample = lambda x: tf.raw_ops.ResizeNearestNeighbor(images=x,
@@ -368,10 +292,9 @@ class TFUpsample(keras.layers.Layer):
class TFConcat(keras.layers.Layer):
# TF version of torch.concat()
def __init__(self, dimension=1, w=None):
super().__init__()
assert dimension == 1, 'convert only NCHW to NHWC concat'
assert dimension == 1, "convert only NCHW to NHWC concat"
self.d = 3
def call(self, inputs):
@@ -395,26 +318,22 @@ def parse_model(d, ch, model, imgsz): # model_dict, input_channels(3)
pass
n = max(round(n * gd), 1) if n > 1 else n # depth gain
if m in [
nn.Conv2d, Conv, DWConv, DWConvTranspose2d, Bottleneck, SPP, SPPF, MixConv2d, Focus, CrossConv,
BottleneckCSP, C3, C3x]:
if m in [nn.Conv2d, Conv, Bottleneck, SPP, SPPF, DWConv, MixConv2d, Focus, CrossConv, BottleneckCSP, C3]:
c1, c2 = ch[f], args[0]
c2 = make_divisible(c2 * gw, 8) if c2 != no else c2
args = [c1, c2, *args[1:]]
if m in [BottleneckCSP, C3, C3x]:
if m in [BottleneckCSP, C3]:
args.insert(2, n)
n = 1
elif m is nn.BatchNorm2d:
args = [ch[f]]
elif m is Concat:
c2 = sum(ch[-1 if x == -1 else x + 1] for x in f)
elif m in [Detect, Segment]:
elif m is Detect:
args.append([ch[x + 1] for x in f])
if isinstance(args[1], int): # number of anchors
args[1] = [list(range(args[1] * 2))] * len(f)
if m is Segment:
args[3] = make_divisible(args[3] * gw, 8)
args.append(imgsz)
else:
c2 = ch[f]
@@ -435,8 +354,7 @@ def parse_model(d, ch, model, imgsz): # model_dict, input_channels(3)
class TFModel:
# TF YOLOv3 model
def __init__(self, cfg='yolov5s.yaml', ch=3, nc=None, model=None, imgsz=(640, 640)): # model, channels, classes
def __init__(self, cfg='yolov3.yaml', ch=3, nc=None, model=None, imgsz=(640, 640)): # model, channels, classes
super().__init__()
if isinstance(cfg, dict):
self.yaml = cfg # model dict
@@ -452,17 +370,11 @@ class TFModel:
self.yaml['nc'] = nc # override yaml value
self.model, self.savelist = parse_model(deepcopy(self.yaml), ch=[ch], model=model, imgsz=imgsz)
def predict(self,
inputs,
tf_nms=False,
agnostic_nms=False,
topk_per_class=100,
topk_all=100,
iou_thres=0.45,
def predict(self, inputs, tf_nms=False, agnostic_nms=False, topk_per_class=100, topk_all=100, iou_thres=0.45,
conf_thres=0.25):
y = [] # outputs
x = inputs
for m in self.model.layers:
for i, m in enumerate(self.model.layers):
if m.f != -1: # if not from previous layer
x = y[m.f] if isinstance(m.f, int) else [x if j == -1 else y[j] for j in m.f] # from earlier layers
@@ -477,18 +389,15 @@ class TFModel:
scores = probs * classes
if agnostic_nms:
nms = AgnosticNMS()((boxes, classes, scores), topk_all, iou_thres, conf_thres)
return nms, x[1]
else:
boxes = tf.expand_dims(boxes, 2)
nms = tf.image.combined_non_max_suppression(boxes,
scores,
topk_per_class,
topk_all,
iou_thres,
conf_thres,
clip_boxes=False)
return (nms,)
return x # output [1,6300,85] = [xywh, conf, class0, class1, ...]
# x = x[0] # [x(1,6300,85), ...] to x(6300,85)
nms = tf.image.combined_non_max_suppression(
boxes, scores, topk_per_class, topk_all, iou_thres, conf_thres, clip_boxes=False)
return nms, x[1]
return x[0] # output only first tensor [1,6300,85] = [xywh, conf, class0, class1, ...]
# x = x[0][0] # [x(1,6300,85), ...] to x(6300,85)
# xywh = x[..., :4] # x(6300,4) boxes
# conf = x[..., 4:5] # x(6300,1) confidences
# cls = tf.reshape(tf.cast(tf.argmax(x[..., 5:], axis=1), tf.float32), (-1, 1)) # x(6300,1) classes
@@ -505,8 +414,7 @@ class AgnosticNMS(keras.layers.Layer):
# TF Agnostic NMS
def call(self, input, topk_all, iou_thres, conf_thres):
# wrap map_fn to avoid TypeSpec related error https://stackoverflow.com/a/65809989/3036450
return tf.map_fn(lambda x: self._nms(x, topk_all, iou_thres, conf_thres),
input,
return tf.map_fn(lambda x: self._nms(x, topk_all, iou_thres, conf_thres), input,
fn_output_signature=(tf.float32, tf.float32, tf.float32, tf.int32),
name='agnostic_nms')
@@ -515,69 +423,50 @@ class AgnosticNMS(keras.layers.Layer):
boxes, classes, scores = x
class_inds = tf.cast(tf.argmax(classes, axis=-1), tf.float32)
scores_inp = tf.reduce_max(scores, -1)
selected_inds = tf.image.non_max_suppression(boxes,
scores_inp,
max_output_size=topk_all,
iou_threshold=iou_thres,
score_threshold=conf_thres)
selected_inds = tf.image.non_max_suppression(
boxes, scores_inp, max_output_size=topk_all, iou_threshold=iou_thres, score_threshold=conf_thres)
selected_boxes = tf.gather(boxes, selected_inds)
padded_boxes = tf.pad(selected_boxes,
paddings=[[0, topk_all - tf.shape(selected_boxes)[0]], [0, 0]],
mode='CONSTANT',
constant_values=0.0)
mode="CONSTANT", constant_values=0.0)
selected_scores = tf.gather(scores_inp, selected_inds)
padded_scores = tf.pad(selected_scores,
paddings=[[0, topk_all - tf.shape(selected_boxes)[0]]],
mode='CONSTANT',
constant_values=-1.0)
mode="CONSTANT", constant_values=-1.0)
selected_classes = tf.gather(class_inds, selected_inds)
padded_classes = tf.pad(selected_classes,
paddings=[[0, topk_all - tf.shape(selected_boxes)[0]]],
mode='CONSTANT',
constant_values=-1.0)
mode="CONSTANT", constant_values=-1.0)
valid_detections = tf.shape(selected_inds)[0]
return padded_boxes, padded_scores, padded_classes, valid_detections
def activations(act=nn.SiLU):
# Returns TF activation from input PyTorch activation
if isinstance(act, nn.LeakyReLU):
return lambda x: keras.activations.relu(x, alpha=0.1)
elif isinstance(act, nn.Hardswish):
return lambda x: x * tf.nn.relu6(x + 3) * 0.166666667
elif isinstance(act, (nn.SiLU, SiLU)):
return lambda x: keras.activations.swish(x)
else:
raise Exception(f'no matching TensorFlow activation found for PyTorch activation {act}')
def representative_dataset_gen(dataset, ncalib=100):
# Representative dataset generator for use with converter.representative_dataset, returns a generator of np arrays
for n, (path, img, im0s, vid_cap, string) in enumerate(dataset):
im = np.transpose(img, [1, 2, 0])
im = np.expand_dims(im, axis=0).astype(np.float32)
im /= 255
yield [im]
input = np.transpose(img, [1, 2, 0])
input = np.expand_dims(input, axis=0).astype(np.float32)
input /= 255
yield [input]
if n >= ncalib:
break
def run(
weights=ROOT / 'yolov5s.pt', # weights path
def run(weights=ROOT / 'yolov3.pt', # weights path
imgsz=(640, 640), # inference size h,w
batch_size=1, # batch size
dynamic=False, # dynamic batch size
):
):
# PyTorch model
im = torch.zeros((batch_size, 3, *imgsz)) # BCHW image
model = attempt_load(weights, device=torch.device('cpu'), inplace=True, fuse=False)
_ = model(im) # inference
model = attempt_load(weights, map_location=torch.device('cpu'), inplace=True, fuse=False)
y = model(im) # inference
model.info()
# TensorFlow model
im = tf.zeros((batch_size, *imgsz, 3)) # BHWC image
tf_model = TFModel(cfg=model.yaml, model=model, nc=model.nc, imgsz=imgsz)
_ = tf_model.predict(im) # inference
y = tf_model.predict(im) # inference
# Keras model
im = keras.Input(shape=(*imgsz, 3), batch_size=None if dynamic else batch_size)
@@ -587,15 +476,146 @@ def run(
LOGGER.info('PyTorch, TensorFlow and Keras models successfully verified.\nUse export.py for TF model export.')
@keras_export("keras.layers.ZeroPadding2D")
class ZeroPadding2D(Layer):
"""Zero-padding layer for 2D input (e.g. picture).
This layer can add rows and columns of zeros
at the top, bottom, left and right side of an image tensor.
Examples:
>>> input_shape = (1, 1, 2, 2)
>>> x = np.arange(np.prod(input_shape)).reshape(input_shape)
>>> print(x)
[[[[0 1]
[2 3]]]]
>>> y = tf.keras.layers.ZeroPadding2D(padding=1)(x)
>>> print(y)
tf.Tensor(
[[[[0 0]
[0 0]
[0 0]
[0 0]]
[[0 0]
[0 1]
[2 3]
[0 0]]
[[0 0]
[0 0]
[0 0]
[0 0]]]], shape=(1, 3, 4, 2), dtype=int64)
Args:
padding: Int, or tuple of 2 ints, or tuple of 2 tuples of 2 ints.
- If int: the same symmetric padding
is applied to height and width.
- If tuple of 2 ints:
interpreted as two different
symmetric padding values for height and width:
`(symmetric_height_pad, symmetric_width_pad)`.
- If tuple of 2 tuples of 2 ints:
interpreted as
`((top_pad, bottom_pad), (left_pad, right_pad))`
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch_size, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch_size, channels, height, width)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be "channels_last".
Input shape:
4D tensor with shape:
- If `data_format` is `"channels_last"`:
`(batch_size, rows, cols, channels)`
- If `data_format` is `"channels_first"`:
`(batch_size, channels, rows, cols)`
Output shape:
4D tensor with shape:
- If `data_format` is `"channels_last"`:
`(batch_size, padded_rows, padded_cols, channels)`
- If `data_format` is `"channels_first"`:
`(batch_size, channels, padded_rows, padded_cols)`
"""
def __init__(self, padding=(1, 1), data_format=None, **kwargs):
super().__init__(**kwargs)
self.data_format = conv_utils.normalize_data_format(data_format)
if isinstance(padding, int):
self.padding = ((padding, padding), (padding, padding))
elif hasattr(padding, "__len__"):
if len(padding) == 4:
padding = ((padding[0], padding[1]), (padding[2], padding[3]))
if len(padding) != 2:
raise ValueError(
f"`padding` should have two elements. Received: {padding}."
)
height_padding = conv_utils.normalize_tuple(
padding[0], 2, "1st entry of padding", allow_zero=True
)
width_padding = conv_utils.normalize_tuple(
padding[1], 2, "2nd entry of padding", allow_zero=True
)
self.padding = (height_padding, width_padding)
else:
raise ValueError(
"`padding` should be either an int, "
"a tuple of 2 ints "
"(symmetric_height_pad, symmetric_width_pad), "
"or a tuple of 2 tuples of 2 ints "
"((top_pad, bottom_pad), (left_pad, right_pad)). "
f"Received: {padding}."
)
self.input_spec = InputSpec(ndim=4)
def compute_output_shape(self, input_shape):
input_shape = tf.TensorShape(input_shape).as_list()
if self.data_format == "channels_first":
if input_shape[2] is not None:
rows = input_shape[2] + self.padding[0][0] + self.padding[0][1]
else:
rows = None
if input_shape[3] is not None:
cols = input_shape[3] + self.padding[1][0] + self.padding[1][1]
else:
cols = None
return tf.TensorShape([input_shape[0], input_shape[1], rows, cols])
elif self.data_format == "channels_last":
if input_shape[1] is not None:
rows = input_shape[1] + self.padding[0][0] + self.padding[0][1]
else:
rows = None
if input_shape[2] is not None:
cols = input_shape[2] + self.padding[1][0] + self.padding[1][1]
else:
cols = None
return tf.TensorShape([input_shape[0], rows, cols, input_shape[3]])
def call(self, inputs):
return backend.spatial_2d_padding(
inputs, padding=self.padding, data_format=self.data_format
)
def get_config(self):
config = {"padding": self.padding, "data_format": self.data_format}
base_config = super().get_config()
return dict(list(base_config.items()) + list(config.items()))
def parse_opt():
parser = argparse.ArgumentParser()
parser.add_argument('--weights', type=str, default=ROOT / 'yolov5s.pt', help='weights path')
parser.add_argument('--weights', type=str, default=ROOT / 'yolov3.pt', help='weights path')
parser.add_argument('--imgsz', '--img', '--img-size', nargs='+', type=int, default=[640], help='inference size h,w')
parser.add_argument('--batch-size', type=int, default=1, help='batch size')
parser.add_argument('--dynamic', action='store_true', help='dynamic batch size')
opt = parser.parse_args()
opt.imgsz *= 2 if len(opt.imgsz) == 1 else 1 # expand
print_args(vars(opt))
print_args(FILE.stem, opt)
return opt
@@ -603,6 +623,6 @@ def main(opt):
run(**vars(opt))
if __name__ == '__main__':
if __name__ == "__main__":
opt = parse_opt()
main(opt)
yolov3/models/yolo.py
+117 -171
View File
@@ -3,29 +3,26 @@
YOLO-specific modules
Usage:
$ python models/yolo.py --cfg yolov5s.yaml
$ python path/to/models/yolo.py --cfg yolov3.yaml
"""
import argparse
import os
import platform
import sys
from copy import deepcopy
from pathlib import Path
FILE = Path(__file__).resolve()
ROOT = FILE.parents[1] # YOLOv3 root directory
ROOT = FILE.parents[1] # root directory
if str(ROOT) not in sys.path:
sys.path.append(str(ROOT)) # add ROOT to PATH
if platform.system() != 'Windows':
ROOT = Path(os.path.relpath(ROOT, Path.cwd())) # relative
# ROOT = ROOT.relative_to(Path.cwd()) # relative
from models.common import *
from models.experimental import *
from utils.autoanchor import check_anchor_order
from utils.general import LOGGER, check_version, check_yaml, make_divisible, print_args
from utils.plots import feature_visualization
from utils.torch_utils import (fuse_conv_and_bn, initialize_weights, model_info, profile, scale_img, select_device,
from utils.torch_utils import (copy_attr, fuse_conv_and_bn, initialize_weights, model_info, scale_img, select_device,
time_sync)
try:
@@ -35,10 +32,8 @@ except ImportError:
class Detect(nn.Module):
# YOLOv3 Detect head for detection models
stride = None # strides computed during build
dynamic = False # force grid reconstruction
export = False # export mode
onnx_dynamic = False # ONNX export parameter
def __init__(self, nc=80, anchors=(), ch=(), inplace=True): # detection layer
super().__init__()
@@ -46,11 +41,11 @@ class Detect(nn.Module):
self.no = nc + 5 # number of outputs per anchor
self.nl = len(anchors) # number of detection layers
self.na = len(anchors[0]) // 2 # number of anchors
self.grid = [torch.empty(0) for _ in range(self.nl)] # init grid
self.anchor_grid = [torch.empty(0) for _ in range(self.nl)] # init anchor grid
self.grid = [torch.zeros(1)] * self.nl # init grid
self.anchor_grid = [torch.zeros(1)] * self.nl # init anchor grid
self.register_buffer('anchors', torch.tensor(anchors).float().view(self.nl, -1, 2)) # shape(nl,na,2)
self.m = nn.ModuleList(nn.Conv2d(x, self.no * self.na, 1) for x in ch) # output conv
self.inplace = inplace # use inplace ops (e.g. slice assignment)
self.inplace = inplace # use in-place ops (e.g. slice assignment)
def forward(self, x):
z = [] # inference output
@@ -60,110 +55,35 @@ class Detect(nn.Module):
x[i] = x[i].view(bs, self.na, self.no, ny, nx).permute(0, 1, 3, 4, 2).contiguous()
if not self.training: # inference
if self.dynamic or self.grid[i].shape[2:4] != x[i].shape[2:4]:
if self.onnx_dynamic or self.grid[i].shape[2:4] != x[i].shape[2:4]:
self.grid[i], self.anchor_grid[i] = self._make_grid(nx, ny, i)
if isinstance(self, Segment): # (boxes + masks)
xy, wh, conf, mask = x[i].split((2, 2, self.nc + 1, self.no - self.nc - 5), 4)
xy = (xy.sigmoid() * 2 + self.grid[i]) * self.stride[i] # xy
wh = (wh.sigmoid() * 2) ** 2 * self.anchor_grid[i] # wh
y = torch.cat((xy, wh, conf.sigmoid(), mask), 4)
else: # Detect (boxes only)
xy, wh, conf = x[i].sigmoid().split((2, 2, self.nc + 1), 4)
xy = (xy * 2 + self.grid[i]) * self.stride[i] # xy
wh = (wh * 2) ** 2 * self.anchor_grid[i] # wh
y = torch.cat((xy, wh, conf), 4)
z.append(y.view(bs, self.na * nx * ny, self.no))
y = x[i].sigmoid()
if self.inplace:
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
else: # for on AWS Inferentia https://github.com/ultralytics/yolov5/pull/2953
xy = (y[..., 0:2] * 2 - 0.5 + self.grid[i]) * self.stride[i] # xy
wh = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i] # wh
y = torch.cat((xy, wh, y[..., 4:]), -1)
z.append(y.view(bs, -1, self.no))
return x if self.training else (torch.cat(z, 1),) if self.export else (torch.cat(z, 1), x)
return x if self.training else (torch.cat(z, 1), x)
def _make_grid(self, nx=20, ny=20, i=0, torch_1_10=check_version(torch.__version__, '1.10.0')):
def _make_grid(self, nx=20, ny=20, i=0):
d = self.anchors[i].device
t = self.anchors[i].dtype
shape = 1, self.na, ny, nx, 2 # grid shape
y, x = torch.arange(ny, device=d, dtype=t), torch.arange(nx, device=d, dtype=t)
yv, xv = torch.meshgrid(y, x, indexing='ij') if torch_1_10 else torch.meshgrid(y, x) # torch>=0.7 compatibility
grid = torch.stack((xv, yv), 2).expand(shape) - 0.5 # add grid offset, i.e. y = 2.0 * x - 0.5
anchor_grid = (self.anchors[i] * self.stride[i]).view((1, self.na, 1, 1, 2)).expand(shape)
if check_version(torch.__version__, '1.10.0'): # torch>=1.10.0 meshgrid workaround for torch>=0.7 compatibility
yv, xv = torch.meshgrid([torch.arange(ny).to(d), torch.arange(nx).to(d)], indexing='ij')
else:
yv, xv = torch.meshgrid([torch.arange(ny).to(d), torch.arange(nx).to(d)])
grid = torch.stack((xv, yv), 2).expand((1, self.na, ny, nx, 2)).float()
anchor_grid = (self.anchors[i].clone() * self.stride[i]) \
.view((1, self.na, 1, 1, 2)).expand((1, self.na, ny, nx, 2)).float()
return grid, anchor_grid
class Segment(Detect):
# YOLOv3 Segment head for segmentation models
def __init__(self, nc=80, anchors=(), nm=32, npr=256, ch=(), inplace=True):
super().__init__(nc, anchors, ch, inplace)
self.nm = nm # number of masks
self.npr = npr # number of protos
self.no = 5 + nc + self.nm # number of outputs per anchor
self.m = nn.ModuleList(nn.Conv2d(x, self.no * self.na, 1) for x in ch) # output conv
self.proto = Proto(ch[0], self.npr, self.nm) # protos
self.detect = Detect.forward
def forward(self, x):
p = self.proto(x[0])
x = self.detect(self, x)
return (x, p) if self.training else (x[0], p) if self.export else (x[0], p, x[1])
class BaseModel(nn.Module):
# YOLOv3 base model
def forward(self, x, profile=False, visualize=False):
return self._forward_once(x, profile, visualize) # single-scale inference, train
def _forward_once(self, x, profile=False, visualize=False):
y, dt = [], [] # outputs
for m in self.model:
if m.f != -1: # if not from previous layer
x = y[m.f] if isinstance(m.f, int) else [x if j == -1 else y[j] for j in m.f] # from earlier layers
if profile:
self._profile_one_layer(m, x, dt)
x = m(x) # run
y.append(x if m.i in self.save else None) # save output
if visualize:
feature_visualization(x, m.type, m.i, save_dir=visualize)
return x
def _profile_one_layer(self, m, x, dt):
c = m == self.model[-1] # is final layer, copy input as inplace fix
o = thop.profile(m, inputs=(x.copy() if c else x,), verbose=False)[0] / 1E9 * 2 if thop else 0 # FLOPs
t = time_sync()
for _ in range(10):
m(x.copy() if c else x)
dt.append((time_sync() - t) * 100)
if m == self.model[0]:
LOGGER.info(f"{'time (ms)':>10s} {'GFLOPs':>10s} {'params':>10s} module")
LOGGER.info(f'{dt[-1]:10.2f} {o:10.2f} {m.np:10.0f} {m.type}')
if c:
LOGGER.info(f"{sum(dt):10.2f} {'-':>10s} {'-':>10s} Total")
def fuse(self): # fuse model Conv2d() + BatchNorm2d() layers
LOGGER.info('Fusing layers... ')
for m in self.model.modules():
if isinstance(m, (Conv, DWConv)) and hasattr(m, 'bn'):
m.conv = fuse_conv_and_bn(m.conv, m.bn) # update conv
delattr(m, 'bn') # remove batchnorm
m.forward = m.forward_fuse # update forward
self.info()
return self
def info(self, verbose=False, img_size=640): # print model information
model_info(self, verbose, img_size)
def _apply(self, fn):
# Apply to(), cpu(), cuda(), half() to model tensors that are not parameters or registered buffers
self = super()._apply(fn)
m = self.model[-1] # Detect()
if isinstance(m, (Detect, Segment)):
m.stride = fn(m.stride)
m.grid = list(map(fn, m.grid))
if isinstance(m.anchor_grid, list):
m.anchor_grid = list(map(fn, m.anchor_grid))
return self
class DetectionModel(BaseModel):
# YOLOv3 detection model
def __init__(self, cfg='yolov5s.yaml', ch=3, nc=None, anchors=None): # model, input channels, number of classes
class Model(nn.Module):
def __init__(self, cfg='yolov3.yaml', ch=3, nc=None, anchors=None): # model, input channels, number of classes
super().__init__()
if isinstance(cfg, dict):
self.yaml = cfg # model dict
@@ -187,13 +107,12 @@ class DetectionModel(BaseModel):
# Build strides, anchors
m = self.model[-1] # Detect()
if isinstance(m, (Detect, Segment)):
if isinstance(m, Detect):
s = 256 # 2x min stride
m.inplace = self.inplace
forward = lambda x: self.forward(x)[0] if isinstance(m, Segment) else self.forward(x)
m.stride = torch.tensor([s / x.shape[-2] for x in forward(torch.zeros(1, ch, s, s))]) # forward
check_anchor_order(m)
m.stride = torch.tensor([s / x.shape[-2] for x in self.forward(torch.zeros(1, ch, s, s))]) # forward
m.anchors /= m.stride.view(-1, 1, 1)
check_anchor_order(m)
self.stride = m.stride
self._initialize_biases() # only run once
@@ -221,6 +140,19 @@ class DetectionModel(BaseModel):
y = self._clip_augmented(y) # clip augmented tails
return torch.cat(y, 1), None # augmented inference, train
def _forward_once(self, x, profile=False, visualize=False):
y, dt = [], [] # outputs
for m in self.model:
if m.f != -1: # if not from previous layer
x = y[m.f] if isinstance(m.f, int) else [x if j == -1 else y[j] for j in m.f] # from earlier layers
if profile:
self._profile_one_layer(m, x, dt)
x = m(x) # run
y.append(x if m.i in self.save else None) # save output
if visualize:
feature_visualization(x, m.type, m.i, save_dir=visualize)
return x
def _descale_pred(self, p, flips, scale, img_size):
# de-scale predictions following augmented inference (inverse operation)
if self.inplace:
@@ -249,6 +181,19 @@ class DetectionModel(BaseModel):
y[-1] = y[-1][:, i:] # small
return y
def _profile_one_layer(self, m, x, dt):
c = isinstance(m, Detect) # is final layer, copy input as inplace fix
o = thop.profile(m, inputs=(x.copy() if c else x,), verbose=False)[0] / 1E9 * 2 if thop else 0 # FLOPs
t = time_sync()
for _ in range(10):
m(x.copy() if c else x)
dt.append((time_sync() - t) * 100)
if m == self.model[0]:
LOGGER.info(f"{'time (ms)':>10s} {'GFLOPs':>10s} {'params':>10s} {'module'}")
LOGGER.info(f'{dt[-1]:10.2f} {o:10.2f} {m.np:10.0f} {m.type}')
if c:
LOGGER.info(f"{sum(dt):10.2f} {'-':>10s} {'-':>10s} Total")
def _initialize_biases(self, cf=None): # initialize biases into Detect(), cf is class frequency
# https://arxiv.org/abs/1708.02002 section 3.3
# cf = torch.bincount(torch.tensor(np.concatenate(dataset.labels, 0)[:, 0]).long(), minlength=nc) + 1.
@@ -256,52 +201,55 @@ class DetectionModel(BaseModel):
for mi, s in zip(m.m, m.stride): # from
b = mi.bias.view(m.na, -1) # conv.bias(255) to (3,85)
b.data[:, 4] += math.log(8 / (640 / s) ** 2) # obj (8 objects per 640 image)
b.data[:, 5:5 + m.nc] += math.log(0.6 / (m.nc - 0.99999)) if cf is None else torch.log(cf / cf.sum()) # cls
b.data[:, 5:] += math.log(0.6 / (m.nc - 0.999999)) if cf is None else torch.log(cf / cf.sum()) # cls
mi.bias = torch.nn.Parameter(b.view(-1), requires_grad=True)
def _print_biases(self):
m = self.model[-1] # Detect() module
for mi in m.m: # from
b = mi.bias.detach().view(m.na, -1).T # conv.bias(255) to (3,85)
LOGGER.info(
('%6g Conv2d.bias:' + '%10.3g' * 6) % (mi.weight.shape[1], *b[:5].mean(1).tolist(), b[5:].mean()))
Model = DetectionModel # retain 'Model' class for backwards compatibility
# def _print_weights(self):
# for m in self.model.modules():
# if type(m) is Bottleneck:
# LOGGER.info('%10.3g' % (m.w.detach().sigmoid() * 2)) # shortcut weights
def fuse(self): # fuse model Conv2d() + BatchNorm2d() layers
LOGGER.info('Fusing layers... ')
for m in self.model.modules():
if isinstance(m, (Conv, DWConv)) and hasattr(m, 'bn'):
m.conv = fuse_conv_and_bn(m.conv, m.bn) # update conv
delattr(m, 'bn') # remove batchnorm
m.forward = m.forward_fuse # update forward
self.info()
return self
class SegmentationModel(DetectionModel):
# segmentation model
def __init__(self, cfg='yolov5s-seg.yaml', ch=3, nc=None, anchors=None):
super().__init__(cfg, ch, nc, anchors)
def autoshape(self): # add AutoShape module
LOGGER.info('Adding AutoShape... ')
m = AutoShape(self) # wrap model
copy_attr(m, self, include=('yaml', 'nc', 'hyp', 'names', 'stride'), exclude=()) # copy attributes
return m
def info(self, verbose=False, img_size=640): # print model information
model_info(self, verbose, img_size)
class ClassificationModel(BaseModel):
# classification model
def __init__(self, cfg=None, model=None, nc=1000, cutoff=10): # yaml, model, number of classes, cutoff index
super().__init__()
self._from_detection_model(model, nc, cutoff) if model is not None else self._from_yaml(cfg)
def _from_detection_model(self, model, nc=1000, cutoff=10):
# Create a classification model from a detection model
if isinstance(model, DetectMultiBackend):
model = model.model # unwrap DetectMultiBackend
model.model = model.model[:cutoff] # backbone
m = model.model[-1] # last layer
ch = m.conv.in_channels if hasattr(m, 'conv') else m.cv1.conv.in_channels # ch into module
c = Classify(ch, nc) # Classify()
c.i, c.f, c.type = m.i, m.f, 'models.common.Classify' # index, from, type
model.model[-1] = c # replace
self.model = model.model
self.stride = model.stride
self.save = []
self.nc = nc
def _from_yaml(self, cfg):
# Create a classification model from a *.yaml file
self.model = None
def _apply(self, fn):
# Apply to(), cpu(), cuda(), half() to model tensors that are not parameters or registered buffers
self = super()._apply(fn)
m = self.model[-1] # Detect()
if isinstance(m, Detect):
m.stride = fn(m.stride)
m.grid = list(map(fn, m.grid))
if isinstance(m.anchor_grid, list):
m.anchor_grid = list(map(fn, m.anchor_grid))
return self
def parse_model(d, ch): # model_dict, input_channels(3)
# Parse a model.yaml dictionary
LOGGER.info(f"\n{'':>3}{'from':>18}{'n':>3}{'params':>10} {'module':<40}{'arguments':<30}")
anchors, nc, gd, gw, act = d['anchors'], d['nc'], d['depth_multiple'], d['width_multiple'], d.get('activation')
if act:
Conv.default_act = eval(act) # redefine default activation, i.e. Conv.default_act = nn.SiLU()
LOGGER.info(f"{colorstr('activation:')} {act}") # print
anchors, nc, gd, gw = d['anchors'], d['nc'], d['depth_multiple'], d['width_multiple']
na = (len(anchors[0]) // 2) if isinstance(anchors, list) else anchors # number of anchors
no = na * (nc + 5) # number of outputs = anchors * (classes + 5)
@@ -309,32 +257,30 @@ def parse_model(d, ch): # model_dict, input_channels(3)
for i, (f, n, m, args) in enumerate(d['backbone'] + d['head']): # from, number, module, args
m = eval(m) if isinstance(m, str) else m # eval strings
for j, a in enumerate(args):
with contextlib.suppress(NameError):
try:
args[j] = eval(a) if isinstance(a, str) else a # eval strings
except NameError:
pass
n = n_ = max(round(n * gd), 1) if n > 1 else n # depth gain
if m in {
Conv, GhostConv, Bottleneck, GhostBottleneck, SPP, SPPF, DWConv, MixConv2d, Focus, CrossConv,
BottleneckCSP, C3, C3TR, C3SPP, C3Ghost, nn.ConvTranspose2d, DWConvTranspose2d, C3x}:
if m in [Conv, GhostConv, Bottleneck, GhostBottleneck, SPP, SPPF, DWConv, MixConv2d, Focus, CrossConv,
BottleneckCSP, C3, C3TR, C3SPP, C3Ghost]:
c1, c2 = ch[f], args[0]
if c2 != no: # if not output
c2 = make_divisible(c2 * gw, 8)
args = [c1, c2, *args[1:]]
if m in {BottleneckCSP, C3, C3TR, C3Ghost, C3x}:
if m in [BottleneckCSP, C3, C3TR, C3Ghost]:
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] for x in f)
# TODO: channel, gw, gd
elif m in {Detect, Segment}:
elif m is Detect:
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)
if m is Segment:
args[3] = make_divisible(args[3] * gw, 8)
elif m is Contract:
c2 = ch[f] * args[0] ** 2
elif m is Expand:
@@ -357,34 +303,34 @@ def parse_model(d, ch): # model_dict, input_channels(3)
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument('--cfg', type=str, default='yolov5s.yaml', help='model.yaml')
parser.add_argument('--batch-size', type=int, default=1, help='total batch size for all GPUs')
parser.add_argument('--cfg', type=str, default='yolov3yaml', help='model.yaml')
parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu')
parser.add_argument('--profile', action='store_true', help='profile model speed')
parser.add_argument('--line-profile', action='store_true', help='profile model speed layer by layer')
parser.add_argument('--test', action='store_true', help='test all yolo*.yaml')
opt = parser.parse_args()
opt.cfg = check_yaml(opt.cfg) # check YAML
print_args(vars(opt))
print_args(FILE.stem, opt)
device = select_device(opt.device)
# Create model
im = torch.rand(opt.batch_size, 3, 640, 640).to(device)
model = Model(opt.cfg).to(device)
model.train()
# Options
if opt.line_profile: # profile layer by layer
model(im, profile=True)
# Profile
if opt.profile:
img = torch.rand(8 if torch.cuda.is_available() else 1, 3, 640, 640).to(device)
y = model(img, profile=True)
elif opt.profile: # profile forward-backward
results = profile(input=im, ops=[model], n=3)
elif opt.test: # test all models
# Test all models
if opt.test:
for cfg in Path(ROOT / 'models').rglob('yolo*.yaml'):
try:
_ = Model(cfg)
except Exception as e:
print(f'Error in {cfg}: {e}')
else: # report fused model summary
model.fuse()
# Tensorboard (not working https://github.com/ultralytics/yolov5/issues/2898)
# from torch.utils.tensorboard import SummaryWriter
# tb_writer = SummaryWriter('.')
# LOGGER.info("Run 'tensorboard --logdir=models' to view tensorboard at http://localhost:6006/")
# tb_writer.add_graph(torch.jit.trace(model, img, strict=False), []) # add model graph