yolov5 and coral TPU

[taloot]

hello, i manage to run yolov5s and yolov5m on coral tpu with very high performace.. but i faced a problem with the compiler,, if i define more than 7 classes the edge compiler fail (if the resolution more than 416) i prefer to run on 512,, which is very good with coral tpu... except its limited classes ,,if i have more than 7 classes then only 416 and less works and if i use less than 7 classes i can go as high as 544

[taloot]

https://www.youtube.com/watch?v=D9IExho8pwo this is updated link

[pirazor]

Hi @taloot, I was able to run 416 model on 98 classes without an issue. Would you mind sharing your postquantized tflite and edgetpu model ? I can maybe give you a suggestion on changing the last transpose operation to 4D.

[taloot]

The problem never happend on 416

It happend if i want more than 416 I m using currently 7 classes and 512 I wish i can run more than 512 coz i never success compiling the 640

On Sun, 27 Jun 2021 at 11:56 PM ENES KARAASLAN @.***> wrote:

Hi @taloot https://github.com/taloot, I was able to run 416 model on 98 classes without an issue. Would you mind sharing your postquantized tflite and edgetpu model ? I can maybe give you a suggestion on changing the last transpose operation to 4D.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/google-coral/edgetpu/issues/405#issuecomment-869222603, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADGSH32RUOBHNAYTHGUYZU3TU6GA5ANCNFSM47LCSQPA .

[taloot]

https://www.sendgb.com/upload/?utm_source=iRw3E1XMmHY this is the files u may want, when i make the freeze graphic i have to specify the resloution if i have more than 7 classes the compile stops at 416... any number above this will generate internal error if i have 7 and less, the max working is 512 keep in mind im using the -a flag in the compiler

[taloot]

161Newds-int8_edgetpu.log Edge TPU Compiler version 15.0.340273435 Input: 161Newds-int8.tflite Output: 161Newds-int8_edgetpu.tflite

Operator Count Status

CONV_2D 86 Mapped to Edge TPU QUANTIZE 24 Mapped to Edge TPU MAX_POOL_2D 3 Mapped to Edge TPU ADD 17 Mapped to Edge TPU SUB 3 Mapped to Edge TPU PAD 6 Mapped to Edge TPU LOGISTIC 86 Mapped to Edge TPU STRIDED_SLICE 9 Mapped to Edge TPU STRIDED_SLICE 4 Only Strided-Slice with unitary strides supported MUL 104 Mapped to Edge TPU TRANSPOSE 3 Operation not supported RESHAPE 6 Mapped to Edge TPU CONCATENATION 18 Mapped to Edge TPU RESIZE_NEAREST_NEIGHBOR 2 Mapped to Edge TPU

[taloot]

Hi @taloot, I was able to run 416 model on 98 classes without an issue. Would you mind sharing your postquantized tflite and edgetpu model ? I can maybe give you a suggestion on changing the last transpose operation to 4D.

try to increases the resolution i m sure will see the problem

[taloot]

`import argparse import logging import os import sys import traceback from copy import deepcopy from pathlib import Path

sys.path.append('./') # to run '$ python *.py' files in subdirectories

import numpy as np import tensorflow as tf import torch import torch.nn as nn import yaml from tensorflow import keras from tensorflow.python.framework.convert_to_constants import convert_variables_to_constants_v2

from models.common import Conv, Bottleneck, SPP, DWConv, Focus, BottleneckCSP, Concat, autopad, C3 from models.experimental import MixConv2d, CrossConv, attempt_load from models.yolo import Detect from utils.datasets import LoadImages from utils.general import make_divisible, check_file, check_dataset from utils.google_utils import attempt_download

logger = logging.getLogger(name)

class tf_BN(keras.layers.Layer):

TensorFlow BatchNormalization wrapper

def __init__(self, w=None):
    super(tf_BN, self).__init__()
    self.bn = keras.layers.BatchNormalization(
        beta_initializer=keras.initializers.Constant(w.bias.numpy()),
        gamma_initializer=keras.initializers.Constant(w.weight.numpy()),
        moving_mean_initializer=keras.initializers.Constant(w.running_mean.numpy()),
        moving_variance_initializer=keras.initializers.Constant(w.running_var.numpy()),
        epsilon=w.eps)

def call(self, inputs):
    return self.bn(inputs)

class tf_Pad(keras.layers.Layer): def init(self, pad): super(tf_Pad, self).init() 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)

class tf_Conv(keras.layers.Layer):

Standard convolution

def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True, w=None):
    # ch_in, ch_out, weights, kernel, stride, padding, groups
    super(tf_Conv, self).__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(
        c2, k, s, 'SAME' if s == 1 else 'VALID', use_bias=False,
        kernel_initializer=keras.initializers.Constant(w.conv.weight.permute(2, 3, 1, 0).numpy()))
    self.conv = conv if s == 1 else keras.Sequential([tf_Pad(autopad(k, p)), conv])
    self.bn = tf_BN(w.bn) if hasattr(w, 'bn') else tf.identity

    # YOLOv5 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):
        self.act = (lambda x: keras.activations.swish(x)) if act else tf.identity

def call(self, inputs):
    return self.act(self.bn(self.conv(inputs)))

class tf_Focus(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):
    # ch_in, ch_out, kernel, stride, padding, groups
    super(tf_Focus, self).__init__()
    self.conv = tf_Conv(c1 * 4, c2, k, s, p, g, act, w.conv)

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
    return self.conv(tf.concat([inputs[:, ::2, ::2, :],
                                inputs[:, 1::2, ::2, :],
                                inputs[:, ::2, 1::2, :],
                                inputs[:, 1::2, 1::2, :]], 3))

class tf_Bottleneck(keras.layers.Layer):

Standard bottleneck

def __init__(self, c1, c2, shortcut=True, g=1, e=0.5, w=None):  # ch_in, ch_out, shortcut, groups, expansion
    super(tf_Bottleneck, self).__init__()
    c_ = int(c2 * e)  # hidden channels
    self.cv1 = tf_Conv(c1, c_, 1, 1, w=w.cv1)
    self.cv2 = tf_Conv(c_, c2, 3, 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 tf_Conv2d(keras.layers.Layer):

Substitution for PyTorch nn.Conv2D

def __init__(self, c1, c2, k, s=1, g=1, bias=True, w=None):
    super(tf_Conv2d, self).__init__()
    assert g == 1, "TF v2.2 Conv2D does not support 'groups' argument"
    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)

class tf_BottleneckCSP(keras.layers.Layer):

CSP Bottleneck https://github.com/WongKinYiu/CrossStagePartialNetworks

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(tf_BottleneckCSP, self).__init__()
    c_ = int(c2 * e)  # hidden channels
    self.cv1 = tf_Conv(c1, c_, 1, 1, w=w.cv1)
    self.cv2 = tf_Conv2d(c1, c_, 1, 1, bias=False, w=w.cv2)
    self.cv3 = tf_Conv2d(c_, c_, 1, 1, bias=False, w=w.cv3)
    self.cv4 = tf_Conv(2 * c_, c2, 1, 1, w=w.cv4)
    self.bn = tf_BN(w.bn)
    self.act = lambda x: keras.activations.relu(x, alpha=0.1)
    self.m = keras.Sequential([tf_Bottleneck(c_, c_, shortcut, g, e=1.0, w=w.m[j]) for j in range(n)])

def call(self, inputs):
    y1 = self.cv3(self.m(self.cv1(inputs)))
    y2 = self.cv2(inputs)
    return self.cv4(self.act(self.bn(tf.concat((y1, y2), axis=3))))

class tf_C3(keras.layers.Layer):

CSP Bottleneck with 3 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(tf_C3, self).__init__()
    c_ = int(c2 * e)  # hidden channels
    self.cv1 = tf_Conv(c1, c_, 1, 1, w=w.cv1)
    self.cv2 = tf_Conv(c1, c_, 1, 1, w=w.cv2)
    self.cv3 = tf_Conv(2 * c_, c2, 1, 1, w=w.cv3)
    self.m = keras.Sequential([tf_Bottleneck(c_, c_, shortcut, g, e=1.0, 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 tf_SPP(keras.layers.Layer):

Spatial pyramid pooling layer used in YOLOv3-SPP

def __init__(self, c1, c2, k=(5, 9, 13), w=None):
    super(tf_SPP, self).__init__()
    c_ = c1 // 2  # hidden channels
    self.cv1 = tf_Conv(c1, c_, 1, 1, w=w.cv1)
    self.cv2 = tf_Conv(c_ * (len(k) + 1), c2, 1, 1, w=w.cv2)
    self.m = [keras.layers.MaxPool2D(pool_size=x, strides=1, padding='SAME') for x in k]

def call(self, inputs):
    x = self.cv1(inputs)
    return self.cv2(tf.concat([x] + [m(x) for m in self.m], 3))

class tf_Detect(keras.layers.Layer): def init(self, nc=80, anchors=(), ch=(), w=None): # detection layer super(tf_Detect, self).init() self.stride = tf.convert_to_tensor(w.stride.numpy(), dtype=tf.float32) self.nc = nc # number of classes 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 = [tf.zeros(1)] self.nl # init grid self.anchors = tf.convert_to_tensor(w.anchors.numpy(), dtype=tf.float32) self.anchor_grid = tf.reshape(tf.convert_to_tensor(w.anchor_grid.numpy(), dtype=tf.float32), [self.nl, 1, -1, 1, 2]) self.m = [tf_Conv2d(x, self.no self.na, 1, w=w.m[i]) for i, x in enumerate(ch)] self.export = False # onnx export self.training = False # set to False after building model for i in range(self.nl): ny, nx = opt.img_size[0] // self.stride[i], opt.img_size[1] // self.stride[i] self.grid[i] = self._make_grid(nx, ny)

def call(self, inputs):
    # x = x.copy()  # for profiling
    z = []  # inference output
    self.training |= self.export
    x = []
    for i in range(self.nl):
        x.append(self.m[i](inputs[i]))
        # x(bs,20,20,255) to x(bs,3,20,20,85)
        ny, nx = opt.img_size[0] // self.stride[i], opt.img_size[1] // self.stride[i]
        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 = 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([[opt.img_size[1], opt.img_size[0]]], dtype=tf.float32)
            wh /= tf.constant([[opt.img_size[1], opt.img_size[0]]], dtype=tf.float32)
            y = tf.concat([xy, wh, y[..., 4:]], -1)
            z.append(tf.reshape(y, [-1, 3 * ny * nx, self.no]))

    return x if self.training else (tf.concat(z, 1), x)

@staticmethod
def _make_grid(nx=20, ny=20):
    # yv, xv = torch.meshgrid([torch.arange(ny), torch.arange(nx)])
    # return torch.stack((xv, yv), 2).view((1, 1, ny, nx, 2)).float()
    xv, yv = tf.meshgrid(tf.range(nx), tf.range(ny))
    return tf.cast(tf.reshape(tf.stack([xv, yv], 2), [1, 1, ny * nx, 2]), dtype=tf.float32)

class tf_Upsample(keras.layers.Layer): def init(self, size, scale_factor, mode, w=None): super(tf_Upsample, self).init() assert scale_factor == 2, "scale_factor must be 2"

self.upsample = keras.layers.UpSampling2D(size=scale_factor, interpolation=mode)

    if opt.tf_raw_resize:
        # with default arguments: align_corners=False, half_pixel_centers=False
        self.upsample = lambda x: tf.raw_ops.ResizeNearestNeighbor(images=x,
                                                                   size=(x.shape[1] * 2, x.shape[2] * 2))
    else:
        self.upsample = lambda x: tf.image.resize(x, (x.shape[1] * 2, x.shape[2] * 2), method=mode)

def call(self, inputs):
    return self.upsample(inputs)

class tf_Concat(keras.layers.Layer): def init(self, dimension=1, w=None): super(tf_Concat, self).init() assert dimension == 1, "convert only NCHW to NHWC concat" self.d = 3

def call(self, inputs):
    return tf.concat(inputs, self.d)

def parse_model(d, ch, model): # model_dict, input_channels(3) logger.info('\n%3s%18s%3s%10s %-40s%-30s' % ('', 'from', 'n', 'params', 'module', 'arguments')) 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)

layers, save, c2 = [], [], ch[-1]  # layers, savelist, ch out
for i, (f, n, m, args) in enumerate(d['backbone'] + d['head']):  # from, number, module, args
    m_str = m
    m = eval(m) if isinstance(m, str) else m  # eval strings
    for j, a in enumerate(args):
        try:
            args[j] = eval(a) if isinstance(a, str) else a  # eval strings
        except:
            pass

    n = max(round(n * gd), 1) if n > 1 else n  # depth gain
    if m in [nn.Conv2d, Conv, Bottleneck, SPP, 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]:
            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 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)
    else:
        c2 = ch[f]

    tf_m = eval('tf_' + m_str.replace('nn.', ''))
    m_ = keras.Sequential([tf_m(*args, w=model.model[i][j]) for j in range(n)]) if n > 1 \
        else tf_m(*args, w=model.model[i])  # module

    torch_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
    np = sum([x.numel() for x in torch_m_.parameters()])  # number params
    m_.i, m_.f, m_.type, m_.np = i, f, t, np  # attach index, 'from' index, type, number params
    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_)
    ch.append(c2)
return keras.Sequential(layers), sorted(save)

class tf_Model(): def init(self, cfg='yolov5s.yaml', ch=3, nc=None, model=None): # model, input channels, number of classes super(tf_Model, self).init() if isinstance(cfg, dict): self.yaml = cfg # model dict else: # is *.yaml 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

    # Define model
    if nc and nc != self.yaml['nc']:
        print('Overriding %s nc=%g with nc=%g' % (cfg, self.yaml['nc'], nc))
        self.yaml['nc'] = nc  # override yaml value
    self.model, self.savelist = parse_model(deepcopy(self.yaml), ch=[ch], model=model)  # model, savelist, ch_out

def predict(self, inputs, profile=False):
    y = []  # outputs
    x = inputs
    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

        x = m(x)  # run
        y.append(x if m.i in self.savelist else None)  # save output

    # Add TensorFlow NMS
    if opt.tf_nms:
        boxes = tf.expand_dims(xywh2xyxy(x[0][..., :4]), 2)
        probs = x[0][:, :, 4:5]
        classes = x[0][:, :, 5:]
        scores = probs * classes
        nms = tf.image.combined_non_max_suppression(
            boxes, scores, opt.topk_per_class, opt.topk_all, opt.iou_thres, opt.score_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
    # return tf.concat([conf, cls, xywh], 1)

def xywh2xyxy(xywh):

Convert nx4 boxes from [x, y, w, h] to [x1, y1, x2, y2] where xy1=top-left, xy2=bottom-right

x, y, w, h = tf.split(xywh, num_or_size_splits=4, axis=-1)
return tf.concat([x - w / 2, y - h / 2, x + w / 2, y + h / 2], axis=-1)

def representative_dataset_gen():

Representative dataset for use with converter.representative_dataset

n = 0
for path, img, im0s, vid_cap in dataset:
    # Get sample input data as a numpy array in a method of your choosing.
    n += 1
    input = np.transpose(img, [1, 2, 0])
    input = np.expand_dims(input, axis=0).astype(np.float32)
    input /= 255.0
    yield [input]
    if n >= opt.ncalib:
        break

if name == "main": parser = argparse.ArgumentParser() parser.add_argument('--cfg', type=str, default='yolov5s.yaml', help='cfg path') parser.add_argument('--weights', type=str, default='yolov5s.pt', help='weights path') parser.add_argument('--img-size', nargs='+', type=int, default=[320, 320], help='image size') # height, width parser.add_argument('--batch-size', type=int, default=1, help='batch size') parser.add_argument('--dynamic-batch-size', action='store_true', help='dynamic batch size') parser.add_argument('--source', type=str, default='../data/coco128.yaml', help='dir of images or data.yaml file') parser.add_argument('--ncalib', type=int, default=100, help='number of calibration images') parser.add_argument('--tfl-int8', action='store_true', dest='tfl_int8', help='export TFLite int8 model') parser.add_argument('--tf-nms', action='store_true', dest='tf_nms', help='TF NMS (without TFLite export)') parser.add_argument('--tf-raw-resize', action='store_true', dest='tf_raw_resize', help='use tf.raw_ops.ResizeNearestNeighbor for resize') parser.add_argument('--topk-per-class', type=int, default=100, help='topk per class to keep in NMS') parser.add_argument('--topk-all', type=int, default=100, help='topk for all classes to keep in NMS') parser.add_argument('--iou-thres', type=float, default=0.5, help='IOU threshold for NMS') parser.add_argument('--score-thres', type=float, default=0.4, help='score threshold for NMS') opt = parser.parse_args() opt.cfg = check_file(opt.cfg) # check file opt.img_size *= 2 if len(opt.img_size) == 1 else 1 # expand print(opt)

# Input
img = torch.zeros((opt.batch_size, 3, *opt.img_size))  # image size(1,3,320,192) iDetection

# Load PyTorch model
model = attempt_load(opt.weights, map_location=torch.device('cpu'), inplace=True, fuse=False)
model.model[-1].export = False  # set Detect() layer export=True
y = model(img)  # dry run
nc = y[0].shape[-1] - 5

# TensorFlow saved_model export
try:
    print('\nStarting TensorFlow saved_model export with TensorFlow %s...' % tf.__version__)
    tf_model = tf_Model(opt.cfg, model=model, nc=nc)
    img = tf.zeros((opt.batch_size, *opt.img_size, 3))  # NHWC Input for TensorFlow

    m = tf_model.model.layers[-1]
    assert isinstance(m, tf_Detect), "the last layer must be Detect"
    m.training = False
    y = tf_model.predict(img)

    inputs = keras.Input(shape=(*opt.img_size, 3), batch_size=None if opt.dynamic_batch_size else opt.batch_size)
    keras_model = keras.Model(inputs=inputs, outputs=tf_model.predict(inputs))
    keras_model.summary()
    path = opt.weights.replace('.pt', '_saved_model')  # filename
    keras_model.save(path, save_format='tf')
    print('TensorFlow saved_model export success, saved as %s' % path)
except Exception as e:
    print('TensorFlow saved_model export failure: %s' % e)
    traceback.print_exc(file=sys.stdout)

# TensorFlow GraphDef export
try:
    print('\nStarting TensorFlow GraphDef export with TensorFlow %s...' % tf.__version__)

    # https://github.com/leimao/Frozen_Graph_TensorFlow
    full_model = tf.function(lambda x: keras_model(x))

    full_model = full_model.get_concrete_function(
        tf.TensorSpec(keras_model.inputs[0].shape, keras_model.inputs[0].dtype))

    frozen_func = convert_variables_to_constants_v2(full_model)
    frozen_func.graph.as_graph_def()
    f = opt.weights.replace('.pt', '.pb')  # filename
    tf.io.write_graph(graph_or_graph_def=frozen_func.graph,
                      logdir=os.path.dirname(f),
                      name=os.path.basename(f),
                      as_text=False)

    print('TensorFlow GraphDef export success, saved as %s' % f)
except Exception as e:
    print('TensorFlow GraphDef export failure: %s' % e)
    traceback.print_exc(file=sys.stdout)

# TFLite model export
if not opt.tf_nms:
    try:
        print('\nStarting TFLite export with TensorFlow %s...' % tf.__version__)

        # fp32 TFLite model export ---------------------------------------------------------------------------------
        # converter = tf.lite.TFLiteConverter.from_keras_model(keras_model)
        # converter.target_spec.supported_ops = [tf.lite.OpsSet.TFLITE_BUILTINS]
        # converter.allow_custom_ops = False
        # converter.experimental_new_converter = True
        # tflite_model = converter.convert()
        # f = opt.weights.replace('.pt', '.tflite')  # filename
        # open(f, "wb").write(tflite_model)

        # fp16 TFLite model export ---------------------------------------------------------------------------------
        converter = tf.lite.TFLiteConverter.from_keras_model(keras_model)
        converter.optimizations = [tf.lite.Optimize.DEFAULT]
        # converter.representative_dataset = representative_dataset_gen
        # converter.target_spec.supported_types = [tf.float16]
        converter.target_spec.supported_ops = [tf.lite.OpsSet.TFLITE_BUILTINS]
        converter.allow_custom_ops = False
        converter.experimental_new_converter = True
        tflite_model = converter.convert()
        f = opt.weights.replace('.pt', '-fp16.tflite')  # filename
        open(f, "wb").write(tflite_model)
        print('\nTFLite export success, saved as %s' % f)

        # int8 TFLite model export ---------------------------------------------------------------------------------
        if opt.tfl_int8:
            # Representative Dataset
            if opt.source.endswith('.yaml'):
                with open(check_file(opt.source)) as f:
                    data = yaml.load(f, Loader=yaml.FullLoader)  # data dict
                    check_dataset(data)  # check
                opt.source = data['train']
            dataset = LoadImages(opt.source, img_size=opt.img_size, auto=False)
            converter = tf.lite.TFLiteConverter.from_keras_model(keras_model)
            converter.optimizations = [tf.lite.Optimize.DEFAULT]
            converter.representative_dataset = representative_dataset_gen
            converter.target_spec.supported_ops = [tf.lite.OpsSet.TFLITE_BUILTINS_INT8]
            converter.inference_input_type = tf.uint8  # or tf.int8
            converter.inference_output_type = tf.uint8  # or tf.int8
            converter.allow_custom_ops = True
            converter.experimental_new_converter = True
            converter.experimental_new_quantizer = False
            tflite_model = converter.convert()
            f = opt.weights.replace('.pt', '-int8.tflite')  # filename
            open(f, "wb").write(tflite_model)
            print('\nTFLite (int8) export success, saved as %s' % f)

    except Exception as e:
        print('\nTFLite export failure: %s' % e)
        traceback.print_exc(file=sys.stdout)`

[pirazor]

Hi @taloot, I was able to run 416 model on 98 classes without an issue. Would you mind sharing your postquantized tflite and edgetpu model ? I can maybe give you a suggestion on changing the last transpose operation to 4D.

try to increases the resolution i m sure will see the problem

Hmm, I am actually able to run the 640 pretrained model. it is slow, but it runs. I couldn't access your files btw, the link may have expired. But, I think my tflite model is compiled slightly different than yours. Can you share the netron images of your graph ?

[taloot]

161512-int8_edgetpu.zip this is my working high speed yolov5s@512

this one im able to reach 20fps solid yolov5s 512x512 = 45ms yolov5s 448x448 = 38ms yolov5s 416x416=26ms yolov5s 320x320 = 18ms yolov5m 448x448 = 90ms thats my score how about urs and i guess 640x640 with yolov5s will be perfect for my application,

[taloot]

Hi @taloot, I was able to run 416 model on 98 classes without an issue. Would you mind sharing your postquantized tflite and edgetpu model ? I can maybe give you a suggestion on changing the last transpose operation to 4D.

try to increases the resolution i m sure will see the problem

Hmm, I am actually able to run the 640 pretrained model. it is slow, but it runs. I couldn't access your files btw, the link may have expired. But, I think my tflite model is compiled slightly different than yours. Can you share the netron images of your graph ?

u success compiler 640 with -a flag? without -a i can but with -a i cant

[pirazor]

@taloot, your model has multiple subgraphs that might be causing IO issues with large tensors being sent to CPU back and forth. Are you using a USB accelerator attached to a PC ? Would you also share the int8 tflite model before you compile. You may not need that last transpose operations which output large tensors into the TPU.

[taloot]

thanks for ur quick response. i have m2 coral, tpu.zip

[taloot]

@taloot, your model has multiple subgraphs that might be causing IO issues with large tensors being sent to CPU back and forth. Are you using a USB accelerator attached to a PC ? Would you also share the int8 tflite model before you compile. You may not need that last transpose operations which output large tensors into the TPU.

i love to map the transpose to tpu,, best i can achive for yolov5s is 38ms 512x512 which is perfect for me,, but the transpose mapping i guess will add littel bit gain

[hjonnala]

@taloot please check this comment if it helps https://github.com/google-coral/edgetpu/issues/419#issuecomment-889878751

[taloot]

Thanks this look like new way to convert the model i will try it and let you know On Fri, 30 Jul 2021 at 9:10 PM Hemanth-Jonnala @.***> wrote:

@taloot https://github.com/taloot please check this comment if it helps #419 (comment) https://github.com/google-coral/edgetpu/issues/419#issuecomment-889878751

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/google-coral/edgetpu/issues/405#issuecomment-890066538, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADGSH36ROTIHHKGYNMYMHL3T2LTKFANCNFSM47LCSQPA .

[taloot]

both ways didnt work for me

im only able to have 7 class on 512 max this is the best result i reached with yolov5

[AkkiSony]

Hi Guys, I converted darknet - Yolov3 model to tflite model using the follwing link - https://github.com/guichristmann/edge-tpu-tiny-yolo

I could successfully convert the model, but when I run inference using Inference.py script, I got no prediction on my image at all. You can find my model along with the classes here. https://drive.google.com/drive/folders/181npG1SDJnMBBQOm_gXbL0XguSA4gGsy?usp=sharing

Please let me know if you have any lead. Thank you in advance. :) @pirazor @taloot

[hjonnala]

@taloot does your issue got resolved with new compiler? If yes, please close the ticket.

[taloot]

no still max size 512

[hjonnala]

@taloot can you please share the tflite model with size 640. Thanks!