只能训练20轮，效果会不会很差

这个方法现在是不能用了嘛

@kidswong999 就是openmv插电脑上识别效果挺好的，但是把openmv连到工控机上，识别效果就不行了

@xhk5 神经网络识别的时候也是和在edge implase上label的时候一样去掉图像两边的吗

这是怎么回事啊，应该怎么解决

@kidswong999 好的好的谢谢啦

import sensor, image, time, os, tf, math, uos, gc,pyb
from pyb import UART
from pyb import LED
red_led = LED(1)
sensor.reset()
sensor.set_pixformat(sensor.RGB565)
sensor.set_framesize(sensor.QVGA)
sensor.set_windowing((240, 240))
sensor.skip_frames(time=2000)
uart = UART(3, 9600)
net = None
labels = None
min_confidence = 0.92
try:
net = tf.load("trained.tflite", load_to_fb=uos.stat('trained.tflite')[6] > (gc.mem_free() - (64*1024)))
except Exception as e:
raise Exception('Failed to load "trained.tflite", did you copy the .tflite and labels.txt file onto the mass-storage device? (' + str(e) + ')')
try:
labels = [line.rstrip('\n') for line in open("labels.txt")]
except Exception as e:
raise Exception('Failed to load "labels.txt", did you copy the .tflite and labels.txt file onto the mass-storage device? (' + str(e) + ')')
colors = [
(255, 0, 0),
( 0, 255, 0),
(255, 255, 0),
( 0, 0, 255),
(255, 0, 255),
( 0, 255, 255),
(255, 255, 255),
]
r=0
a=0
count=0
b=0
t=0
flag=0
A=C=D=E=H=I=J=K=L=M=N=O=0
clock = time.clock()
while(True):
img = sensor.snapshot()
if uart.any():
red_led.on()
time.sleep_ms(500)
red_led.off()
A=C=D=E=H=I=J=K=L=M=N=O=0
start = pyb.millis()
count=0
m=uart.read(1).decode()
if m!='o':
a=int(m)

    while( count <=6 ):
        detected = True
        img = sensor.snapshot()
        if pyb.millis() - start>5000:
            uart.write("d")
            count=7
        for i, detection_list in enumerate(net.detect(img, thresholds=[(math.ceil(min_confidence * 255), 255)])):

            if (i == 0):

                continue

            if (len(detection_list) == 0):

                continue


            print("********** %s **********" % labels[i])
            for d in detection_list:
                [x, y, w, h] = d.rect()
                center_x = math.floor(x + (w / 2))
                center_y = math.floor(y + (h / 2))
                print('x %d\ty %d' % (center_x, center_y))
            if center_y>100:
                if labels[i]=="旺仔" :
                    if center_x<=60:
                        A=A+1
                    elif center_x>60 and center_x<=165:
                        C=C+1
                    else:
                        D=D+1
                if labels[i]=="王老吉" :
                    if center_x<=60:
                        E=E+1
                    elif center_x>60 and center_x<=165:
                        H=H+1
                    else:
                        I=I+1
                if labels[i]=="雪花" :
                    if center_x<=60:
                        J=J+1
                    elif center_x>60 and center_x<=165:
                        K=K+1
                    else:
                        L=L+1
                if labels[i]=="AD" :
                    if center_x<=60:
                        M=M+1
                    elif center_x>60 and center_x<=165:
                        N=N+1
                    else:
                        O=O+1
                count=count+1
        #if detected:
        #    flag=1
        #if flag==1:
        #    uart.write("d")
        #    flag=0
    list=[A,C,D,E,H,I,J,K,L,M,N,O]
    Z=max(list)
    if Z<=3 and Z>0 :

        uart.write("d")
        A=C=D=E=H=I=J=K=L=M=N=O=0

    elif Z>3:
        if A==Z :
            uart.write("A")

        if C==Z :
            uart.write("C")
        if D==Z :
            uart.write("D")
        if E==Z :
            uart.write("E")
        if H==Z :
            uart.write("H")
        if I==Z :
            uart.write("I")
        if J==Z :
            uart.write("J")
        if K==Z :
            uart.write("K")
        if L==Z :
            uart.write("L")
        if M==Z :
            uart.write("M")
        if N==Z :
            uart.write("N")
        if O==Z :
            uart.write("O")
        A=C=D=E=H=I=J=K=L=M=N=O=0

# Edge Impulse - OpenMV Object Detection Example

import sensor, image, time, os, tf, math, uos, gc

sensor.reset()                         # Reset and initialize the sensor.
sensor.set_pixformat(sensor.RGB565)    # Set pixel format to RGB565 (or GRAYSCALE)
sensor.set_framesize(sensor.QVGA)      # Set frame size to QVGA (320x240)
sensor.set_windowing((240, 240))       # Set 240x240 window.
sensor.skip_frames(time=2000)          # Let the camera adjust.

net = None
labels = None
min_confidence = 0.5

try:
    # load the model, alloc the model file on the heap if we have at least 64K free after loading
    net = tf.load("trained.tflite", load_to_fb=uos.stat('trained.tflite')[6] > (gc.mem_free() - (64*1024)))
except Exception as e:
    raise Exception('Failed to load "trained.tflite", did you copy the .tflite and labels.txt file onto the mass-storage device? (' + str(e) + ')')

try:
    labels = [line.rstrip('\n') for line in open("labels.txt")]
except Exception as e:
    raise Exception('Failed to load "labels.txt", did you copy the .tflite and labels.txt file onto the mass-storage device? (' + str(e) + ')')

colors = [ # Add more colors if you are detecting more than 7 types of classes at once.
    (255,   0,   0),
    (  0, 255,   0),
    (255, 255,   0),
    (  0,   0, 255),
    (255,   0, 255),
    (  0, 255, 255),
    (255, 255, 255),
]

clock = time.clock()
while(True):
    clock.tick()

    img = sensor.snapshot()

    # detect() returns all objects found in the image (splitted out per class already)
    # we skip class index 0, as that is the background, and then draw circles of the center
    # of our objects

    for i, detection_list in enumerate(net.detect(img, thresholds=[(math.ceil(min_confidence * 255), 255)])):
        if (i == 0): continue # background class
        if (len(detection_list) == 0): continue # no detections for this class?

        print("********** %s **********" % labels[i])
        for d in detection_list:
            [x, y, w, h] = d.rect()
            center_x = math.floor(x + (w / 2))
            center_y = math.floor(y + (h / 2))
            print('x %d\ty %d' % (center_x, center_y))
            img.draw_circle((center_x, center_y, 12), color=colors[i], thickness=2)

    print(clock.fps(), "fps", end="\n\n")

@kidswong999 找不同呀，左中右放三个未知的物品，找出不一样的一个

@kidswong999 但是神经网络不是要预先录入已知的物品的嘛，我说的这个三个物品是未知的