识别垃圾种类增多到10种，IMPLUSE网站建的模型识别准确率差，怎么改善？或者有其他训练网站吗？

# Edge Impulse - OpenMV Image Classification Example

import sensor, image, time, os, tf

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.
from pyb import UART
net = "trained.tflite"
labels = [line.rstrip('\n') for line in open("labels.txt")]
uart = UART(3, 115200)
uart.init(115200, bits=8, parity=None, stop=1)
clock = time.clock()
while(True):
    if uart.any():
        a=uart.readline().strip()
        clock.tick()
        time.sleep(1000)
        img = sensor.snapshot()
        for obj in tf.classify(net, img, min_scale=1.0, scale_mul=0.8, x_overlap=0.5, y_overlap=0.5):
            print("**********\nPredictions at [x=%d,y=%d,w=%d,h=%d]" % obj.rect())
            img.draw_rectangle(obj.rect())
            predictions_list = list(zip(labels, obj.output()))
            for i in range(len(predictions_list)):
                print("%s = %f" % (predictions_list[i][0], predictions_list[i][1]))
                max_size = predictions_list[0][1]
                zubie=0
                for i in range(len(predictions_list)):
                    if predictions_list[i][1] >= max_size:
                       zubie=i+1
                       max_size=predictions_list[i][1]
                    print(max_size)
                    print(zubie)
        if a==b'\x01':
           data = bytearray([zubie])
           uart.write(data)
           print(zubie)

为什么加了串口通信的语句之后，识别结果的概率和没加串口通信语句的概率差别这么大，没加的时候特别精确，但是加了之后老是出错

# Edge Impulse - OpenMV Image Classification Example

import sensor, image, time, os, tf

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 = "trained.tflite"
labels = [line.rstrip('\n') for line in open("labels.txt")]

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

    img = sensor.snapshot()

    # default settings just do one detection... change them to search the image...
    for obj in tf.classify(net, img, min_scale=1.0, scale_mul=0.8, x_overlap=0.5, y_overlap=0.5):
        print("**********\nPredictions at [x=%d,y=%d,w=%d,h=%d]" % obj.rect())
        img.draw_rectangle(obj.rect())
        # This combines the labels and confidence values into a list of tuples
        predictions_list = list(zip(labels, obj.output()))

        for i in range(len(predictions_list)):
            print("%s = %f" % (predictions_list[i][0], predictions_list[i][1]))

    print(clock.fps(), "fps")

下面的串行终端显示的是yaoshi这组数据最大应该最后显示2但是他先是了3