Frozzenizzy 2020-04-25 15:52 采纳率: 0%
浏览 238

keras用2层cnn和2层全连接搭建食物图像识别的神经网络acc一直在50上不去。

现在用10x750+10x250张图片做训练测试
总体有101x750+101x250
不知道这样测试出来的网络能不能直接套用在全部的数据上(因为电脑原因跑的比较慢才这样测试的)
用10类别测试的时候 acc比较低而且把待测图片放入测试文件中的时候会出现100%但指向另外类别的结果

import sys
import os
from keras.preprocessing.image import ImageDataGenerator
from keras import optimizers
from keras.models import Sequential
from keras.layers import Dropout, Flatten, Dense, Activation
from keras.layers.convolutional import Convolution2D, MaxPooling2D
from keras import callbacks
import time
import tensorflow as tf

config = tf.ConfigProto(gpu_options=tf.GPUOptions(allow_growth=True))
sess = tf.Session(config=config)

start = time.time()

DEV = False
argvs = sys.argv
argc = len(argvs)

if argc > 1 and (argvs[1] == "--development" or argvs[1] == "-d"):
  DEV = True

if DEV:
  epochs = 2
else:
  epochs = 10

train_data_path = 'data/train'
validation_data_path = 'data/test'

"""
Parameters
"""
img_width, img_height = 100, 100
batch_size = 32
samples_per_epoch = 7500
validation_steps = 490
nb_filters1 = 32
nb_filters2 = 64
conv1_size = 3
conv2_size = 2
pool_size = 2
classes_num = 10
lr = 0.0004

model = Sequential()
model.add(Convolution2D(
  nb_filters1,
  conv1_size,
  conv1_size,
  border_mode ="same",
  input_shape=(img_width, img_height, 3)
))
model.add(Activation("relu"))
model.add(MaxPooling2D(
  pool_size=(pool_size, pool_size)
))

model.add(Convolution2D(nb_filters2, conv2_size, conv2_size, border_mode ="same"))
model.add(Activation("relu"))
model.add(Convolution2D(nb_filters2, conv2_size, conv2_size, border_mode ="same"))
model.add(Activation("relu"))
model.add(MaxPooling2D(pool_size=(pool_size, pool_size), dim_ordering='th'))

model.add(Flatten())
model.add(Dense(256))
model.add(Activation("relu"))
model.add(Dropout(0.5))
model.add(Dense(classes_num, activation='softmax'))

model.compile(loss='categorical_crossentropy',
              optimizer=optimizers.RMSprop(lr=lr),
              metrics=['accuracy'])

train_datagen = ImageDataGenerator(
    rescale=1. / 255,
    shear_range=0.2,
    zoom_range=0.2,
    horizontal_flip=True)

test_datagen = ImageDataGenerator(rescale=1. / 255)

train_generator = train_datagen.flow_from_directory(
    train_data_path,
    target_size=(img_height, img_width),
    batch_size=batch_size,
    class_mode='categorical')

validation_generator = test_datagen.flow_from_directory(
    validation_data_path,
    target_size=(img_height, img_width),
    batch_size=batch_size,
    class_mode='categorical')

"""
Tensorboard log
"""
log_dir = './training101tf-log/'
tb_cb = callbacks.TensorBoard(log_dir=log_dir, histogram_freq=0)
cbks = [tb_cb]

model.fit_generator(
    train_generator,
    samples_per_epoch=samples_per_epoch,
    epochs=epochs,
    validation_data=validation_generator,
    callbacks=cbks,
    validation_steps=validation_steps)

target_dir = './models_training101/'
if not os.path.exists(target_dir):
  os.mkdir(target_dir)
model.save('./models_training101/model.h5')
model.save_weights('./models_training101/weights.h5')

#Calculate execution time
end = time.time()
dur = end-start

if dur<60:
    print("Execution Time:",dur,"seconds")
elif dur>60 and dur<3600:
    dur=dur/60
    print("Execution Time:",dur,"minutes")
else:
    dur=dur/(60*60)
    print("Execution Time:",dur,"hours")

  • 写回答

1条回答 默认 最新

  • threenewbee 2020-04-25 15:59
    关注

    你的网络不太合理,应该conv+pooling+conv+pooling逐渐缩小,而且你的样本也太少

    评论

报告相同问题?

悬赏问题

  • ¥15 matlab数字图像处理频率域滤波
  • ¥15 在abaqus做了二维正交切削模型,给刀具添加了超声振动条件后输出切削力为什么比普通切削增大这么多
  • ¥15 ELGamal和paillier计算效率谁快?
  • ¥15 file converter 转换格式失败 报错 Error marking filters as finished,如何解决?
  • ¥15 ubuntu系统下挂载磁盘上执行./提示权限不够
  • ¥15 Arcgis相交分析无法绘制一个或多个图形
  • ¥15 关于#r语言#的问题:差异分析前数据准备,报错Error in data[, sampleName1] : subscript out of bounds请问怎么解决呀以下是全部代码:
  • ¥15 seatunnel-web使用SQL组件时候后台报错,无法找到表格
  • ¥15 fpga自动售货机数码管(相关搜索:数字时钟)
  • ¥15 用前端向数据库插入数据,通过debug发现数据能走到后端,但是放行之后就会提示错误