百度AI Studio首頁https://aistudio.baidu.com/aistudio/index

一毫别、課程評價(jià)

1身弊、每日課程分為理論部分和實(shí)踐部分辟汰，課程內(nèi)容由淺至深列敲，循序漸進(jìn)。每次課后都有作業(yè)帖汞，用以鞏固課上所學(xué)的知識酿炸，親自上手搭建網(wǎng)絡(luò)，調(diào)整參數(shù)涨冀，不斷提高準(zhǔn)確率填硕，滿滿的成就感。

2鹿鳖、課程微信群學(xué)習(xí)氛圍濃郁扁眯，班班和助教們都很專業(yè)，幾乎所有的問題發(fā)到群里后翅帜，很快就能得到班班姻檀，助教和同學(xué)們的幫助。

3涝滴、免費(fèi)绣版，完全免費(fèi)，paddlepaddle為每位同學(xué)承擔(dān)1699元/人學(xué)費(fèi)歼疮，每天運(yùn)行AI Studio項(xiàng)目就能獲得12小時(shí)免費(fèi)GPU資源（Tesla V100）杂抽，連續(xù)5天還能額外獲得48小時(shí)，完全就是白嫖啊韩脏。

4缩麸、獎(jiǎng)勵(lì)多多，完成作業(yè)+打比賽就可以獲得結(jié)業(yè)證書赡矢，還有小度音響杭朱、小度在家、深度學(xué)習(xí)書籍等各種獎(jiǎng)勵(lì)吹散。

二弧械、課程內(nèi)容

Day 1：新冠疫情可視化

1、本地安裝PaddlePaddle：https://www.paddlepaddle.org.cn/documentation/docs/zh/install/index_cn.html
2空民、使用Pyecharts進(jìn)行全國疫情實(shí)時(shí)數(shù)據(jù)

#安裝pyecharts
#pip install -i https://pypi.tuna.tsinghua.edu.cn/simple pyecharts
import json
import datetime
from pyecharts.charts import Map
from pyecharts import options as opts
from pyecharts.charts import Pie

# 讀原始數(shù)據(jù)文件
today = datetime.date.today().strftime('%Y%m%d')   #20200315
datafile = 'work/'+ today + '.json'
with open(datafile, 'r', encoding='UTF-8') as file:
    json_array = json.loads(file.read())

# 分析全國實(shí)時(shí)確診數(shù)據(jù)：'confirmedCount'字段
china_data = []
for province in json_array:
    china_data.append((province['provinceShortName'], province['confirmedCount']))
china_data = sorted(china_data, key=lambda x: x[1], reverse=True)                 #reverse=True,表示降序刃唐，反之升序

print(china_data)
# 全國疫情地圖

labels = [data[0] for data in china_data]
counts = [data[1] for data in china_data]

data_pair = [list(z) for z in zip(labels, counts)]

pie = Pie(init_opts=opts.InitOpts(width='900px',height='500px'))
pie.add(
    series_name="新冠病例統(tǒng)計(jì)",
    data_pair=data_pair,
    radius="55%",
    center=["30%", "70%"],
    label_opts=opts.LabelOpts(is_show=False,position="center")).set_global_opts(
        title_opts=opts.TitleOpts(title='全國實(shí)時(shí)確診數(shù)據(jù)',subtitle='數(shù)據(jù)來源：丁香園 '),
        legend_opts=opts.LegendOpts(is_show=False),).set_series_opts(
            tooltip_opts=opts.TooltipOpts(trigger="item", formatter="{a} <br/>: {c} (kkk3fzq%)"),
            label_opts=opts.LabelOpts(formatter="袭景: {c}",
        )
    )
pie.render(path='/home/aistudio/work/pie.html')

全國實(shí)時(shí)確診數(shù)據(jù)

Day 2：手勢識別

DNN網(wǎng)絡(luò)結(jié)構(gòu)（經(jīng)過N次訓(xùn)練后唁桩，測試集準(zhǔn)確率為0.982）
數(shù)據(jù)集地址：
https://aistudio.baidu.com/aistudio/datasetdetail/2182

#定義DNN網(wǎng)絡(luò)
class MyDNN(fluid.dygraph.Layer):
    def __init__(self):
        super(MyDNN,self).__init__()
        self.conv1 = Conv2D(num_channels=3,num_filters=6,filter_size=2,stride=2,act='relu')
        self.pool1 = Pool2D(pool_size=2,pool_type='max',pool_stride=2)
        self.conv2 = Conv2D(num_channels=6,num_filters=16,filter_size=3,stride=2,act='relu')
        self.pool2 = Pool2D(pool_size=2,pool_type='max',pool_stride=2)
        self.linear1 = Linear(16*6*6,256,act='relu')
        self.linear2 = Linear(256,10,act='softmax')
    def forward(self,input):
        x = self.conv1(input)
        x = self.pool1(x)
        x = self.conv2(x)
        x = self.pool2(x)
        x = fluid.layers.reshape(x,shape=[-1,16*6*6])
        x = self.linear1(x)
        x = fluid.layers.dropout(x,dropout_prob=0.7)
        y = self.linear2(x)
        return y

手勢識別

Day 3：車牌識別

CNN網(wǎng)絡(luò)結(jié)構(gòu)（經(jīng)過N次訓(xùn)練后，測試集準(zhǔn)確率為0.978）
數(shù)據(jù)集地址：
https://aistudio.baidu.com/aistudio/datasetdetail/23617

#定義網(wǎng)絡(luò)
class MyLeNet(fluid.dygraph.Layer):
    def __init__(self):
        super(MyLeNet,self).__init__()
        self.hidden1_1 = Conv2D(num_channels=1,num_filters=28,filter_size=5,stride=1,act='relu')
        self.hidden1_2 = Pool2D(2,pool_type='max',pool_stride=1)
        self.hidden2_1 = Conv2D(num_channels=28,num_filters=32,filter_size=3,stride=1,act='relu')
        self.hidden2_2 = Pool2D(2,pool_type='max',pool_stride=1)
        self.hidden3 = Conv2D(num_channels=32,num_filters=32,filter_size=3,stride=1,act='relu')
        self.hidden4 = Linear(32*10*10,65,act='softmax')
    def forward(self,input):
        x = self.hidden1_1(input)
        x = self.hidden1_2(x)
        x = self.hidden2_1(x)
        x = self.hidden2_2(x)
        x = self.hidden3(x)
        x = fluid.layers.reshape(x,shape=[-1,32*10*10])
        y = self.hidden4(x)
        return y

車牌識別

Day 4：口罩分類

VGG16網(wǎng)絡(luò)結(jié)構(gòu)（準(zhǔn)確率為1.0）
數(shù)據(jù)集地址：
https://aistudio.baidu.com/aistudio/datasetdetail/22392
https://aistudio.baidu.com/aistudio/datasetdetail/23615

class ConvPool(fluid.dygraph.Layer):
    '''卷積+池化'''
    def __init__(self,
                 num_channels,
                 num_filters,
                 filter_size,
                 pool_size,
                 pool_stride,
                 groups,
                 pool_padding=0,
                 pool_type='max',
                 conv_stride=1,
                 conv_padding=0,
                 act=None):
        super(ConvPool, self).__init__()  

        self._conv2d_list = []

        for i in range(groups):
            conv2d = self.add_sublayer(   #返回一個(gè)由所有子層組成的列表耸棒。
                'bb_%d' % i,
                fluid.dygraph.Conv2D(
                num_channels=num_channels[i], #通道數(shù)
                num_filters=num_filters,   #卷積核個(gè)數(shù)
                filter_size=filter_size,   #卷積核大小
                stride=conv_stride,        #步長
                padding=conv_padding,      #padding大小荒澡，默認(rèn)為0
                act=act)
            )
            self._conv2d_list.append(conv2d)   

        self._pool2d = fluid.dygraph.Pool2D(
            pool_size=pool_size,           #池化核大小
            pool_type=pool_type,           #池化類型，默認(rèn)是最大池化
            pool_stride=pool_stride,       #池化步長
            pool_padding=pool_padding      #填充大小
            )

    def forward(self, inputs):
        x = inputs
        for conv in self._conv2d_list:
            x = conv(x)
        x = self._pool2d(x)
        return x

class VGGNet(fluid.dygraph.Layer):
    '''
    VGG網(wǎng)絡(luò)
    '''
    def __init__(self):
        super(VGGNet, self).__init__()
        self.c2p1_64 = ConvPool(num_channels=[3,64],num_filters=64,filter_size=3,pool_size=2,pool_stride=2,groups=2,conv_stride=1,conv_padding=1,act='relu')
        self.c2p2_128 = ConvPool(num_channels=[64,128],num_filters=128,filter_size=3,pool_size=2,pool_stride=2,groups=2,conv_stride=1,conv_padding=1,act='relu')
        self.c3p3_256 = ConvPool(num_channels=[128,256,256],num_filters=256,filter_size=3,pool_size=2,pool_stride=2,groups=3,conv_stride=1,conv_padding=1,act='relu')
        self.c3p4_512 = ConvPool(num_channels=[256,512,512],num_filters=512,filter_size=3,pool_size=2,pool_stride=2,groups=3,conv_stride=1,conv_padding=1,act='relu')
        self.c3p5_512 = ConvPool(num_channels=[512,512,512],num_filters=512,filter_size=3,pool_size=2,pool_stride=2,groups=3,conv_stride=1,conv_padding=1,act='relu')
        self.linear1 = fluid.dygraph.Linear(512*7*7,4096,act='relu')
        self.linear2 = fluid.dygraph.Linear(4096,4096,act='relu')
        self.linear3 = fluid.dygraph.Linear(4096,2,act='softmax')
    def forward(self, inputs):
        """前向計(jì)算"""
        x = self.c2p1_64(inputs)
        x = self.c2p2_128(x)
        x = self.c3p3_256(x)
        x = self.c3p4_512(x)
        x = self.c3p5_512(x)
        x = fluid.layers.reshape(x,shape=[-1,512*7*7])
        x = self.linear1(x)
        x = fluid.layers.dropout(x,dropout_prob=0.5)
        x = self.linear2(x)
        x = fluid.layers.dropout(x,dropout_prob=0.5)
        y = self.linear3(x)
        return y

口罩分類

Day 5：PaddleHub體驗(yàn)

【比賽】人流密度檢測
數(shù)據(jù)集地址：https://aistudio.baidu.com/aistudio/datasetdetail/1917

Day 6：PaddleSlim模型壓縮

PaddleSlim代碼地址： https://github.com/PaddlePaddle/PaddleSlim
文檔地址：https://paddlepaddle.github.io/PaddleSlim/

#安裝paddleslim
#pip install paddleslim

#1. 導(dǎo)入依賴

import paddle
import paddle.fluid as fluid
import paddleslim as slim
import numpy as np

#2. 構(gòu)建模型
#該章節(jié)構(gòu)造一個(gè)用于對MNIST數(shù)據(jù)進(jìn)行分類的分類模型与殃，選用MobileNetV1单山，并將輸入大小設(shè)置為[1, 28, 28]碍现，輸出類別數(shù)為10。 為了方便展示示例米奸，我們在paddleslim.models下預(yù)定義了用于構(gòu)建分類模型的方法昼接，執(zhí)行以下代碼構(gòu)建分類模型：

use_gpu = fluid.is_compiled_with_cuda()
exe, train_program, val_program, inputs, outputs = slim.models.image_classification("MobileNet", [1, 28, 28], 10, use_gpu=use_gpu)
place = fluid.CUDAPlace(0) if fluid.is_compiled_with_cuda() else fluid.CPUPlace()

#3 定義輸入數(shù)據(jù)
#為了快速執(zhí)行該示例，我們選取簡單的MNIST數(shù)據(jù)悴晰，Paddle框架的paddle.dataset.mnist包定義了MNIST數(shù)據(jù)的下載和讀取慢睡。 代碼如下：

import paddle.dataset.mnist as reader
train_reader = paddle.batch(
        reader.train(), batch_size=128, drop_last=True)
test_reader = paddle.batch(
        reader.test(), batch_size=128, drop_last=True)
data_feeder = fluid.DataFeeder(inputs, place)
 
#4. 訓(xùn)練和測試
#先定義訓(xùn)練和測試函數(shù)，正常訓(xùn)練和量化訓(xùn)練時(shí)只需要調(diào)用函數(shù)即可铡溪。在訓(xùn)練函數(shù)中執(zhí)行了一個(gè)epoch的訓(xùn)練漂辐，因?yàn)镸NIST數(shù)據(jù)集數(shù)據(jù)較少，一個(gè)epoch就可將top1精度訓(xùn)練到95%以上棕硫。

def train(prog):
    iter = 0
    for data in train_reader():
        acc1, acc5, loss = exe.run(prog, feed=data_feeder.feed(data), fetch_list=outputs)
        if iter % 100 == 0:
            print('train iter={}, top1={}, top5={}, loss={}'.format(iter, acc1.mean(), acc5.mean(), loss.mean()))
        iter += 1
        
def test(prog):
    iter = 0
    res = [[], []]
    for data in test_reader():
        acc1, acc5, loss = exe.run(prog, feed=data_feeder.feed(data), fetch_list=outputs)
        if iter % 100 == 0:
            print('test iter={}, top1={}, top5={}, loss={}'.format(iter, acc1.mean(), acc5.mean(), loss.mean()))
        res[0].append(acc1.mean())
        res[1].append(acc5.mean())
        iter += 1
    print('final test result top1={}, top5={}'.format(np.array(res[0]).mean(), np.array(res[1]).mean()))
#調(diào)用train函數(shù)訓(xùn)練分類網(wǎng)絡(luò)髓涯，train_program是在第2步：構(gòu)建網(wǎng)絡(luò)中定義的

train(train_program)

train iter=0, top1=0.0390625, top5=0.46875, loss=2.89760327339
train iter=100, top1=0.9296875, top5=0.9921875, loss=0.196367427707
train iter=200, top1=0.96875, top5=1.0, loss=0.125141501427
train iter=300, top1=0.9609375, top5=0.9921875, loss=0.158306568861
train iter=400, top1=0.9375, top5=1.0, loss=0.178206339478

#調(diào)用test函數(shù)測試分類網(wǎng)絡(luò)，val_program是在第2步：構(gòu)建網(wǎng)絡(luò)中定義的哈扮。

test(val_program)

test iter=0, top1=0.9765625, top5=1.0, loss=0.0672305747867
final test result top1=0.962439894676, top5=0.998597741127

#5. 量化模型
#按照配置在train_program和val_program中加入量化和反量化op.

place = exe.place
# quant_program =         #請?jiān)诖螖?shù)添加你的代碼
# val_quant_program =     #請?jiān)诖螖?shù)添加你的代碼
quant_program = slim.quant.quant_aware(train_program, exe.place, for_test=False)
val_quant_program = slim.quant.quant_aware(val_program, exe.place, for_test=True)

2020-04-05 21:52:28,981-INFO: quant_aware config {'moving_rate': 0.9, 'weight_quantize_type': 'channel_wise_abs_max', 'is_full_quantize': False, 'dtype': 'int8', 'weight_bits': 8, 'window_size': 10000, 'activation_bits': 8, 'quantize_op_types': ['conv2d', 'depthwise_conv2d', 'mul'], 'not_quant_pattern': ['skip_quant'], 'activation_quantize_type': 'moving_average_abs_max', 'for_tensorrt': False}
2020-04-05 21:52:30,032-INFO: quant_aware config {'moving_rate': 0.9, 'weight_quantize_type': 'channel_wise_abs_max', 'is_full_quantize': False, 'dtype': 'int8', 'weight_bits': 8, 'window_size': 10000, 'activation_bits': 8, 'quantize_op_types': ['conv2d', 'depthwise_conv2d', 'mul'], 'not_quant_pattern': ['skip_quant'], 'activation_quantize_type': 'moving_average_abs_max', 'for_tensorrt': False}

#6 訓(xùn)練和測試量化后的模型?
#微調(diào)量化后的模型纬纪，訓(xùn)練一個(gè)epoch后測試。

train(quant_program)

train iter=0, top1=0.96875, top5=1.0, loss=0.0995958149433
train iter=100, top1=0.96875, top5=1.0, loss=0.0631555095315
train iter=200, top1=0.96875, top5=1.0, loss=0.100415810943
train iter=300, top1=0.9921875, top5=0.9921875, loss=0.0579719617963
train iter=400, top1=0.953125, top5=1.0, loss=0.122109659016

#測試量化后的模型滑肉，和3.2 訓(xùn)練和測試中得到的測試結(jié)果相比包各，精度相近，達(dá)到了無損量化赦邻。

test(val_quant_program)

test iter=0, top1=0.984375, top5=1.0, loss=0.0304987411946
final test result top1=0.973056912422, top5=0.99919873476

三髓棋、心得體會

課程難度適中，作業(yè)提供baseline惶洲，降低了學(xué)習(xí)的難度。在作業(yè)中手寫了DNN膳犹，LeNet恬吕，VGG16等網(wǎng)絡(luò)，并且通過修改網(wǎng)絡(luò)結(jié)構(gòu)和參數(shù)得到了較高的準(zhǔn)確率须床，是一次很好的實(shí)踐經(jīng)歷铐料。在課程的學(xué)習(xí)中，對深度學(xué)習(xí)的一些概念有了更深的理解豺旬，實(shí)踐的經(jīng)驗(yàn)對畢業(yè)設(shè)計(jì)也有很大幫助钠惩。例如可以通過dropout和正則化等方法解決過擬合問題；模型的準(zhǔn)確率不僅僅與網(wǎng)絡(luò)結(jié)構(gòu)有關(guān)族阅，還和batch_size篓跛，learning_rate等參數(shù)密不可分

四、PaddlePaddle打卡營預(yù)告

1坦刀、下一期打卡營

課程時(shí)間：4月22日
課程內(nèi)容：python和AI打卡營愧沟，適合無人工智能背景蔬咬，無編程基礎(chǔ)、無學(xué)習(xí)氛圍沐寺、無計(jì)算資源又想好好學(xué)習(xí)林艘、天天向上的純小白
課程簡介：基礎(chǔ)語法、進(jìn)階運(yùn)用混坞、深度學(xué)習(xí)工具和paddlehub 創(chuàng)意賽
課程報(bào)名：關(guān)注飛槳PaddlePaddle公眾號

2狐援、籌備中打卡營

1、論文復(fù)現(xiàn)營：以CV領(lǐng)域檢測任務(wù)究孕、GAN咕村、視頻分類等經(jīng)典方向的最新頂會論文（如，CVPR）為案例蚊俺，介紹論文研讀懈涛、框架搭建、模型優(yōu)化等方面內(nèi)容泳猬，帶你進(jìn)入科研大軍
2批钠、競賽輔導(dǎo)營：以CV領(lǐng)域研究最應(yīng)用最廣的目標(biāo)檢測為題，講解目標(biāo)檢測兩階段（R-CNN得封、SPP-Net埋心、Fast R-CNN、Faster R-CNN忙上、FPN拷呆、Mask-RCNN）和一階段方法（YOLO、SSD）疫粥，介紹大賽的塞梯解讀茬斧、數(shù)據(jù)分析、模型選取梗逮、模型優(yōu)化项秉、后處理調(diào)優(yōu)的方面經(jīng)驗(yàn)知識