 *image source from [unsplash.com](https://images.unsplash.com/photo-1556546395-b63c28e30e86?ixlib=rb-1.2.1&ixid=eyJhcHBfaWQiOjEyMDd9&auto=format&fit=crop&w=1189&q=80) by Sergio souza* 众所周知,Tensorflow入门之所以困难,与其采用的Graph 和 Session 模式有关,这与原生的 Python 代码简单、直观的印象格格不入。同时,由于计算仅仅发生在Session里面,所以初始化参数和变量的时候没办法将结果打印出来,以至于调试起来也十分困难。 当然Google官方也意识到了这点,于是引入了Eager模式,在这个模式下tensorflow的常量和变量可以直接计算并打印出来,甚至还可以和numpy数组混合计算。本文代码参考官方教程(from [github](https://github.com/tensorflow/docs/blob/master/site/en/guide/eager.ipynb) with Apache License 2.0) 同样的,为了方便与读者交流,所有的代码都放在了这里: https://github.com/zht007/tensorflow-practice ### 1. 激活Eager模式 激活Eager模式也非常简单,仅几行代码。 ```python import tensorflow as tf tf.enable_eager_execution() tfe = tf.contrib.eager ``` 注意,eager模式在程序开始就要激活,且不能与普通模式混用。另外tfe在后面优化器(Optimizer)的时候需要用到,故先在这里定义了。 ### 2. Eger模式上手 Eger模式下,定义的变量或者常量可以直接打印出来 ```python a = tf.constant([[1, 2], [3, 4]]) print('a=',a) b = tf.Variable(np.zeros((2,2))) print('\n b=',b) c = tf.Variable([[6, 7], [8, 9]]) print('\n c=',c) -------output------- a= tf.Tensor( [[1 2] [3 4]], shape=(2, 2), dtype=int32) b= <tf.Variable 'Variable:0' shape=(2, 2) dtype=float64, numpy= array([[0., 0.], [0., 0.]])> c= <tf.Variable 'Variable:0' shape=(2, 2) dtype=int32, numpy= array([[6, 7], [8, 9]], dtype=int32)> ``` 可以直接转换成我们熟悉的numpy arrary ```python print(c.numpy()) ---output--- [[6 7] [8 9]] ``` 当然也可以直接计算并输出结果,甚至可以与numpy arrary 混合计算。 ```python x = tf.Variable([[6, 7], [8.0, 9.0]],dtype ="float32") y = np.array([[1,2], [3,4]],dtype ="float32") print(tf.matmul(x,y)) ----output---- tf.Tensor( [[27. 40.] [35. 52.]], shape=(2, 2), dtype=float32) ``` ### 3. Eager 模式下训练线性回归模型 最后我们用[Tensor Flow 在Eager模式下](https://steemit.com/cn-stem/@hongtao/tensorflow-keras)训练线性回归模型,该模型我们之前已经用Tensorflow和Keras训练过了,感兴趣的朋友可以参照[之前的文章](https://steemit.com/cn-stem/@hongtao/tensorflow-keras)进行对比。 #### 3.1 创建模拟数据 与之前的数据一样,此处数据是100万个带噪音的线性数据,100万个点用plt是画不出来的,图中随机采样了250个点  我们定义一个帮助函数方便以batch的形式这100万个数据点中随机抽取batch size大小的数据进行训练 ```python def next_batch(x_data, batch_size): batch_index = np.random.randint(len(x_data),size=(BATCH_SIZE)) x_train = x_data[batch_index] y_train = y_true[batch_index] return x_train, y_train ``` #### 3.2 定义变量 此处与普通模式下的tensorflow变量没有任何区别 ```python w_tfe = tf.Variable(np.random.uniform()) b_tfe = tf.Variable(np.random.uniform(1,10) ``` #### 3.3 线性函数 在普通模式下的tensorflow中我们需要定义计算图谱,这里我们直接以 python 函数的形式,定义要训练的线性回归函数。 ```python def linear_regression(inputs): return inputs * w_tfe + b_tfe ``` #### 3.4 损失函数 同样的,MS(Mean Square)损失函数也要以python 函数的形式定义,而不是计算图谱。 ```python def mean_square_fn(model_fn, inputs, labels): return tf.reduce_sum(tf.pow(model_fn(inputs) - labels, 2)) / (2 * BATCH_SIZE) ``` #### 3.5 优化器 同样使用Gradient Descent 的优化器,不同在于,普通模式下我们创建一个计算图谱train = optimizer.minimize(error), 在Eager模式下,要用tfe.implicit_gradients()来返回一个函数。 ```python optimizer = tf.train.GradientDescentOptimizer(learning_rate=0.001) grad = tfe.implicit_gradients(mean_square_fn) ``` #### 3.6 模型训练 由于没有计算图谱,所以也不需要初始化变量,也不需在Session下运行,而是类似于原生 Python 函数的形式将数据传入"Optimizer模型"函数。训练完成之后,w 和 b 的参数也自动保存下来,不必在Session中提取。 ```python for step in range(BATCHS): x_train, y_train = next_batch(x_data, BATCH_SIZE) optimizer.apply_gradients(grad(linear_regression, x_train, y_train)) ``` #### 3.7 验证训练结果 直接将最终的 w 和 b 带入线性函数,训练结果也非常符合预期。  ### 4. 总结 Eager 模式下的 Tensorflow 与原生的 Python 代码非常类似,可以直接计算并打印结果,创建和训练模型的过程也类似于python函数的创建和调用。Eager 和Keras API 都是Tensorflow 2.0 官方主推的 Tensorflow使用方式,相信在不久的将来,获取我们就再也看不到 init = tf.global_variables_initializer() 还有 with tf.Session() as sess:这样的类似八股文一样的关键词啦。 ------ 参考资料 [1] [Google Tensorflow 官方文档和教程](https://www.tensorflow.org/guide/eager?hl=zh-cn) [2] [Github TensorFlow-Examples](https://github.com/aymericdamien/TensorFlow-Examples) ------ 相关文章 [Tensorflow入门——线性回归](https://steemit.com/cn-stem/@hongtao/tensorflow) [Tensorflow入门——Keras简介和上手](https://steemit.com/cn-stem/@hongtao/tensorflow-keras) [AI学习笔记——Tensorflow入门](https://steemit.com/cn-stem/@hongtao/ai-tensorflow) ------ 同步到我的简书 https://www.jianshu.com/u/bd506afc6fc1
author | hongtao |
---|---|
permlink | tensorflow-eager-python |
category | cn-stem |
json_metadata | {"community":"busy","app":"busy/2.5.6","format":"markdown","tags":["cn-stem","ai","tensorflow","python","busy"],"users":["hongtao"],"links":["https://images.unsplash.com/photo-1556546395-b63c28e30e86?ixlib=rb-1.2.1&ixid=eyJhcHBfaWQiOjEyMDd9&auto=format&fit=crop&w=1189&q=80","https://github.com/tensorflow/docs/blob/master/site/en/guide/eager.ipynb","https://github.com/zht007/tensorflow-practice","https://steemit.com/cn-stem/@hongtao/tensorflow-keras","https://steemit.com/cn-stem/@hongtao/tensorflow-keras","https://www.tensorflow.org/guide/eager?hl=zh-cn","https://github.com/aymericdamien/TensorFlow-Examples","https://steemit.com/cn-stem/@hongtao/tensorflow","https://steemit.com/cn-stem/@hongtao/tensorflow-keras","https://steemit.com/cn-stem/@hongtao/ai-tensorflow"],"image":["https://images.unsplash.com/photo-1556546395-b63c28e30e86?ixlib=rb-1.2.1&ixid=eyJhcHBfaWQiOjEyMDd9&auto=format&fit=crop&w=1000&q=80","https://ws1.sinaimg.cn/large/006tNc79gy1g2nb3l0wo6j30ac07074k.jpg","https://ws4.sinaimg.cn/large/006tNc79gy1g2nbrf39krj30d709kwf6.jpg"]} |
created | 2019-05-02 14:37:06 |
last_update | 2019-05-02 14:37:06 |
depth | 0 |
children | 3 |
last_payout | 2019-05-09 14:37:06 |
cashout_time | 1969-12-31 23:59:59 |
total_payout_value | 1.666 HBD |
curator_payout_value | 0.505 HBD |
pending_payout_value | 0.000 HBD |
promoted | 0.000 HBD |
body_length | 4,484 |
author_reputation | 3,241,267,862,629 |
root_title | Tensorflow入门——Eager模式像原生Python一样简洁优雅 |
beneficiaries | [] |
max_accepted_payout | 1,000,000.000 HBD |
percent_hbd | 10,000 |
post_id | 84,132,670 |
net_rshares | 4,619,096,482,188 |
author_curate_reward | "" |
voter | weight | wgt% | rshares | pct | time |
---|---|---|---|---|---|
wackou | 0 | 77,357,149,577 | 1.29% | ||
tombstone | 0 | 211,675,608,943 | 0.86% | ||
delegate.lafona | 0 | 429,146,723,231 | 20% | ||
team | 0 | 82,330,099,404 | 10% | ||
lola-carola | 0 | 392,712,503 | 2.16% | ||
eric-boucher | 0 | 8,038,514,027 | 2.16% | ||
anwenbaumeister | 0 | 195,457,898 | 4.32% | ||
liberosist | 0 | 1,226,467,164 | 4.32% | ||
psygambler | 0 | 249,380,734 | 2.16% | ||
lemouth | 0 | 52,116,201,308 | 10% | ||
tumutanzi | 0 | 3,794,916,714 | 0.5% | ||
rwilday | 0 | 57,185,214 | 100% | ||
lamouthe | 0 | 11,623,827,265 | 20% | ||
lk666 | 0 | 816,876,473 | 2.16% | ||
justyy | 0 | 49,820,122,483 | 2.07% | ||
whoib | 0 | 618,062,553 | 70% | ||
curie | 0 | 1,407,224,152,120 | 4.32% | ||
hendrikdegrote | 0 | 58,436,159,911 | 4.32% | ||
vact | 0 | 85,097,127,341 | 4.32% | ||
steemstem | 0 | 801,385,967,597 | 20% | ||
dashfit | 0 | 449,824,680 | 2.16% | ||
gangstayid | 0 | 97,781,623 | 2.16% | ||
busy.org | 0 | 29,987,918 | 0.31% | ||
vodonik | 0 | 83,013,965 | 6.6% | ||
dna-replication | 0 | 5,597,496,767 | 20% | ||
gmedley | 0 | 314,250,993 | 2.16% | ||
coldhair | 0 | 1,154,450,671 | 0.5% | ||
diebaasman | 0 | 2,552,079,805 | 12% | ||
moksamol | 0 | 469,136,307 | 2.16% | ||
getrichordie | 0 | 142,578,588 | 2.16% | ||
thatsweeneyguy | 0 | 107,223,337 | 2.16% | ||
szokerobert | 0 | 116,955,348 | 0.86% | ||
bloom | 0 | 90,812,998,160 | 20% | ||
eurogee | 0 | 83,513,235 | 2% | ||
iansart | 0 | 996,104,257 | 2.16% | ||
jiujitsu | 0 | 1,927,744,575 | 2.16% | ||
lekang | 0 | 589,307,881 | 2.16% | ||
helo | 0 | 30,602,118,684 | 10% | ||
samminator | 0 | 9,210,582,603 | 10% | ||
zerotoone | 0 | 459,221,792 | 2.16% | ||
locikll | 0 | 3,601,506,708 | 8.64% | ||
mahdiyari | 0 | 17,086,143,122 | 10% | ||
lorenzor | 0 | 5,870,690,702 | 50% | ||
aboutyourbiz | 0 | 846,367,155 | 4.32% | ||
giuato | 0 | 107,617,147 | 2.16% | ||
alexander.alexis | 0 | 11,190,931,507 | 20% | ||
jonmagnusson | 0 | 151,676,432 | 1.08% | ||
jayna | 0 | 479,996,257 | 0.64% | ||
suesa | 0 | 87,199,466,218 | 25% | ||
cryptokrieg | 0 | 780,078,579 | 4.32% | ||
rival | 0 | 2,225,588,306 | 2% | ||
tensor | 0 | 2,510,252,275 | 2.16% | ||
slickhustler007 | 0 | 85,497,560 | 2.16% | ||
corsica | 0 | 9,460,150,297 | 20% | ||
makrotheblack | 0 | 102,423,531 | 2.16% | ||
trenz | 0 | 223,221,492 | 1.5% | ||
fancybrothers | 0 | 349,533,113 | 6% | ||
allcapsonezero | 0 | 2,107,177,770 | 2.16% | ||
howo | 0 | 28,337,169,991 | 10% | ||
tsoldovieri | 0 | 1,445,941,517 | 10% | ||
bluemoon | 0 | 1,134,411,379 | 4.32% | ||
nitego | 0 | 91,178,701 | 1.29% | ||
hotsteam | 0 | 3,623,694,130 | 10% | ||
neumannsalva | 0 | 628,033,372 | 2.16% | ||
wargof | 0 | 207,959,490 | 10% | ||
abigail-dantes | 0 | 341,651,031,896 | 20% | ||
phogyan | 0 | 87,319,355 | 2.16% | ||
esteemguy | 0 | 215,482,491 | 20% | ||
zonguin | 0 | 1,208,121,886 | 5% | ||
alexzicky | 0 | 5,932,255,999 | 5% | ||
mountain.phil28 | 0 | 3,582,968,323 | 25% | ||
jasonbu | 0 | 12,443,919,545 | 25% | ||
coolbuddy | 0 | 0 | 1% | ||
tuoficinavirtual | 0 | 99,223,106 | 25% | ||
iamphysical | 0 | 16,242,800,363 | 90% | ||
zest | 0 | 3,866,216,152 | 10% | ||
felixrodriguez | 0 | 666,683,701 | 10% | ||
revo | 0 | 1,665,649,601 | 2.16% | ||
azulear | 0 | 484,037,229 | 100% | ||
felicenavidad | 0 | 209,667,441 | 50% | ||
psicoluigi | 0 | 409,014,743 | 50% | ||
lmon | 0 | 4,531,772,145 | 50% | ||
jadabug | 0 | 1,512,474,735 | 1% | ||
massivevibration | 0 | 3,116,804,473 | 5% | ||
accelerator | 0 | 5,395,604,650 | 0.33% | ||
eurodale | 0 | 193,501,653 | 2.16% | ||
reaverza | 0 | 1,334,835,897 | 15% | ||
clweeks | 0 | 183,545,387 | 2.59% | ||
superbing | 0 | 728,282,273 | 7.24% | ||
dokter-purnama | 0 | 258,532,617 | 2.16% | ||
erikkun28 | 0 | 0 | 1% | ||
cryptononymous | 0 | 1,202,625,637 | 2.16% | ||
minloulou | 0 | 812,871,681 | 5% | ||
dailystats | 0 | 2,214,080,808 | 7.36% | ||
jlsplatts | 0 | 220,996,964 | 0.64% | ||
poodai | 0 | 167,869,234 | 2.16% | ||
markmorbidity | 0 | 99,174,135 | 2.16% | ||
cryptocurrencyhk | 0 | 330,439,111 | 20% | ||
emdesan | 0 | 139,789,457 | 10% | ||
peaceandwar | 0 | 642,466,869 | 2.16% | ||
enzor | 0 | 321,372,606 | 10% | ||
joendegz | 0 | 267,530,551 | 2.16% | ||
carloserp-2000 | 0 | 28,230,975,659 | 100% | ||
carlos84 | 0 | 4,458,437,326 | 100% | ||
gra | 0 | 7,643,424,864 | 20% | ||
jianan | 0 | 1,123,404,218 | 7.2% | ||
imisstheoldkanye | 0 | 1,791,127,272 | 1% | ||
shayekh2 | 0 | 100,835,566 | 50% | ||
pinksteam | 0 | 1,198,050,578 | 10% | ||
aalok | 0 | 104,897,626 | 26% | ||
dranren | 0 | 1,986,841,630 | 100% | ||
cnbuddy | 0 | 13,904,322,879 | 1% | ||
nicole-st | 0 | 182,459,244 | 2.16% | ||
teukurival | 0 | 180,791,756 | 2.16% | ||
drmake | 0 | 2,326,457,446 | 2.16% | ||
anxin | 0 | 108,840,954 | 7.21% | ||
guga34 | 0 | 476,310,013 | 15% | ||
pechichemena | 0 | 98,123,783 | 0.86% | ||
amestyj | 0 | 5,300,023,393 | 100% | ||
sandracarrascal | 0 | 148,702,417 | 100% | ||
bitinvdig0 | 0 | 741,435,155 | 24% | ||
skycae | 0 | 547,987,373 | 4.32% | ||
sireh | 0 | 85,266,735 | 0.43% | ||
egotheist | 0 | 200,607,286 | 2% | ||
kenadis | 0 | 5,135,971,106 | 20% | ||
esaia.mystic | 0 | 155,827,222 | 4.32% | ||
robotics101 | 0 | 1,903,793,808 | 20% | ||
tristan-muller | 0 | 78,517,832 | 20% | ||
gentleshaid | 0 | 3,969,019,679 | 10% | ||
thescubageek | 0 | 264,112,156 | 2.16% | ||
fejiro | 0 | 214,605,238 | 10% | ||
paradigmprospect | 0 | 151,029,161 | 0.32% | ||
nunesso | 0 | 35,151,657,584 | 50% | ||
danaedwards | 0 | 431,299,672 | 4.32% | ||
ivymalifred | 0 | 2,098,972,990 | 50% | ||
sco | 0 | 19,140,422,528 | 20% | ||
douglimarbalzan | 0 | 430,283,218 | 100% | ||
ennyta | 0 | 951,282,637 | 50% | ||
rharphelle | 0 | 159,503,147 | 25% | ||
gordon92 | 0 | 252,557,046 | 2.16% | ||
bitcoinportugal | 0 | 75,682,775 | 2.16% | ||
stahlberg | 0 | 945,312,802 | 2.16% | ||
gabrielatravels | 0 | 244,938,296 | 1.08% | ||
reizak | 0 | 333,701,362 | 1.72% | ||
zlatkamrs | 0 | 224,611,564 | 4.1% | ||
monie | 0 | 538,241,690 | 100% | ||
eliaschess333 | 0 | 10,712,242,059 | 50% | ||
shoganaii | 0 | 184,661,204 | 10% | ||
darkiche | 0 | 75,919,045 | 10% | ||
ydavgonzalez | 0 | 411,517,146 | 5% | ||
payger | 0 | 85,996,869 | 2.16% | ||
langford | 0 | 351,555,518 | 20% | ||
mattiarinaldoni | 0 | 0 | 1% | ||
hijosdelhombre | 0 | 2,035,414,647 | 2.16% | ||
mathowl | 0 | 5,040,260,302 | 10% | ||
shinedojo | 0 | 477,096,681 | 4.32% | ||
hongtao | 0 | 2,438,736,305 | 97% | ||
gaming.yer | 0 | 503,193,440 | 100% | ||
bennettitalia | 0 | 298,163,803 | 1.08% | ||
maiyude | 0 | 136,172,128 | 0.5% | ||
suesa-random | 0 | 12,772,518,772 | 50% | ||
steem-familia | 0 | 501,542,356 | 100% | ||
lacher-prise | 0 | 189,586,262 | 10% | ||
terrylovejoy | 0 | 3,662,932,854 | 8% | ||
jcalero | 0 | 142,514,033 | 4.32% | ||
wisewoof | 0 | 118,762,343 | 2.16% | ||
neneandy | 0 | 4,778,899,691 | 4.32% | ||
olajidekehinde | 0 | 80,976,594 | 10% | ||
real2josh | 0 | 151,833,906 | 10% | ||
gribouille | 0 | 566,514,456 | 20% | ||
traviseric | 0 | 246,771,337 | 50% | ||
woolfe19861008 | 0 | 86,269,221 | 7.21% | ||
yrmaleza | 0 | 370,940,422 | 50% | ||
stemng | 0 | 6,574,719,798 | 10% | ||
mininthecity | 0 | 156,395,563 | 3.45% | ||
edprivat | 0 | 1,796,706,086 | 0.15% | ||
trixie | 0 | 80,250,128 | 10% | ||
kingabesh | 0 | 505,309,239 | 10% | ||
bobandtom | 0 | 431,658,322 | 50% | ||
evangelista.yova | 0 | 493,634,557 | 100% | ||
miguelangel2801 | 0 | 788,472,159 | 50% | ||
dailychina | 0 | 2,070,412,954 | 7.23% | ||
didic | 0 | 2,520,377,112 | 2.16% | ||
jenniferjulieth | 0 | 422,712,588 | 100% | ||
operahoser | 0 | 317,459,819 | 0.64% | ||
asonintrigue | 0 | 136,428,431 | 2.16% | ||
emiliomoron | 0 | 5,511,215,976 | 50% | ||
dexterdev | 0 | 2,123,463,165 | 10% | ||
intellihandling | 0 | 2,319,485,897 | 50% | ||
oghie | 0 | 638,006,144 | 50% | ||
geopolis | 0 | 1,285,099,864 | 20% | ||
ajfernandez | 0 | 387,374,824 | 100% | ||
dongfengman | 0 | 609,554,485 | 7.21% | ||
robertbira | 0 | 2,124,050,973 | 5% | ||
bearded-benjamin | 0 | 58,878,939,189 | 50% | ||
teekingtv | 0 | 426,292,182 | 5% | ||
alexdory | 0 | 1,006,839,740 | 8% | ||
vegan.niinja | 0 | 244,752,516 | 2.16% | ||
flugschwein | 0 | 4,620,580,745 | 19% | ||
benleemusic | 0 | 106,498,108 | 0.43% | ||
cyprianj | 0 | 967,076,649 | 20% | ||
shentrading | 0 | 98,323,351 | 0.9% | ||
francostem | 0 | 2,715,049,321 | 20% | ||
ivan-g | 0 | 545,267,362 | 2.16% | ||
endopediatria | 0 | 692,879,897 | 20% | ||
croctopus | 0 | 1,538,148,981 | 100% | ||
ingmarvin | 0 | 437,739,779 | 100% | ||
joelagbo | 0 | 103,757,234 | 5% | ||
emmanuel293 | 0 | 99,142,956 | 25% | ||
cryptofuwealth | 0 | 131,069,439 | 11% | ||
djoi | 0 | 106,819,891 | 5% | ||
ethanlee | 0 | 158,161,775 | 6.1% | ||
morbyjohn | 0 | 60,595,646 | 7% | ||
alix96 | 0 | 422,894,880 | 100% | ||
ambitiouslife | 0 | 234,335,582 | 2.16% | ||
positiveninja | 0 | 461,746,050 | 2.16% | ||
tomastonyperez | 0 | 12,619,182,387 | 50% | ||
bil.prag | 0 | 125,638,059 | 0.21% | ||
jingis07 | 0 | 168,589,647 | 2.16% | ||
elvigia | 0 | 10,329,058,756 | 50% | ||
scoora82 | 0 | 838,896,858 | 24% | ||
qberry | 0 | 2,058,861,025 | 2.16% | ||
gabyoraa | 0 | 80,225,364 | 2.16% | ||
lesmouths-travel | 0 | 967,917,750 | 13% | ||
ezravandi | 0 | 2,787,373,385 | 1% | ||
cjunros | 0 | 111,792,862 | 2.16% | ||
markko | 0 | 10,868,091 | 50% | ||
wales | 0 | 488,567,051 | 2.16% | ||
effofex | 0 | 2,158,454,909 | 10% | ||
luiscd8a | 0 | 1,058,795,316 | 80% | ||
eniolw | 0 | 264,695,799 | 5% | ||
de-stem | 0 | 8,115,521,177 | 19.8% | ||
elsll | 0 | 102,355,909 | 4.32% | ||
geadriana | 0 | 675,377,390 | 15% | ||
elpdl | 0 | 537,018,340 | 100% | ||
derbesserwisser | 0 | 177,178,821 | 100% | ||
serylt | 0 | 3,912,670,443 | 19.6% | ||
bavi | 0 | 135,621,748 | 2.16% | ||
hiddenblade | 0 | 386,778,016 | 3.45% | ||
josedelacruz | 0 | 6,010,726,451 | 50% | ||
joseangelvs | 0 | 2,332,948,251 | 100% | ||
viannis | 0 | 1,237,574,073 | 50% | ||
majapesi | 0 | 246,859,215 | 50% | ||
menoski | 0 | 1,215,372,130 | 5% | ||
erickyoussif | 0 | 2,188,628,879 | 100% | ||
michaelwrites | 0 | 227,773,973 | 10% | ||
deholt | 0 | 818,839,483 | 17% | ||
archaimusic | 0 | 135,820,511 | 10% | ||
smacommunity | 0 | 162,466,700 | 2.16% | ||
musicvoter | 0 | 4,059,479,432 | 1% | ||
edanya | 0 | 87,286,055 | 2.16% | ||
goodway | 0 | 144,741,742 | 1% | ||
ntowl | 0 | 87,958,936 | 1.29% | ||
temitayo-pelumi | 0 | 1,499,521,955 | 20% | ||
andrick | 0 | 454,903,416 | 50% | ||
sweet-jenny8 | 0 | 1,204,623,845 | 7.21% | ||
yusvelasquez | 0 | 965,744,404 | 50% | ||
winkandwoo | 0 | 407,589,265 | 100% | ||
doctor-cog-diss | 0 | 330,758,285 | 20% | ||
alexworld | 0 | 492,273,467 | 25% | ||
gracelbm | 0 | 200,580,377 | 2.16% | ||
acont | 0 | 245,068,822 | 50% | ||
niouton | 0 | 177,965,414 | 0.86% | ||
blockurator | 0 | 76,245,199 | 0.43% | ||
elimao | 0 | 429,865,285 | 100% | ||
schroders | 0 | 1,633,379,775 | 1.29% | ||
anaestrada12 | 0 | 23,160,823,894 | 100% | ||
steemzeiger | 0 | 985,243,506 | 19.8% | ||
yorgermadison | 0 | 349,080,745 | 100% | ||
alexjunior | 0 | 370,862,623 | 100% | ||
somegaming | 0 | 91,231,835 | 4.32% | ||
moneybaby | 0 | 1,025,644,253 | 5% | ||
cryptocopy | 0 | 229,653,492 | 2.16% | ||
antunez25 | 0 | 429,776,745 | 100% | ||
haf67 | 0 | 382,145,955 | 100% | ||
chavas | 0 | 464,832,349 | 100% | ||
longer | 0 | 344,015,576 | 50% | ||
blewitt | 0 | 1,336,732,709 | 0.3% | ||
kafupraise | 0 | 133,393,318 | 34% | ||
biomimi | 0 | 188,011,142 | 40% | ||
ibk-gabriel | 0 | 120,765,827 | 10% | ||
scrawly | 0 | 605,850,556 | 4.32% | ||
drsensor | 0 | 1,525,647,680 | 8% | ||
mirzantorres | 0 | 395,199,177 | 50% | ||
mrnightmare89 | 0 | 1,934,008,944 | 26% | ||
ilovecryptopl | 0 | 565,846,701 | 3.45% | ||
purelyscience | 0 | 116,555,959 | 10% | ||
eglinson | 0 | 329,680,433 | 100% | ||
uzcateguiazambra | 0 | 447,747,497 | 100% | ||
yomismosoy | 0 | 357,959,770 | 50% | ||
casiloko | 0 | 247,536,860 | 50% | ||
bflanagin | 0 | 368,917,920 | 2.16% | ||
ubaldonet | 0 | 4,594,453,365 | 65% | ||
asmeira | 0 | 512,894,221 | 100% | ||
garrillo | 0 | 337,400,026 | 100% | ||
lillywilton | 0 | 1,023,911,115 | 20% | ||
yestermorrow | 0 | 1,992,115,159 | 6% | ||
mary11 | 0 | 542,200,806 | 75% | ||
laiyuehta | 0 | 102,194,895 | 5.18% | ||
hansmast | 0 | 313,601,777 | 2.16% | ||
turtlegraphics | 0 | 625,282,759 | 7.22% | ||
wallyt | 0 | 74,939,128 | 1.72% | ||
wstanley226 | 0 | 73,170,933 | 50% | ||
reinaseq | 0 | 5,633,326,001 | 100% | ||
yaelg | 0 | 2,534,123,415 | 5% | ||
pfernandezpetit | 0 | 352,680,190 | 100% | ||
mgarrillogonzale | 0 | 399,749,621 | 100% | ||
rubenp | 0 | 537,575,803 | 100% | ||
jeferc | 0 | 536,410,262 | 100% | ||
steeming-hot | 0 | 0 | 0.01% | ||
tommasobusiello | 0 | 241,791,270 | 2.16% | ||
clement.poiret | 0 | 239,393,156 | 4.32% | ||
ogsenti | 0 | 69,835,439 | 100% | ||
fran.frey | 0 | 1,814,451,462 | 50% | ||
emsteemians | 0 | 103,181,337 | 10% | ||
perpetuum-lynx | 0 | 424,208,382 | 19.6% | ||
jrevilla | 0 | 151,831,397 | 50% | ||
annaabi | 0 | 277,267,462 | 2.16% | ||
emperorhassy | 0 | 534,444,758 | 10% | ||
moniroy | 0 | 1,616,202,568 | 50% | ||
skorup87 | 0 | 22,087,143 | 12% | ||
trang | 0 | 379,242,547 | 2.16% | ||
stem-espanol | 0 | 80,825,896,549 | 100% | ||
praditya | 0 | 1,645,093,060 | 24% | ||
aleestra | 0 | 2,003,053,578 | 100% | ||
rhethypo | 0 | 134,823,405 | 2.16% | ||
antigourmet | 0 | 155,324,418 | 2.16% | ||
predict-crypto | 0 | 94,659,692 | 0.08% | ||
chickenmeat | 0 | 225,649,531 | 2.16% | ||
javier.dejuan | 0 | 4,815,079,748 | 20% | ||
witnesstools | 0 | 601,452,107 | 7.22% | ||
sciencetech | 0 | 118,074,815 | 2% | ||
hirally | 0 | 449,610,894 | 100% | ||
emynb | 0 | 342,436,500 | 100% | ||
fanta-steem | 0 | 501,152,128 | 30% | ||
reverseacid | 0 | 317,352,293 | 2.16% | ||
giulyfarci52 | 0 | 1,096,583,605 | 50% | ||
eu-id | 0 | 620,154,223 | 10% | ||
ilovecoding | 0 | 596,691,830 | 7.22% | ||
votes4minnows | 0 | 855,414,473 | 1.5% | ||
andohyara | 0 | 127,707,290 | 5% | ||
ambercookie | 0 | 85,261,206 | 90% | ||
alvin0617 | 0 | 339,013,501 | 2.16% | ||
solarphasing | 0 | 354,210,103 | 5% | ||
stem.witness | 0 | 17,755,981,220 | 20% | ||
sarhugo | 0 | 61,997,614 | 2.16% | ||
hdu | 0 | 855,794,005 | 1% | ||
steemexpress | 0 | 2,129,855,660 | 3.73% | ||
steemfuckeos | 0 | 376,297,223 | 7.22% | ||
alex-hm | 0 | 1,575,160,669 | 50% | ||
wilmer14molina | 0 | 1,545,873,228 | 100% | ||
zerofive | 0 | 1,687,102,659 | 100% | ||
eugenialobo | 0 | 511,459,328 | 100% | ||
ballesteroj | 0 | 371,550,435 | 100% | ||
jcmontilva | 0 | 471,057,568 | 100% | ||
rodriguezr | 0 | 394,348,040 | 100% | ||
marbely20 | 0 | 442,959,793 | 100% | ||
moyam | 0 | 538,301,503 | 100% | ||
emilycg | 0 | 388,315,509 | 100% | ||
darys | 0 | 472,716,535 | 100% | ||
sibaja | 0 | 480,237,345 | 100% | ||
balcej | 0 | 538,225,018 | 100% | ||
lmanjarres | 0 | 457,111,866 | 100% | ||
anaka | 0 | 539,359,694 | 100% | ||
benhurg | 0 | 538,307,612 | 100% | ||
judisa | 0 | 539,997,982 | 100% | ||
juddarivv | 0 | 538,300,408 | 100% | ||
mariamo | 0 | 418,359,315 | 100% | ||
kimmorales | 0 | 462,914,249 | 100% | ||
loraine25 | 0 | 442,568,637 | 100% | ||
kingnosa | 0 | 91,667,196 | 50% | ||
cameravisual | 0 | 6,856,076,023 | 50% | ||
amin-ove | 0 | 129,404,135 | 50% | ||
goodcontentbot | 0 | 64,053,326 | 50% | ||
huilco | 0 | 537,950,102 | 100% | ||
hanyseek | 0 | 52,256,945 | 50% | ||
herculean | 0 | 56,785,290 | 50% | ||
naythan | 0 | 94,379,295 | 2.16% | ||
combatsports | 0 | 1,595,476,398 | 4.32% | ||
electrodo | 0 | 5,142,373,192 | 1.68% | ||
jent | 0 | 342,345,548 | 70% | ||
steemean | 0 | 317,568,009 | 10% | ||
lorasia | 0 | 162,575,304 | 50% | ||
donasys | 0 | 57,276,692 | 50% | ||
mtfmohammad | 0 | 99,183,152 | 25% | ||
bewithbreath | 0 | 832,830,162 | 2.5% | ||
chrisluke | 0 | 134,194,104 | 26% | ||
joannar | 0 | 114,617,288 | 25% | ||
pflanzenlilly | 0 | 248,232,033 | 50% | ||
sapphire.app | 0 | 1,927,059,242 | 50% | ||
nicephoto | 0 | 2,842,570,804 | 3.45% | ||
naturalproducts | 0 | 699,211,173 | 25% | ||
gutenmorganism | 0 | 1,017,395,558 | 50% | ||
faberleggenda | 0 | 539,998,604 | 100% | ||
mohaaking | 0 | 142,705,471 | 50% | ||
gustavoagt | 0 | 248,570,461 | 97% | ||
vaccinusveritas | 0 | 6,740,826,768 | 50% | ||
moz333 | 0 | 319,241,568 | 2.16% | ||
patris | 0 | 127,794,753 | 2.16% | ||
lionsmane | 0 | 0 | 1% | ||
alfatron777 | 0 | 204,714,034 | 1.72% | ||
sembdelgado | 0 | 75,832,657 | 50% | ||
cojp | 0 | 70,074,358 | 50% | ||
descalante | 0 | 147,867,454 | 50% | ||
map0226 | 0 | 3,224,850,168 | 100% |
你那里天气如何?想一展歌喉吗?好声音@cn-voice欢迎你~假如我的留言打扰到你,请回复“取消”。
author | cnbuddy |
---|---|
permlink | re-hongtao-tensorflow-eager-python-20190502t150346320z |
category | cn-stem |
json_metadata | "" |
created | 2019-05-02 15:03:48 |
last_update | 2019-05-02 15:03:48 |
depth | 1 |
children | 0 |
last_payout | 2019-05-09 15:03:48 |
cashout_time | 1969-12-31 23:59:59 |
total_payout_value | 0.000 HBD |
curator_payout_value | 0.000 HBD |
pending_payout_value | 0.000 HBD |
promoted | 0.000 HBD |
body_length | 50 |
author_reputation | -1,449,160,991,441 |
root_title | Tensorflow入门——Eager模式像原生Python一样简洁优雅 |
beneficiaries | [] |
max_accepted_payout | 1,000,000.000 HBD |
percent_hbd | 10,000 |
post_id | 84,133,976 |
net_rshares | 0 |
Congratulations @hongtao! You have completed the following achievement on the Steem blockchain and have been rewarded with new badge(s) : <table><tr><td>https://steemitimages.com/60x70/http://steemitboard.com/@hongtao/voted.png?201905041412</td><td>You received more than 8000 upvotes. Your next target is to reach 9000 upvotes.</td></tr> </table> <sub>_You can view [your badges on your Steem Board](https://steemitboard.com/@hongtao) and compare to others on the [Steem Ranking](http://steemitboard.com/ranking/index.php?name=hongtao)_</sub> <sub>_If you no longer want to receive notifications, reply to this comment with the word_ `STOP`</sub> **Do not miss the last post from @steemitboard:** <table><tr><td><a href="https://steemit.com/steemmeetupaachen/@steemitboard/steemitboard-to-support-the-german-speaking-community-meetups"><img src="https://steemitimages.com/64x128/https://cdn.steemitimages.com/DQmeoNp9iCaCfd2D6TqnWa3Aky2mU4Fm3xaSmjTM91YoNBS/image.png"></a></td><td><a href="https://steemit.com/steemmeetupaachen/@steemitboard/steemitboard-to-support-the-german-speaking-community-meetups">SteemitBoard to support the german speaking community meetups</a></td></tr></table> > You can upvote this notification to help all Steem users. Learn how [here](https://steemit.com/steemitboard/@steemitboard/http-i-cubeupload-com-7ciqeo-png)!
author | steemitboard |
---|---|
permlink | steemitboard-notify-hongtao-20190504t144532000z |
category | cn-stem |
json_metadata | {"image":["https://steemitboard.com/img/notify.png"]} |
created | 2019-05-04 14:45:30 |
last_update | 2019-05-04 14:45:30 |
depth | 1 |
children | 0 |
last_payout | 2019-05-11 14:45:30 |
cashout_time | 1969-12-31 23:59:59 |
total_payout_value | 0.000 HBD |
curator_payout_value | 0.000 HBD |
pending_payout_value | 0.000 HBD |
promoted | 0.000 HBD |
body_length | 1,354 |
author_reputation | 38,975,615,169,260 |
root_title | Tensorflow入门——Eager模式像原生Python一样简洁优雅 |
beneficiaries | [] |
max_accepted_payout | 1,000,000.000 HBD |
percent_hbd | 10,000 |
post_id | 84,247,499 |
net_rshares | 0 |
<div class='text-justify'> <div class='pull-left'> <center> <br /> <img width='200' src='https://res.cloudinary.com/drrz8xekm/image/upload/v1553698283/weenlqbrqvvczjy6dayw.jpg'> </center> <br/> </div> This post has been voted on by the **SteemSTEM** curation team and voting trail. It is elligible for support from <b><a href='https://www.steemstem.io/#!/@curie'>@curie</a></b>.<br /> If you appreciate the work we are doing, then consider supporting our witness [**stem.witness**](https://steemconnect.com/sign/account_witness_vote?approve=1&witness=stem.witness). Additional witness support to the [**curie witness**](https://steemconnect.com/sign/account_witness_vote?approve=1&witness=curie) would be appreciated as well.<br /> For additional information please join us on the [**SteemSTEM discord**]( https://discord.gg/BPARaqn) and to get to know the rest of the community!<br /> Please consider setting <b><a href='https://www.steemstem.io/#!/@steemstem'>@steemstem</a></b> as a beneficiary to your post to get a stronger support.<br /> Please consider using the <b><a href='https://www.steemstem.io'>steemstem.io</a></b> app to get a stronger support.</div>
author | steemstem |
---|---|
permlink | re-hongtao-tensorflow-eager-python-20190504t123841985z |
category | cn-stem |
json_metadata | {"app":"bloguable-bot"} |
created | 2019-05-04 12:38:45 |
last_update | 2019-05-04 12:38:45 |
depth | 1 |
children | 0 |
last_payout | 2019-05-11 12:38:45 |
cashout_time | 1969-12-31 23:59:59 |
total_payout_value | 0.000 HBD |
curator_payout_value | 0.000 HBD |
pending_payout_value | 0.000 HBD |
promoted | 0.000 HBD |
body_length | 1,174 |
author_reputation | 262,017,435,115,313 |
root_title | Tensorflow入门——Eager模式像原生Python一样简洁优雅 |
beneficiaries | [] |
max_accepted_payout | 1,000,000.000 HBD |
percent_hbd | 10,000 |
post_id | 84,241,623 |
net_rshares | 0 |