你有哪些deep learning(rnn、cnn)调参的经验?
hzwer
北京大学 旷视 算法 hzwer.com
https://www.zhihu.com/question/41631631/answer/859040970
有的同学在打比赛的时候,从头到尾只维护若干份代码,每次载入前一次的训练参数,改一下代码再炼,俗称老丹。这样会有几个问题:某次引入一个 bug,过了很久才发现,然后不知道影响范围;得到一个好模型,但是不知道它是怎么来的;忘了自己的 baseline,不知道改动是正面还是负面。
w = np.random.uniform(low=-scale, high=scale, size=[n_in,n_out])
-
Xavier初始法,适用于普通激活函数(tanh,sigmoid):scale = np.sqrt(3/n)
-
He初始化,适用于ReLU:scale = np.sqrt(6/n)
w = np.random.randn(n_in,n_out) * stdev # stdev为高斯分布的标准差,均值设为0
-
Xavier初始法,适用于普通激活函数 (tanh,sigmoid):stdev = np.sqrt(n)
-
He初始化,适用于ReLU:stdev = np.sqrt(2/n)
X -= np.mean(X, axis = 0) # zero-center
X /= np.std(X, axis = 0) # normalize
-
同样的参数,不同的初始化方式
-
不同的参数,通过cross-validation,选取最好的几组
-
同样的参数,模型训练的不同阶段,即不同迭代次数的模型。
-
不同的模型,进行线性融合. 例如RNN和传统模型.
-
Regularizing and Optimizing LSTM Language Models(LSTM 的训练技巧)
-
Massive Exploration of Neural Machine Translation Architectures(NMT 里各个超参的影响)
-
Training Tips for the Transformer Model(训练 Transformer 时会发生的各种现象)
-
RoBERTa: A Robustly Optimized BERT Pretraining Approach(BERT 预训练技巧,虽然跟大部分人没啥关系)
-
Training ImageNet in 1 Hour(大批量训练技巧)
-
Bag of Tricks for Image Classification with Convolutional Neural Networks(各种训练技巧集大成)
-
Bag of Freebies for Training Object Detection Neural Networks(同上)
-
EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks(当前对参数利用最有效的 CNN,类似地还有一篇 EfficientDet)
<pre style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"><p style="max-width: 100%;min-height: 1em;letter-spacing: 0.544px;white-space: normal;color: rgb(0, 0, 0);font-family: -apple-system-font, system-ui, "Helvetica Neue", "PingFang SC", "Hiragino Sans GB", "Microsoft YaHei UI", "Microsoft YaHei", Arial, sans-serif;widows: 1;line-height: 1.75em;box-sizing: border-box !important;overflow-wrap: break-word !important;"><strong style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="max-width: 100%;letter-spacing: 0.5px;font-size: 14px;box-sizing: border-box !important;overflow-wrap: break-word !important;"><strong style="max-width: 100%;font-size: 16px;letter-spacing: 0.544px;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="max-width: 100%;letter-spacing: 0.5px;box-sizing: border-box !important;overflow-wrap: break-word !important;"></span></strong></span></strong></p>
<pre style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"><section style="max-width: 100%;min-height: 1em;letter-spacing: 0.544px;white-space: normal;color: rgb(0, 0, 0);font-family: -apple-system-font, system-ui, "Helvetica Neue", "PingFang SC", "Hiragino Sans GB", "Microsoft YaHei UI", "Microsoft YaHei", Arial, sans-serif;widows: 1;line-height: 1.75em;box-sizing: border-box !important;overflow-wrap: break-word !important;"><strong style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="max-width: 100%;letter-spacing: 0.5px;font-size: 14px;box-sizing: border-box !important;overflow-wrap: break-word !important;"><strong style="max-width: 100%;font-size: 16px;letter-spacing: 0.544px;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="max-width: 100%;letter-spacing: 0.5px;box-sizing: border-box !important;overflow-wrap: break-word !important;">—</span></strong>完<strong style="max-width: 100%;font-size: 16px;letter-spacing: 0.544px;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="max-width: 100%;letter-spacing: 0.5px;font-size: 14px;box-sizing: border-box !important;overflow-wrap: break-word !important;"><strong style="max-width: 100%;font-size: 16px;letter-spacing: 0.544px;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="max-width: 100%;letter-spacing: 0.5px;box-sizing: border-box !important;overflow-wrap: break-word !important;">—</span></strong></span></strong></span></strong></section><section style="max-width: 100%;letter-spacing: 0.544px;white-space: normal;font-family: -apple-system-font, system-ui, "Helvetica Neue", "PingFang SC", "Hiragino Sans GB", "Microsoft YaHei UI", "Microsoft YaHei", Arial, sans-serif;widows: 1;box-sizing: border-box !important;overflow-wrap: break-word !important;"><section powered-by="xiumi.us" style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"><section style="margin-top: 15px;margin-bottom: 25px;max-width: 100%;opacity: 0.8;box-sizing: border-box !important;overflow-wrap: break-word !important;"><section style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"><section style="max-width: 100%;letter-spacing: 0.544px;box-sizing: border-box !important;overflow-wrap: break-word !important;"><section powered-by="xiumi.us" style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"><section style="margin-top: 15px;margin-bottom: 25px;max-width: 100%;opacity: 0.8;box-sizing: border-box !important;overflow-wrap: break-word !important;"><section style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"><section style="margin-bottom: 15px;padding-right: 0em;padding-left: 0em;max-width: 100%;min-height: 1em;color: rgb(127, 127, 127);font-size: 12px;font-family: sans-serif;line-height: 25.5938px;letter-spacing: 3px;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="max-width: 100%;color: rgb(0, 0, 0);box-sizing: border-box !important;overflow-wrap: break-word !important;"><strong style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="max-width: 100%;font-size: 16px;font-family: 微软雅黑;caret-color: red;box-sizing: border-box !important;overflow-wrap: break-word !important;">为您推荐</span></strong></span></section><section style="margin: 5px 32px;padding-right: 0em;padding-left: 0em;max-width: 100%;min-height: 1em;font-family: sans-serif;letter-spacing: 0px;opacity: 0.8;line-height: normal;box-sizing: border-box !important;overflow-wrap: break-word !important;">如何评价何恺明团队的最新工作RegNet?<br style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;" /></section><section style="margin-top: 5px;margin-bottom: 5px;padding-right: 0em;padding-left: 0em;max-width: 100%;min-height: 1em;font-family: sans-serif;letter-spacing: 0px;opacity: 0.8;line-height: normal;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="max-width: 100%;color: rgb(87, 107, 149);-webkit-tap-highlight-color: rgba(0, 0, 0, 0);cursor: pointer;font-size: 14px;box-sizing: border-box !important;overflow-wrap: break-word !important;">研究生从入门到放弃!不好意思老板,我这周没进展</span></section><section style="margin-top: 5px;margin-bottom: 5px;padding-right: 0em;padding-left: 0em;max-width: 100%;min-height: 1em;font-family: sans-serif;letter-spacing: 0px;opacity: 0.8;line-height: normal;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="max-width: 100%;color: rgb(87, 107, 149);-webkit-tap-highlight-color: rgba(0, 0, 0, 0);cursor: pointer;font-size: 14px;box-sizing: border-box !important;overflow-wrap: break-word !important;">有了这个神器,轻松用 Python 写个 App</span></section><section style="margin-top: 5px;margin-bottom: 5px;padding-right: 0em;padding-left: 0em;max-width: 100%;min-height: 1em;font-family: sans-serif;letter-spacing: 0px;opacity: 0.8;line-height: normal;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="color: rgb(87, 107, 149);-webkit-tap-highlight-color: rgba(0, 0, 0, 0);cursor: pointer;max-width: 100%;font-size: 14px;box-sizing: border-box !important;overflow-wrap: break-word !important;">MIT最新深度学习入门课,安排起来!</span></section><section style="margin-top: 5px;margin-bottom: 5px;padding-right: 0em;padding-left: 0em;max-width: 100%;min-height: 1em;font-family: sans-serif;letter-spacing: 0px;opacity: 0.8;line-height: normal;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="font-size: 14px;">一个AI PhD的毕业随感</span></section></section></section></section></section></section></section></section></section>
本篇文章来源于: 深度学习这件小事
本文为原创文章,版权归知行编程网所有,欢迎分享本文,转载请保留出处!
内容反馈