WHITE PAPER RNN

Size: px

Start display at page:

Download "WHITE PAPER RNN"

そよひろもり
4 years ago
Views:

1 WHITE PAPER RNN

2 ii RNN? ARIMA RNN Recurrent Neural Network RNN RNN RNN RNN SAS Viya RNN Appendix A. CAS Server B. CAS Action Python API SWAT DLPy C

3 1 1 SAS Viya RNN Recurrent Neural Network ARIMA RNN RNN SAS Viya RNN 2 RNN? 2.1 ARIMA RNN ARIMA AR Autoregressive model x t = a 1 x t 1 + a 2 x t t x t 1 x t 2 x t x t 1 x t 2 x t AR ARIMA ARIMA AR ARIMA

4 2 S t S t+1 S t+2 x t1 x t+1 x t+2 AR (a 1, a 2 ) 1 ARIMA RNN 3 y = 1 f 1 (x) + 2 f 2 (x) +... x y y = f 1 ( f 2 ( f 3 (x))) 1 2 3

5 3 入力層隠れ層出力層 x 1 x 2 y x 3 ノード x D 1 層 2 層 3 層 4 層 RNN Recurrent Neural Network RNN RNN 隠れ層 x t 1 x t Montufar, Guido F., et al. "On the number of linear regions of deep neural networks." Advances in neural information processing systems. 2014

6 4 x t 1 x t h t h t+1 h t+2 h t+10 x t+10 x t x t+1 x t RNN Recurrent Neural Network RNN 隠れ層入力出力 x t h t h 展開 t h t+1 h t+2 x t x t+1 x t+2

7 5 RNN = RNN RNN LSTM Long Short-Term Memory LSTM RNN LSTM RNN LSTM LSTM ブロック入力ゲート出力ゲート活性化活性化 x t CEC 関数関数 h t 忘却ゲート LSTM LSTM 3 RNN RNN RNN RNN

8 6 3.1 RNN RNN RNN ARIMA ( ARIMA RNN) RNN RNN RNN RNN RNN 3.2 RNN RNN RNN RNN 3 36 RNN RNN RNN RNN ARIMA RNN 6 1 =1 1 =12 1 =1 1 =365

9 7 RNN RNN RNN RNN 3.3 RNN RNN RNN RNN RNN RNN RNN RNN RNN RNN 時系列モデリングにおける RNN の適用可能性判定学習オブザベーション数が多い Yes No ビジネス的にブラックボックス化されたモデルが許容される RNN は適さない Yes No RNN の長所が現れ一般的に検証できた場合例えば言語処理等 RNN は適さない Yes No RNN が適している RNN と他のモデルを比較し精度の高いモデルを採用すべき

10 8 RNN RNN RNN 4 SAS Viya RNN SAS Viya CAS Cloud Analytic Services CAS REST API Python R Jupyter Notebook Python SAS Viya Python SWAT DLPy SWAT DLPy SAS Viya Python SAS REST API Python SAS Viya SAS Viya SWAT DLPy Appendix SAS kwh SAS Viya RNN 4 Solar_Power 1 Global_Horizontal_Irradiance Hour 24 Solar_Elevation_Angle Day_Time_Indicator RNN

11 9 4.1 RNN CAS csv Pandas 'Solar_ Power' 26 RNN SAS Viya CAS Table from_ pandas() Pandas SAS Viya CAS Table Python Pandas Python SAS Viya

$timeseries_partion() 8 7 {x1, x 2, x 3,.$

12 10 SAS Viya RNN DLPy prepare_ subsequences() DLPy timeseries_partion() 8 7 {x1, x 2, x 3,...} {x 11, x 12, x 13, x 14 } {x 11, x 12, x 13, x 14 } 1 8

13 11 SAS data step SAS data step 4.2 RNN 2 RNN LSTM SAS Viya RNN 2 15 RNN Sequential() solar_forecast add() "std=" STD STD 0 1 LSTM "output_type=" samelength "init=" XAVIER 9 2 encoding "reversed=" 1 2 "reversed=false" "act=" IDENTITY 10 "error=" NORMAL 9 Glorot, X., and Bengio, Y. Understanding the difficulty of training deep feedforward neural networks. In Proceedings of the thirteenth international conference on artificial intelligence and statistics, pp ,

14 12 RNN RNN RNN RNN ( ) 4

13 RNN mini_batch_size RNN 4 max_epochs 150 150 ADAM AdamSolver() ADAM learning_rate 0.

15 13 RNN mini_batch_size RNN 4 max_epochs ADAM AdamSolver() ADAM learning_rate 0.01 learning_rate_policy step gamma 0.2 step_size gamma 30 crip_grad

14 12 input_length 5 target_length token_size token_size 1 1 4 token_size 5 sequence_opt 3

16 14 12 input_length 5 target_length token_size token_size token_size 5 sequence_opt 3 prepare_subsequences() token_size train() "seed=" 1234 SAS Viya GPU GPU CPU GPU CPU train() "gpu=1" GPU RNN SAS Viya

15 4.3 1 score() "init_weights=" "copy_ vars=" 'Solar_Power'

17 score() "init_weights=" "copy_ vars=" 'Solar_Power' 'datetime' solar_ forecast_score_out score() "gpu=1" GPU RNN _DL_Pred_2

18 16 WAPE 13 RMSE 14 WAPE=6.17% WAPE 13 WAPE: Weighted Absolute Percentage Error % 0 14 RMSE: Root Mean Squared Error 2 1 1

19 17 5 SAS Viya RNN Recurrent Neural Network ARIMA RNN RNN SAS Viya RNN SAS Viya RNN RNN DLPy DLPy

20 18 Appendix A. CAS Server SAS Viya SAS Cloud Analytic Services CAS CAS SMP/MPP SAS Viya GPU SAS Viya CAS CAS # 1 import swat import dlpy sess = swat.cas( host, port, user, password) image_table = dlpy.imagetable.load_files( sess, path='./imgs/', casout={'name':'my_images'}) swat dlpy CAS CAS # 2 from dlpy import applications vgg16model = applications.vgg16( sess, pre_trained_weights=true, pre_trained_weights_file="./vgg_ilsvrc_16_layers.caffemodel.h5", include_top=true) CAS result # 3 result = vgg16model.predict(image_table)

21 19 Jupyter CAS ❶ ❷ ❸ Memory / Storate Memory Viya Client (Jupyter) Viya Server ❶ ローカルにあるファイルやデータを読み込んで CAS サーバーに格納する (# コード 1) ❷ CAS サーバー上にモデルを構築する (# コード 2) ❸ 分析を実行し結果を取得する (# コード 3) B. CAS Action Python API SWAT DLPy CAS CAS WebAPI WebAPI REST Python WebAPI SWAT Scripting Wrapper for Analytics Transfer SWAT WebAPI Python API API REST API CAS Action Action ActionSet ImageActionSet DeeplearnActionSet Action SWAT ActionSet Actionset Action SWAT SWAT DLPy DLPy SWAT DeeplearnActionSet Keras PyTorch Pythonic DNN ONNX SWAT CAS Action ActionSet SWAT DLPy

22 20 Client App DLPy Viya Client (Jupyter) deeplearn SWAT Image addlayer augumentimages buildmodel compareimages : : ActionSet Action Viya Server CAS C. SAS Viya Getting Started with SAS Viya 3.4 for Python &docsetid=caspg3&docsettarget=titlepage.htm&locale=en SWAT API SWAT DLPy API DLPy DLPy An Introduction to SAS Viya 3.4 Programming &docsetid=pgmdiff&docsettarget=titlepage.htm&locale=en

23 21 SAS Cloud Analytic Services 3.4: Fundamentals setid=casfun&docsettarget=titlepage.htm&locale=ja Deep Learning Action Set API setid=casdlpg&docsettarget=cas-deeplearn-tblofactions.htm&locale=ja