Beginning Anomaly Detection Using

IRU

HSRFK

LQ

UDQJH QXPBHSRFKV 

IRU

L LPDJHVODEHOV 

LQ HQXPHUDWH

WUDLQBORDGHU 

LPDJHV LPDJHVWR GHYLFH

ODEHOV ODEHOVWR GHYLFH

)RUZDUGSDVV

RXWSXWV PRGHO LPDJHV

ORVV FULWHULRQ RXWSXWVODEHOV

%DFNZDUGDQGRSWLPL]H

RSWLPL]HU]HURBJUDG

ORVVEDFNZDUG

RSWLPL]HUVWHS

LI

L   

SULQW

(SRFK>^`^`@6WHS>^`^`@/RVV^I` 

IRUPDW HSRFKQXPBHSRFKVLWRWDOBVWHS

ORVVLWHP

Figure 3-63.  You initialize the model, your loss function, and your optimizer, and 

then you start the training process

Chapter 3   IntroduCtIon to deep LearnIng

---

119

It might take a while, but you should see something like Figure 

3-64

.

After training is done, you can test your model and find the AUC score (Figure 

3-65

).

Figure 3-64.  The output of the training process

Chapter 3   IntroduCtIon to deep LearnIng

---

120