Beginning Anomaly Detection Using

IRU

I

LQ UDQJH

 

QRUPDO QRUPDOLORF>QSUDQGRPSHUPXWDWLRQ OHQ QRUPDO @

GDWDBVHW SGFRQFDW >QRUPDO>@DQRPDOLHV@

[BWUDLQ[BWHVW WUDLQBWHVWBVSOLW GDWDBVHWWHVWBVL]H 

UDQGRPBVWDWH 

[BWUDLQ [BWUDLQVRUWBYDOXHV E\ > 7LPH @

[BWHVW [BWHVWVRUWBYDOXHV E\ > 7LPH @

\BWUDLQ [BWUDLQ>&ODVV@

\BWHVW [BWHVW>&ODVV@

[BWUDLQKHDG 

Figure B-39.  The creation of the training and testing data sets

appendix B   intro to pytorch

---

397

Figure B-40.  The output of the code in Figure 

B-39

The output should look somewhat like Figure 

B-40

.

appendix B   intro to pytorch

---

398

[BWUDLQ QSDUUD\ [BWUDLQ UHVKDSH [BWUDLQVKDSH>@

[BWUDLQVKDSH>@

[BWHVW QSDUUD\ [BWHVW UHVKDSH [BWHVWVKDSH>@

[BWHVWVKDSH>@

\BWUDLQ QSDUUD\ \BWUDLQ UHVKDSH \BWUDLQVKDSH>@

\BWHVW QSDUUD\ \BWHVW UHVKDSH \BWHVWVKDSH>@

SULQW

6KDSHV?Q[BWUDLQV?Q\BWUDLQV?Q [BWUDLQVKDSH

\BWUDLQVKDSH

SULQW

[BWHVWV?Q\BWHVWV?Q [BWHVWVKDSH

\BWHVWVKDSH

Figure B-41.  Reshaping the training and testing data sets so you can pass them 

into the model

After defining your data sets, you need to reshape the values so that your neural 

network can accept them (see Figure 

B-41

).

The output should look like Figure 

B-42

.

Figure B-42.  The output of the code in Figure 

B-41

appendix B   intro to pytorch

---

399

Now you can define your model (Figure 

B-43

 and Figure 

B-44

).