SVIG/python-bootcamp
7
0
Fork 1
Code Issues Pull Requests 1 Releases Wiki Activity

Update page 'Solutions to the questions concerning R language'

xuanyi.wang
2023-05-22 21:03:13 +02:00
parent 8230bd20bf
commit 9ffe79af0a
1 changed files with 4 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
Solutions to the questions concerning R language.md
+4 -2
@@ -378,8 +378,7 @@ summary(tune.out)
## 2 0.01778 0.3675 0.05076
## 3 0.03162 0.3575 0.05177
## 4 0.05623 0.3425 0.04937
## 5 0.10000 0.3150 0.05231
## 6 0.17783 0.2487 0.03929
****## 6 0.17783 0.2487 0.03929
## 7 0.31623 0.2088 0.05684
## 8 0.56234 0.2088 0.05653
## 9 1.00000 0.2000 0.06095
@@ -421,4 +420,7 @@ table(OJ.test$Purchase, test.pred)
```
## [1] 0.1741
```
Tuning reduces the training error to 15.12% and test error to 17.4% which is worse than radial kernel but slightly better than linear kernel.
**viii.**
Overall, radial basis kernel seems to be producing minimum misclassification error on both train and test data.