marcorealacci/master-degree-notes
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

vault backup: 2025-01-04 16:30:37

Browse source
This commit is contained in:
Marco Realacci 2025-01-04 16:30:37 +01:00
parent c4a223f25d
commit 89bdca19a6
8 changed files with 125 additions and 62 deletions
Download patch file Download diff file Expand all files Collapse all files

2
Foundation of data science/notes/3.1 Multi Class Logistic Regression.md
View file

@ -46,7 +46,7 @@ sample x1 e x2
- $f(x2)=[0.01, 0.98, 0.01]$ (il modello per questo sample invece è molto sicuro)
Calcoliamo la Cross Entropy!
$$-[[1(-0.91)+0(-1.6)+0(-0.91)]+[0(-4.6)+0(-0.02)+0(-4.6)]] = 0.93$$
$$-[[1(-0.91)+0(-1.6)+0(-0.91)]+[0(-4.6)+1(-0.02)+0(-4.6)]] = 0.93$$
quei numeri strani sono semplicemente i logaritmi di 0.4, 0.2 ecc., abbiamo semplicemente applicato la formula sopra!
##### Softmax optimization