ridge.fit1 <- glmnet(x,y, alpha = 0)  
  
ridge.fitt <- coef(glmnet(x, y, alpha = 0, lambda = 3))
ridge.pred <- predict(ridge.fit1, s = 3, type = 'coefficients')

What's the difference between ridge.fitt and ridge.pred?