Loss¶
- TS_PCA.loss.SumVarifoldLoss(K_lst)¶
Generate a sum of Varifold loss function related to the given varifold kernels
- Parameters:
K_lst (list of kernel functions (X,mask_X,Y,mask_Y,b)-> array of the shape of b) – \(K(X,Y)b\) where X and Y are array of size (n_samples,d+1), \(K(X,Y)\) is the kernel matrix \((k(x_i,y_j))\) and b is an array of shape (n_samples,d) with d the dimension of the problem
- Returns:
- Sum of Varifoldloss(s_x,mask_s_x,t_x,mask_t_x)->float
where s_x,t_x are time series graph (array of size (n_samples,d+1)) and mask_s_x,mask_t_x are time series graph mask (array of size (n_samples,1))
- TS_PCA.loss.VarifoldLoss(K)¶
Generate Varifold loss function related to the given varifold kernel
- Parameters:
K (kernel function (X,mask_X,Y,mask_Y,b)-> array of the shape of b) – \(K(X,Y)b\) where X and Y are array of size (n_samples,d+1), \(K(X,Y)\) is the kernel matrix \((k(x_i,y_j))\) and b is an array of shape (n_samples,d) with d the dimension of the problem
- Returns:
- Varifoldloss(s_x,mask_s_x,t_x,mask_t_x)->float
where s_x,t_x are time series graph (array of size (n_samples,d+1)) and mask_s_x,mask_t_x are time series graph mask (array of size (n_samples,1))