fa786e5f2a8016725991cf2d02eebe6368dc04b3,dipy/reconst/dti.py,,wls_fit_tensor,#Any#Any#,1154

Before Change

    tol = 1e-6
    data = np.asarray(data)
    data_flat = data.reshape((-1, data.shape[-1]))
    dti_params = np.empty((len(data_flat), 4, 3))

    //obtain OLS fitting matrix
    //U,S,V = np.linalg.svd(design_matrix, False)
    //math: beta_ols = inv(X.T*X)*X.T*y
    //math: ols_fit = X*beta_ols*inv(y)
    //ols_fit = np.dot(U, U.T)
    ols_fit = _ols_fit_matrix(design_matrix)
    min_diffusivity = tol / -design_matrix.min()

    for param, sig in zip(dti_params, data_flat):
        param[0], param[1:] = _wls_iter(ols_fit, design_matrix, sig,
                                        min_diffusivity)

    dti_params.shape = data.shape[:-1] + (12,)
    return dti_params


def _wls_iter(ols_fit, design_matrix, sig, min_diffusivity):
    """ Helper function used by wls_fit_tensor.

After Change

    log_s = np.log(data)
    w = np.exp(np.einsum("...ij,...j", ols_fit, log_s))
    evals, evecs = decompose_tensor(
        from_lower_triangular(
            np.einsum("...ij,...j",
                      pinv_vec(design_matrix * w[..., None]),
                      w * log_s)
        ),
        min_diffusivity=tol / -design_matrix.min()
    )

    dti_params = np.concatenate((evals[..., None, :], evecs), axis=-2)
    return dti_params.reshape(data.shape[:-1] + (12,))

Instances