This paper explores the connection between quantum computing and kernel methods in machine learning, focusing on the idea of encoding inputs into quantum states as a nonlinear feature map. The authors interpret this process as mapping data to a quantum Hilbert space, where a quantum computer can analyze the input data. They discuss two approaches for building quantum models for classification: the implicit approach, where a quantum device estimates inner products of quantum states to compute a kernel function, and the explicit approach, where a variational quantum circuit is used as a linear model in the Hilbert space. The paper illustrates these ideas with a feature map based on squeezing in a continuous-variable system and visualizes the working principle using 2-dimensional mini-benchmark datasets. The authors also investigate the linear separability of data in the feature Hilbert space and provide proofs for the linear independence of feature vectors under the squeezing feature map.This paper explores the connection between quantum computing and kernel methods in machine learning, focusing on the idea of encoding inputs into quantum states as a nonlinear feature map. The authors interpret this process as mapping data to a quantum Hilbert space, where a quantum computer can analyze the input data. They discuss two approaches for building quantum models for classification: the implicit approach, where a quantum device estimates inner products of quantum states to compute a kernel function, and the explicit approach, where a variational quantum circuit is used as a linear model in the Hilbert space. The paper illustrates these ideas with a feature map based on squeezing in a continuous-variable system and visualizes the working principle using 2-dimensional mini-benchmark datasets. The authors also investigate the linear separability of data in the feature Hilbert space and provide proofs for the linear independence of feature vectors under the squeezing feature map.