Information output program, information output method, and information processing device

Abstract

To enable provision of active space information that contribute to quantum chemistry computation.SOLUTION: An information output program disclosed herein causes a computer to perform processes of: acquiring occupancy data including a time series of occupancy of each of multiple molecular orbits, performing principal component analysis of the occupancy data, and outputting information on an active space corresponding to a subset of the multiple molecular orbits used for quantum chemistry computation on the basis of a result of the principal component analysis.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an information output program, an information output method, and an information processing device. [Background technology]

[0002] The molecular orbital method is known as one of the approximation methods for quantum chemical calculations. In the molecular orbital method, the electron orbitals that extend throughout the molecule, known as molecular orbitals, are approximately constructed by linear combinations of the electron orbitals of each atom, known as atomic orbitals.

[0003] For example, in the Hartree-Fock (HF) method, the wave functions and orbital energies of molecular orbitals can be calculated using successive approximations. In the HF model, electrons are accommodated in orbitals in order of decreasing orbital energy.

[0004] Of these, the orbital occupied by an electron with the highest energy is called the HOMO (Highest Occupied Molecular Orbital), and the empty orbital with the lowest energy is called the LUMO (Lowest Unoccupied Molecular Orbital).

[0005] In the molecular orbital method, in order to reduce the amount of calculation, instead of using a set of all molecular orbitals in calculations by an optimization method such as a variational calculation, a subset of molecular orbitals, a so-called "active space," may be designated and a variational calculation for optimization may be performed.

[0006] One known method for specifying such an active space is the "HOMO-m / LUMO+n" type, in which m orbitals are specified in descending order from the HOMO and n orbitals are specified in ascending order from the LUMO. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] International Publication No. 2022 / 097298 [Patent Document 2] Japanese Patent Application Laid-Open No. 2012-32908 [Patent Document 3] US Patent Application Publication No. 2020 / 0349459 [Patent Document 4] US Patent Application Publication No. 2016 / 0378955 Summary of the Invention [Problem to be solved by the invention]

[0008] However, since there are no clear rules for determining the active space, it is left to the user's subjective judgment, and therefore the specification of the active space may not be appropriate in terms of the accuracy and amount of calculation of the quantum chemical calculation.

[0009] In one aspect, the present invention aims to provide an information output program, an information output method, and an information processing device that can provide information on an active space that contributes to quantum chemical calculations. [Means for solving the problem]

[0010] An information output program according to one aspect causes a computer to execute the following processes: acquire occupancy data including a time series of occupancy numbers of each of a plurality of molecular orbitals; perform principal component analysis on the occupancy data; and output information on an active space corresponding to a subset of the plurality of molecular orbitals used for quantum chemical calculations based on the results of the principal component analysis. [Effects of the Invention]

[0011] According to one embodiment, it is possible to provide information on the active space that is useful for quantum chemical calculations. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a block diagram illustrating an example of the functional configuration of a server device. [Figure 2]FIG. 2 is a schematic diagram illustrating an example of a molecular orbital. [Figure 3] FIG. 3 is a schematic diagram illustrating an example of an active space. [Figure 4] FIG. 4 is a schematic diagram illustrating one aspect of the problem-solving approach. [Figure 5] FIG. 5 is a flowchart showing the procedure of the process of generating the occupation number data. [Figure 6] FIG. 6 is a flowchart showing the procedure of the information output process. [Figure 7] FIG. 7 is a diagram illustrating an example of a hardware configuration. DETAILED DESCRIPTION OF THE INVENTION

[0013] Hereinafter, examples for implementing the information output program, information output method, and information processing device according to the present disclosure will be described with reference to the accompanying drawings. Note that this example merely illustrates one example or aspect, and the structure, action, function, properties, characteristics, methods, uses, etc. according to the present disclosure are not limited by such examples.

[0014] Example 1 Fig. 1 is a block diagram showing an example of the functional configuration of a server device 10. Fig. 1 illustrates the server device 10 that provides an information output function for realizing the provision of information on active spaces that contribute to quantum chemical calculations.

[0015] <Terminology> Before describing the functional configuration example of the server device 10 shown in FIG. 1, some of the terms related to the information output function in the field of quantum chemistry will be explained below.

[0016] (1)Molecular orbital method In the field of quantum chemistry, solving the Schrödinger equation inevitably becomes a many-body problem because the objects of interest are various compounds. Therefore, since it is not realistic to obtain an exact solution, various approximation methods are introduced. The molecular orbital method is one such approximation method and is one of the fundamental concepts of current quantum chemical calculations.

[0017] In the molecular orbital method, the electron orbital (molecular orbital) that extends throughout the molecule is approximately constructed by a linear combination of the electron orbitals (atomic orbitals) of each atom. In the most basic HF method, the wave function (φ i ) and orbital energy (ε i ) can be obtained using the successive approximation method. Note that "i" may refer to the index of the molecular orbital.

[0018] FIG. 2 is a schematic diagram illustrating an example of a molecular orbital. For example, FIG. 2 shows eight molecular orbitals with i=1 to 8 corresponding to orbital energies (ε1) to (ε8). As shown in FIG. 2, electrons are arranged in each of the eight molecular orbitals with i=1 to 8 in ascending order of orbital energy. Up to two electrons can be accommodated per molecular orbital. In this case, the spin state when two electrons are accommodated is limited to antiparallel by the Pauli exclusion factor.

[0019] (2) Number of occupants The number of electrons placed in each molecular orbital is called the "occupancy number." In the HF model, the occupation number can only take integer values ​​of 0, 1, or 2, but in post-HF models such as the CCSD (Coupled Cluster Singles and Doubles) method, the occupation number can take real values ​​between 0 and 2 for a single molecular orbital.

[0020] (3) HOMO and LUMO In the HF model, electrons are accommodated in orbitals in ascending orbital energy order, with the orbital with the highest energy occupied by an electron being called the highest occupied molecular orbital (HOMO), while the orbital with the lowest energy unoccupied is called the lowest unoccupied molecular orbital (LUMO).

[0021] (4)Active space In molecular orbital theory, expectation values ​​of physical quantities such as orbital energies can be calculated using optimization techniques such as calculus of variations. While strictly speaking, all molecular orbitals can be used for optimization, it is common to narrow down the scope by selecting a subset of the orbitals in order to reduce the computational complexity. This narrowed-down subset of molecular orbitals is called the "active space."

[0022] Fig. 3 is a schematic diagram illustrating an example of an active space. Similar to Fig. 2, Fig. 3 shows molecular orbitals corresponding to orbital energies (ε1) to (ε8). Furthermore, Fig. 3 shows the molecular orbitals corresponding to the active space among the eight molecular orbitals, distinguished by hatching.

[0023] As shown in Figure 3, of the eight molecular orbitals i = 1 to 8, the five molecular orbitals i = 1 to 5 have electrons assigned to them. Of these, the molecular orbital i = 5 with the highest orbital energy ε5 is the HOMO. On the other hand, of the three molecular orbitals i = 6 to 8 with no electrons assigned to them, the one with the lowest orbital energy ε6 is the LUMO.

[0024] For example, in the example shown in FIG. 3, the active space is exemplified as a range that combines one molecular orbital (i=4) in descending order from the HOMO and one molecular orbital (i=7) in ascending order from the LUMO, that is, a subset of four molecular orbitals, i=4 to 7.

[0025] <One aspect of the issue> As explained in the Background Art section above, there are no clear rules for determining the active space, and it is left to the user's subjective judgment. Therefore, the specification of the active space may not be appropriate in terms of the accuracy and amount of calculation of the quantum chemical calculation.

[0026] In other words, the user's subjective judgment is left to estimate the range of molecular orbitals arranged in descending order from the HOMO that will have a large impact on the accuracy of quantum chemical calculations, and further, the range of molecular orbitals arranged in ascending order from the LUMO that will have a large impact on the accuracy of quantum chemical calculations.

[0027] However, due to constraints such as computational resources and computational time, the permissible range of (m, n) can only be determined empirically, and it is not easy even for experts to determine which molecular orbitals obtained by HF calculations etc. have the greatest impact on the accuracy of quantum chemical calculations.

[0028] For example, even if experts predict interactions between molecular orbitals using various information on molecular orbitals obtained from tools for visualizing molecular orbitals or HF calculations, it is difficult to determine the impact that individual molecular orbitals will have on quantum chemical calculations.

[0029] Such an active space specified subjectively by the user may not be appropriate in terms of the accuracy and amount of calculation required for quantum chemical calculations.

[0030] <One aspect of the problem-solving approach> Therefore, the information output function according to this embodiment acquires occupancy data including a time series in the optimization process of occupancy vectors corresponding to the occupancies of each of a plurality of molecular orbitals, and outputs information on the active space based on the results of applying principal component analysis to the occupancy data.

[0031] Here, the above-mentioned occupation number data may be pre-convergence data obtained in a process such as variational optimization in molecular orbital methods. This is because the variational calculation has an objective function such as energy minimization, and even during convergence, it contains characteristic changes that are in line with that objective. Note that although variational calculation is used as an example here, optimization may also be applied to calculations other than variational calculations, and therefore the term "variational calculation, etc." may be used to include variational calculations used in optimization and similar calculations.

[0032] For example, it is unlikely that the occupancy of a low-energy occupied orbital or a high-energy unoccupied orbital will change from 2 or 0. On the other hand, the closer an orbital is to the HOMO / LUMO boundary, the closer its energy value is, so the more frequently the occupancy changes, resulting in variations over time.

[0033] By applying principal component analysis to such occupancy data, which reduces high-dimensional data to low-dimensional data by extracting features that best express the overall variation, it becomes possible to distinguish between molecular orbitals that have a large impact on the accuracy of quantum chemical calculations and those that do not.

[0034] Figure 4 is a schematic diagram illustrating one aspect of the problem-solving approach. Figure 4 shows a graph in which data points corresponding to an occupancy vector whose elements are the occupancy numbers x1 to x3 of three molecular orbitals (i = 1 to 3) are plotted with black dots for each iteration of a variational calculation, etc. Furthermore, Figure 4 shows data axes t1 to t3 corresponding to the first to third principal components obtained as a result of applying principal component analysis.

[0035] As shown in Figure 4, the data group of occupancy vectors is concentrated and distributed on a two-dimensional plane formed by two data axes, t1 and t2. This makes it clear that the variation in the data group of three-dimensional occupancy vectors, X1 to X3, can be well expressed using only two axes, (t1, t2).

[0036] For the sake of convenience, an example in which the number of dimensions, i.e., the number of molecular orbitals, is "3" is shown in Fig. 4. However, compared to such a three-dimensional space, the occupation number vector that can be obtained from an actual process such as variational optimization can be higher dimensional.

[0037] However, even if the dimension of the occupation vector increases, the tendency that the occupation numbers of low-energy occupied orbitals and high-energy unoccupied orbitals hardly change remains, so it is clear that a similar variation in dimension concentration occurs.

[0038] In this way, it is possible to identify molecular orbitals where changes in occupancy occur frequently based on the data axis on which variations are concentrated, and it is therefore possible to output information on molecular orbitals that have a large impact on the accuracy of quantum chemical calculations and those that do not.

[0039] On the one hand, it is obvious that the accuracy of calculations can be improved by selecting, as the active space, molecular orbitals that have a large influence on the accuracy of quantum chemical calculations among all molecular orbitals. Furthermore, even if the number of molecular orbitals specified in the active space is reduced due to constraints such as computational resources and calculation time, it is also obvious that the deterioration of calculation accuracy can be suppressed by selecting, as the active space, molecular orbitals that have a large influence on the accuracy of quantum chemical calculations.

[0040] Therefore, the information output function according to this embodiment can provide information on active spaces that contribute to quantum chemical calculations in various aspects, such as calculation accuracy, calculation resources, and calculation time. By providing such information on active spaces, it is possible to achieve a balance between the accuracy and required time of quantum chemical calculations in the molecular orbital method, even without advanced specialized knowledge of quantum chemical calculations or proficiency with quantum chemical calculation software. Therefore, it is possible to automate the specification of active spaces while eliminating dependency on individual users.

[0041] <Overall structure> FIG. 1 shows, as just one example of a use case, an example in which the server device 10 provides the above-mentioned information output function based on occupation number data obtained in the process of a variational calculation of a quantum chemical calculation executed by the client terminal 30.

[0042] The server device 10 is an example of an information processing device that provides the above-mentioned information output function. For example, the server device 10 can be realized as a SaaS (Software as a Service) type application. This allows the above-mentioned information output function to be provided as a cloud service. In addition, the server device 10 does not prevent the above-mentioned information output function from being provided on-premise.

[0043] The client terminal 30 is an example of a computer that receives the information output function. Users of this information output function may be anyone involved in performing quantum chemical calculations using molecular orbital methods. For example, they may include employees of manufacturers of chemical products, pharmaceuticals, etc., or experts such as developers.

[0044] <Configuration of client terminal 30> Next, an example of the functional configuration of the client terminal 30 according to this embodiment will be described. In Fig. 1, blocks related to functions related to the function of generating occupancy number data that the client terminal 30 has are diagrammatically illustrated.

[0045] 1, the client terminal 30 includes a receiving unit 31, a quantum chemistry calculation unit 33, and an output unit 35. Note that Fig. 1 only illustrates an extract of functional units related to the function corresponding to the above-described function of generating occupancy number data, and the client terminal 30 may be provided with functional units other than those illustrated.

[0046] The receiving unit 31 is a processing unit that receives various requests. For example, the receiving unit 31 can receive a request to execute a quantum chemical calculation via a user interface (not shown).

[0047] When receiving such a request, the receiving unit 31 can receive input of "compound data" that expresses the three-dimensional structure of the molecule that is the target of the quantum chemical calculation. For example, the compound data may include the types of atoms that constitute the compound and the XYZ coordinates of the atoms. Furthermore, input of "specified conditions" such as parameters to be used when performing the quantum chemical calculation, such as the number of iterations and the active space, may be received.

[0048] The quantum chemistry calculation unit 33 is a processing unit that executes quantum chemistry calculations. In one embodiment, the quantum chemistry calculation unit 33 can generate the above-mentioned occupancy number data by executing software that realizes quantum chemistry calculations in accordance with the above-mentioned compound data and the above-mentioned specified conditions. Such quantum chemistry calculation software may be any existing software, regardless of whether it is open source or from a specific vendor.

[0049] Here, the quantum chemistry calculation performed by the quantum chemistry calculation unit 33 may be distinguished from the quantum chemistry calculation performed according to the specification of the active space after the active space is determined, for the following reasons, and the algorithms and parameters used by the two may be different.

[0050] In one aspect, the quantum chemistry calculations executed by the quantum chemistry calculation unit 33 do not need to be repeated until the optimization of the variation calculations in the molecular orbital method or the like converges. From this aspect, as one of the above-mentioned specified conditions, the number of iterations of the variation calculations or the like can be specified as any number equal to or greater than 1 iteration.

[0051] In another aspect, the quantum chemistry calculation performed by the quantum chemistry calculation unit 33 is sufficient in terms of calculation accuracy to the extent that orbital energies can be calculated. From this perspective, one of the above-mentioned specified conditions may be to specify an algorithm that is faster than the quantum chemistry calculation performed after the active space is determined. For example, the active space to be applied to the VQE (Variational Quantum Eigensolver) method may be obtained from the calculation process of a post-HF model, such as the faster CCSD method.

[0052] Another aspect is that in scenarios where the scale of a compound or other factors requires a large amount of calculation for all molecular orbitals in order to balance the computational resources, it may be impossible to perform even a single iteration of a variational calculation. In such cases, one of the above-mentioned conditions can be specified as a HOMO-m / LUMO+n type active space. Additionally, by specifying domain decomposition, such as DMET (Density Matrix Embedding Theory), it is possible to obtain the results of a single iteration of a variational calculation.

[0053] The output unit 35 is a functional unit that outputs various types of information. In one aspect, the output unit 35 can display, output as sound, or print out the active space information output by the information output unit 19 of the server device 10.

[0054] Although the example of presenting active space information to the user has been given here as one aspect of information output, it goes without saying that the active space information can also be output to software or services that perform quantum chemical calculations according to the active space specification. In this case, it is also possible to skip the user's confirmation or editing of the active space and have the quantum chemical calculations performed.

[0055] <Configuration of Server Device 10> Next, an example of the functional configuration of the server device 10 according to this embodiment will be described. In Fig. 1, blocks related to functions related to the information output function of the server device 10 are diagrammatically shown.

[0056] 1, the server device 10 includes a data acquisition unit 11, a PCA (Principal Component Analysis) execution unit 13, an importance calculation unit 15, a trajectory selection unit 17, and an information output unit 19. Note that Fig. 1 only shows a selection of functional units related to the functions corresponding to the information output function described above, and the server device 10 may include functional units other than those shown in the figure.

[0057] The data acquisition unit 11 is a processing unit that acquires the above-mentioned occupancy number data. In one embodiment, the data acquisition unit 11 can acquire the above-mentioned occupancy number data from the results of quantum chemistry calculations performed by the quantum chemistry calculation unit 33 of the client terminal 30. The acquisition of such occupancy number data may be performed on demand, or may be performed automatically in cooperation with the quantum chemistry calculation unit 33 of the client terminal 30.

[0058] The PCA execution unit 13 is a processing unit that executes principal component analysis (PCA). Such PCA may be implemented using any variation of a model, including a linear model, and may be executed using any software for multivariate analysis. In one embodiment, the PCA execution unit 13 executes PCA on the occupancy data acquired by the data acquisition unit 11, as described below.

[0059] That is, following the idea of ​​principal component analysis (PCA), we consider a data matrix X∈R consisting of n data points with p attributes. n×p Consider compressing the attribute dimension of data point x from p to q (q≦p). i =(x i1 ,x i2 ,…,x ip ), where i = 1, 2, ..., n, we assume the use of an occupancy vector containing the occupancies of p orbitals.

[0060] The purpose of principal component analysis is to extract components that best represent the characteristics of the structure of the original data by some kind of linear transformation. Let T∈R be the transformed data matrix. n×p , the factor loading matrix representing the linear transformation for compression is W∈R p×p Then, there is a relationship between them: T=XW. The vector t after data transformation i =(t i1 ,t i2 ,…,t ip ), i=1, 2, ..., n are the principal component scores, and T is called the principal component score matrix. Each matrix can be explicitly expressed as the following equations (1) to (3).

[0061]

number

[0062] In the above dimensionality reduction, the data point x i =(x i1 ,x i2 ,…,x ip ), i=1, 2, ..., n, we require that we do not lose as much information as possible about the variation of the principal component score vector t i We find and apply a linear transformation W that maximizes the sample variance of . This is equivalent to determining a q-dimensional space that best preserves the appearance of the original data points distributed in p-dimensional space.

[0063] According to the theory of principal component analysis (PCA), the transformation matrix Wq for such dimensionality reduction can be obtained by solving the eigenvalue problem of the variance-covariance matrix S of p attribute variables, i.e., the following equation (4):

[0064]

number

[0065] where σ ij = (ij=1, 2, ..., p) is the value of the covariance when the ith and jth attribute variables are viewed as random variables. When the eigenvalue of the square matrix S is λ and the eigenvector is w (which may be normalized to |w| = 1), the solution to the equation Sw = λw can be easily found using existing technology.

[0066] The eigenvectors corresponding to these eigenvalues ​​are arranged in order of magnitude and are called λ 1、 λ2,…,λ p (λ1 ≥ λ2,…,≥ λ p ) and w 1、 w2,…,w p Then, by selecting q (≦p) eigenvalues ​​and eigenvectors in descending order from the largest eigenvalue, the dimension of the original data can be reduced. Data conversion formula: T q=XW q The transformation matrix of W q =(w 1、 w2,…,w q ), w i ∈R p The column vectors can be arranged as shown below, but at the same time, the principal component score matrix T can also be determined as T q =(t 1、 t2,…,t q ), t i ∈R n will be pruned.

[0067] Here, the principal component score t1 corresponding to the largest eigenvalue (first eigenvalue) is called the first principal component, and is considered to best represent the fluctuation of the original data. j is called the j-th principal component, but as j increases, the eigenvalue becomes smaller, so its influence is thought to weaken. It is also known that the j-th eigenvalue calculated here matches the variance of the j-th principal component, and the axes of each principal component are all orthogonal.

[0068] A vector containing the occupancy numbers of p orbitals is given by the data point x i =(x i1 ,x i2 ,…,x ip ), i = 1, 2, ..., n, the dimension can be compressed from p to q. Here, the first principal component t1 = XW1(t1∈R n ,X∈R n×p ,W1∈R p×1 ) is expressed as a matrix as shown in the following equation (5). The first principal component represents the component of each data point when the direction in which the variance of the original data is maximum is taken as the coordinate axis.

[0069]

number

[0070] The importance calculation unit 15 is a processing unit that calculates the importance of molecular orbitals based on the eigenvalues ​​and eigenvectors of the first principal component to a predetermined number of principal components obtained as a result of the principal component analysis by the PCA execution unit 13.

[0071] The first principal component corresponds to the component that most dominates the variation of the original data, and quantifies the influence of the number of occupied orbitals of p in the original data on this component. In the above equation (5), the number of occupied orbitals of the i-th orbital is given by the coefficient w j1 , j=1, 2, ..., p are multiplied, so it is possible to quantify it with some function of this coefficient. Also, there is no need to consider whether the fluctuation in the number of occupancies is positive or negative, and the variance of the first principal component is the variance of the number of occupancies w i1 2 Since it is a linear sum with coefficients (assuming that the fluctuations in the number of orbitals occupied are independent), w j1 2 It can be seen that such a coefficient w is suitable as a criterion for quantification. j1 2 can be calculated as the importance of the molecular orbital.

[0072] So far, the calculation of importance based on only the first principal component has been illustrated, but it is also possible to calculate importance based on multiple principal components. For example, when realizing a comprehensive evaluation including the first principal component to the K-th principal component, the importance index v(j) of trajectory j can be calculated according to the following formula (6). This is done by multiplying the λ of the k-th eigenvalue by k Since λ is equal to the variance of the k-th principal component score, k The value of w was also used in the calculation method using only the first principal component. j1 2 This can correspond to a proportional distribution of the magnitude of the

[0073]

number

[0074] In addition, |w k | 2By normalizing v(j) as =1, 2, ..., K, this index can be simplified as shown in the following formula (7). When actually determining the activation space, this v(j) can be calculated for each trajectory j. This v(j) corresponds to the sum of squares of the factor loadings of the first to Kth principal components in principal component analysis, taken for the jth attribute variable.

[0075]

number

[0076] The orbital selection unit 17 is a processing unit that selects, from among a plurality of molecular orbitals, molecular orbitals whose importance calculated by the importance calculation unit 15 corresponds to a predetermined number of top molecular orbitals. In one embodiment, the orbital selection unit 17 can select a predetermined number of top molecular orbitals j in descending order of the importance v(j) of the molecular orbitals calculated by the importance calculation unit 15. This makes it possible to automatically determine the active space.

[0077] The information output unit 19 is a processing unit that executes information output to the client terminal 30. As merely one example, the information output unit 19 can output, as active space information, to the client terminal 30, a list of the indices j of the molecular orbitals selected by the orbital selection unit 17, a list of the importance v(j) of each molecular orbital j, the variance of the number of occupancies of each molecular orbital j, or a combination of these. Note that, although an example of outputting active space information regarding molecular orbitals selected by the orbital selection unit 17 has been given here, it is also possible to allow a user to edit the active space used in quantum chemical calculations by outputting the active space information of all orbitals j.

[0078] <Processing flow> Next, the flow of the processes executed by each device according to this embodiment will be described. Here, (1) the process of generating occupancy number data executed by the client terminal 30 will be described, followed by (2) the process of outputting information executed by the server device 10.

[0079] (1) Generation of occupancy data 5 is a flowchart showing the process of generating occupation number data. This process can be started when the receiving unit 31 receives a request for executing a quantum chemical calculation, as an example.

[0080] 5, the quantum chemistry calculation unit 33 reads the compound data input when the request is received, and also reads specified conditions such as the number of iterations of the variation calculation and the like, the active space, etc. (steps S101 and S102). Subsequently, the quantum chemistry calculation unit 33 sets the initial state of the quantum chemistry calculation (step S103).

[0081] Thereafter, the quantum chemistry calculation unit 33 executes loop processing 1, which repeats the following step S104 a number of times corresponding to the number of iterations N included in the specified conditions read in step S102. That is, the quantum chemistry calculation unit 33 executes the n-th variational optimization, etc. (step S104).

[0082] By repeating this loop process 1, a data matrix consisting of a time series of n points of an occupation number vector containing the occupation numbers of p molecular orbitals as elements is obtained as occupation number data.

[0083] Thereafter, the quantum chemistry calculation unit 33 outputs the occupation number data obtained in the loop process 1 to the server device 10 (step S105), and ends the process.

[0084] (2) Information output processing 6 is a flowchart showing the procedure of the information output process. This process can be started when the above-mentioned occupancy number data is acquired from the client terminal 30, as an example.

[0085] 6, the PCA execution unit 13 reads the data matrix X of the occupancy numbers of all orbits acquired by the data acquisition unit 11 (step S301). Subsequently, the PCA execution unit 13 calculates the variance-covariance matrix S of the occupancy numbers from the data matrix X (step S302).

[0086] Then, the PCA execution unit 13 executes numerical calculations of an algorithm that solves the eigenvalue problem for the variance-covariance matrix S (step S303). After that, the PCA execution unit 13 executes numerical calculations of the eigenvalue λ obtained as a result of the calculation in step S303. i and the corresponding eigenvector w i λ i The data are sorted in descending order of size (step S304).

[0087] Next, the importance calculation unit 15 calculates the importance index v(j) of each trajectory j from the eigenvalues ​​and eigenvectors of the first to K-th principal components (step S305). Then, the trajectory selection unit 17 selects a predetermined number L of trajectories j in descending order of importance index v(j) (step S306).

[0088] Thereafter, the information output unit 19 outputs the active space information including the indexes of the L trajectories j and a list of the importance v(j) of each trajectory j to the client terminal 30 (step S307), and ends the process.

[0089] <One aspect of the effect> As described above, the server device 10 according to this embodiment acquires occupation number data including a time series of occupation number vectors corresponding to the occupation numbers of each of a plurality of molecular orbitals, and outputs information on the active space based on the results of applying principal component analysis to the occupation number data.

[0090] Therefore, the server device 10 according to this embodiment can provide information on active spaces that contribute to quantum chemical calculations in various aspects, such as calculation accuracy, calculation resources, and calculation time. By providing such information on active spaces, it is possible to achieve a balance between the accuracy and required time of quantum chemical calculations in the molecular orbital method, even without advanced specialized knowledge of quantum chemical calculations or proficiency in quantum chemical calculation software. Therefore, it is possible to automate the specification of active spaces while eliminating dependency on individual users.

[0091] <Example 2> Although the first embodiment of the present disclosure has been described above, various applications are possible, and further, the present disclosure may be implemented in various different forms other than the first embodiment described above.

[0092] <Exercise creative ability> The matters described in the first embodiment, such as the types of algorithms and parameters of quantum chemical calculations or principal component analysis, are merely examples and can be changed. Also, the order of processing in the flowchart described in the first embodiment can be changed within a consistent range.

[0093] <System> The information including the processing procedures, control procedures, specific names, various data, and parameters shown in the above documents and drawings can be changed as desired unless otherwise specified. For example, one or more of the functional units of the server device 10, namely, the data acquisition unit 11, the PCA execution unit 13, the importance calculation unit 15, the trajectory selection unit 17, and the information output unit 19, may be configured as separate devices.

[0094] Furthermore, the components of each device shown in the figure are functional concepts and do not necessarily have to be physically configured as shown. In other words, the specific form of distribution and integration of each device is not limited to that shown. In other words, all or part of them can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc. Note that each configuration may also be a physical configuration.

[0095] For example, in the above-mentioned first embodiment, an example was given in which the generation of the above-mentioned occupancy number data was executed by the client terminal 30, but the generation of the above-mentioned occupancy number data can also be executed by the server device 10. Furthermore, in the above-mentioned first embodiment, an example was given in which the server device 10 outputs the active space information to the client terminal 30, but the server device 10 may also execute quantum chemical calculations based on the active space information.

[0096] Furthermore, each processing function performed by each device can be realized, in whole or in part, by a CPU (Central Processing Unit) and a program analyzed and executed by the CPU, or can be realized as hardware using wired logic.

[0097] <Hardware> Next, an example of the hardware configuration of the computer described in the first and second embodiments will be described. Fig. 7 is a diagram showing an example of the hardware configuration. As shown in Fig. 7, the server device 10 includes a communication device 10a, a storage device 10b, a memory 10c, and a processor 10d. The components shown in Fig. 7 may be interconnected via a bus or the like.

[0098] The communication device 10a is a network interface card, etc. The storage device 10b is a storage device such as a hard disk drive (HDD) or a solid state drive (SSD). For example, the storage device 10b stores programs and databases that operate the functions shown in FIG.

[0099] The processor 10d reads out a program that executes the same processing as the processing unit shown in FIG. 1 from the storage device 10b or the like and loads it into the memory 10c, thereby operating a process that executes the functions described in FIG.

[0100] Such a process realizes the same functions as the processing units of the server device 10. For example, the processor 10d reads from the storage device 10b or the like a program having the same functions as the data acquisition unit 11, the PCA execution unit 13, the importance calculation unit 15, the trajectory selection unit 17, the information output unit 19, etc. Then, the processor 10d executes a process that executes the same processing as the data acquisition unit 11, the PCA execution unit 13, the importance calculation unit 15, the trajectory selection unit 17, the information output unit 19, etc.

[0101] In this way, the server device 10 operates as an information processing device that executes an information output method by reading and executing a program. The server device 10 can also realize the same functions as those of the first embodiment by reading the program from a recording medium using a medium reading device and executing the read program. Note that the program in the second embodiment is not limited to being executed by the server device 10. For example, the functions of the present disclosure can be similarly applied to cases where another computer or server executes the program, or where these execute the program in cooperation with each other.

[0102] The above program can be distributed via a network such as the Internet. The above program can also be recorded on any recording medium and executed by a computer by reading it from the recording medium. For example, the recording medium can be a hard disk, a flexible disk (FD), a CD-ROM, a magneto-optical disk (MO), a digital versatile disk (DVD), or the like.

[0103] The following additional notes are provided regarding the embodiments including the above examples.

[0104] (Appendix 1) Obtaining occupation number data including a time series of the occupation number of each of a plurality of molecular orbitals, performing a principal component analysis on the occupancy count data; outputting information on an active space corresponding to a subset of the plurality of molecular orbitals to be used for quantum chemical calculations based on the results of the principal component analysis; An information output program that causes a computer to execute a process.

[0105] (Appendix 2) The information output program according to appendix 1, further causing the computer to execute a process of calculating importance of the molecular orbital based on eigenvalues ​​and eigenvectors of a first principal component to a predetermined number of principal components obtained as a result of the principal component analysis.

[0106] (Supplementary Note 3) The computer is further caused to execute a process of selecting molecular orbitals having a predetermined number of top importance levels from the plurality of molecular orbitals, 3. The information output program according to claim 2, wherein the outputting process includes a process of outputting information relating to the molecular orbital selected in the selecting process.

[0107] (Appendix 4) The information output program according to appendix 1, wherein the output process includes a process of outputting information about the index of the molecular orbital, the importance of the molecular orbital, the variance of the number of occupancies of the molecular orbital, or a combination thereof.

[0108] (Appendix 5) The information output program according to appendix 1, wherein the occupation number data corresponds to an execution result obtained by iterative calculation such as variational optimization in molecular orbital method.

[0109] (Supplementary Note 6) The information output program according to Supplementary Note 5, wherein the occupation number data is pre-convergence data obtained in the process of the iterative calculation.

[0110] (Appendix 7) Obtaining occupation number data including a time series of the occupation number of each of a plurality of molecular orbitals, performing a principal component analysis on the occupancy count data; outputting information on an active space corresponding to a subset of the plurality of molecular orbitals to be used for quantum chemical calculations based on the results of the principal component analysis; An information output method characterized in that the processing is executed by a computer.

[0111] (Appendix 8) The information output method according to appendix 7, characterized in that the computer further executes a process of calculating the importance of the molecular orbital based on the eigenvalues ​​and eigenvectors of the first principal component to a predetermined number of principal components obtained as a result of the principal component analysis.

[0112] (Supplementary Note 9) The computer further executes a process of selecting molecular orbitals having a predetermined number of top importance levels from the plurality of molecular orbitals, 9. The information output method according to claim 8, wherein the outputting step includes a step of outputting information relating to the molecular orbital selected in the selecting step.

[0113] (Appendix 10) The information output method according to appendix 7, wherein the output process includes a process of outputting information about the index of the molecular orbital, the importance of the molecular orbital, the variance of the number of occupancies of the molecular orbital, or a combination thereof.

[0114] (Supplementary Note 11) The information output method according to Supplementary Note 7, wherein the occupation number data corresponds to an execution result obtained by iterative calculation such as variational optimization in molecular orbital method.

[0115] (Supplementary Note 12) The information output method according to Supplementary Note 11, wherein the occupation number data is pre-convergence data obtained in the process of the iterative calculation.

[0116] (Appendix 13) Obtaining occupation number data including a time series of the occupation number of each of a plurality of molecular orbitals, performing a principal component analysis on the occupancy count data; outputting information on an active space corresponding to a subset of the plurality of molecular orbitals to be used for quantum chemical calculations based on the results of the principal component analysis; An information processing device comprising a control unit that executes processing.

[0117] (Supplementary Note 14) The information processing device according to Supplementary Note 13, wherein the control unit further executes a process of calculating the importance of the molecular orbital based on the eigenvalues ​​and eigenvectors of a first principal component to a predetermined number of principal components obtained as a result of the principal component analysis.

[0118] (Supplementary Note 15) The control unit further executes a process of selecting molecular orbitals having a predetermined number of top importance levels from the plurality of molecular orbitals, 15. The information processing device according to claim 14, wherein the outputting process includes a process of outputting information relating to the molecular orbital selected in the selecting process.

[0119] (Appendix 16) The information processing device according to appendix 13, wherein the output process includes a process of outputting information regarding the index of the molecular orbital, the importance of the molecular orbital, the variance of the number of occupancies of the molecular orbital, or a combination thereof.

[0120] (Supplementary Note 17) The information processing device according to Supplementary Note 13, wherein the occupation number data corresponds to an execution result obtained by iterative calculation such as variational optimization in molecular orbital method.

[0121] (Supplementary Note 18) The information processing device according to Supplementary Note 17, wherein the occupation number data is pre-convergence data obtained in the process of the iterative calculation. [Explanation of symbols]

[0122] 10 Server device 11 Data Acquisition Section 13 PCA Executive Department 15 Importance calculation part 17 Orbit selection section 19 Information output section 30 client terminals 31 Reception 33 Quantum Chemistry Computation Department 35 Output section

Claims

1. acquiring occupation number data including a time series of the occupation number of each of a plurality of molecular orbitals; performing a principal component analysis on the occupancy data; outputting information on an active space corresponding to a subset of the plurality of molecular orbitals to be used for quantum chemical calculations based on the results of the principal component analysis; An information output program that causes a computer to execute a process.

2. 2. The information output program according to claim 1, further causing the computer to execute a process of calculating importance of the molecular orbital based on eigenvalues ​​and eigenvector values ​​of a first principal component to a predetermined number of principal components obtained as a result of the principal component analysis.

3. further causing the computer to execute a process of selecting molecular orbitals that correspond to a predetermined number of top importance levels from the plurality of molecular orbitals; 3. The information output program according to claim 2, wherein the output process includes a process of outputting information relating to the molecular orbital selected in the selection process.

4. 4. The information output program according to claim 1, wherein the output process includes a process of outputting information about the index of the molecular orbital, the importance of the molecular orbital, the variance of the number of occupancies of the molecular orbital, or a combination thereof.

5. 4. The information output program according to claim 1, wherein the occupation number data corresponds to an execution result obtained by iterative calculation such as variational optimization in molecular orbital method.

6. 6. The information output program according to claim 5, wherein the occupation number data is data obtained before convergence during the iterative calculation.

7. acquiring occupation number data including a time series of the occupation number of each of a plurality of molecular orbitals; performing a principal component analysis on the occupancy data; outputting information on an active space corresponding to a subset of the plurality of molecular orbitals to be used for quantum chemical calculations based on the results of the principal component analysis; An information output method characterized in that the processing is executed by a computer.

8. acquiring occupation number data including a time series of the occupation number of each of a plurality of molecular orbitals; performing a principal component analysis on the occupancy data; outputting information on an active space corresponding to a subset of the plurality of molecular orbitals to be used for quantum chemical calculations based on the results of the principal component analysis; An information processing device comprising a control unit that executes processing.

Images