In the neural network, one or more nodes connected through the link may relatively form the relationship between an input node and an output node. Concepts of the input node and the output node are relative and a predetermined node which has the output node relationship with respect to one node may have the input node relationship in the relationship with another node and vice versa. As described above, the relationship of the input node to the output node may be generated based on the link. One or more output nodes may be connected to one input node through the link and vice versa.

In the relationship of the input node and the output node connected through one link, a value of data of the output node may be determined based on data input in the input node. Here, a link connecting the input node and the output node to each other may have a weight. The weight may be variable and the weight is variable by a user or an algorithm in order for the neural network to perform a desired function. For example, when one or more input nodes are mutually connected to one output node by the respective links, the output node may determine an output node value based on values input in the input nodes connected with the output node and the weights set in the links corresponding to the respective input nodes.

As described above, in the neural network, one or more nodes are connected to each other through one or more links to form a relationship of the input node and output node in the neural network. A characteristic of the neural network may be determined according to the number of nodes, the number of links, correlations between the nodes and the links, and values of the weights granted to the respective links in the neural network. For example, when the same number of nodes and links exist and there are two neural networks in which the weight values of the links are different from each other, it may be recognized that two neural networks are different from each other.

The neural network may be constituted by a set of one or more nodes. A subset of the nodes constituting the neural network may constitute a layer. Some of the nodes constituting the neural network may constitute one layer based on the distances from the initial input node. For example, a set of nodes of which distance from the initial input node is n may constitute n layers. The distance from the initial input node may be defined by the minimum number of links which should be passed through for reaching the corresponding node from the initial input node. However, a definition of the layer is predetermined for description and the order of the layer in the neural network may be defined by a method different from the aforementioned method. For example, the layers of the nodes may be defined by the distance from a final output node.

The initial input node may mean one or more nodes in which data is directly input without passing through the links in the relationships with other nodes among the nodes in the neural network. Alternatively, in the neural network, in the relationship between the nodes based on the link, the initial input node may mean nodes which do not have other input nodes connected through the links. Similarly thereto, the final output node may mean one or more nodes which do not have the output node in the relationship with other nodes among the nodes in the neural network. Further, a hidden node may mean nodes constituting the neural network other than the initial input node and the final output node.

In the neural network according to an exemplary embodiment of the present disclosure, the number of nodes of the input layer may be the same as the number of nodes of the output layer, and the neural network may be a neural network of a type in which the number of nodes decreases and then, increases again from the input layer to the hidden layer. Further, in the neural network according to another exemplary embodiment of the present disclosure, the number of nodes of the input layer may be smaller than the number of nodes of the output layer, and the neural network may be a neural network of a type in which the number of nodes decreases from the input layer to the hidden layer. Further, in the neural network according to yet another exemplary embodiment of the present disclosure, the number of nodes of the input layer may be larger than the number of nodes of the output layer, and the neural network may be a neural network of a type in which the number of nodes increases from the input layer to the hidden layer. The neural network according to still yet another exemplary embodiment of the present disclosure may be a neural network of a type in which the neural networks are combined.

A deep neural network (DNN) may refer to a neural network that includes a plurality of hidden layers in addition to the input and output layers. When the deep neural network is used, the latent structures of data may be determined. That is, latent structures of photos, text, video, voice, and music (e.g., what objects are in the photo, what the content and feelings of the text are, what the content and feelings of the voice are) may be determined. The deep neural network may include a convolutional neural network (CNN), a recurrent neural network (RNN), an auto encoder, generative adversarial networks (GAN), a restricted Boltzmann machine (RBM), a deep belief network (DBN), a Q network, a U network, a Siam network, a Generative Adversarial Network (GAN), and the like. The description of the deep neural network described above is just an example and the present disclosure is not limited thereto.

In an exemplary embodiment of the present disclosure, the network function may include the auto encoder. The auto encoder may be a kind of artificial neural network for outputting output data similar to input data. The auto encoder may include at least one hidden layer and odd hidden layers may be disposed between the input and output layers. The number of nodes in each layer may be reduced from the number of nodes in the input layer to an intermediate layer called a bottleneck layer (encoding), and then expanded symmetrical to reduction to the output layer (symmetrical to the input layer) in the bottleneck layer. The auto encoder may perform non-linear dimensional reduction. The number of input and output layers may correspond to a dimension after preprocessing the input data. The auto encoder structure may have a structure in which the number of nodes in the hidden layer included in the encoder decreases as a distance from the input layer increases. When the number of nodes in the bottleneck layer (a layer having a smallest number of nodes positioned between an encoder and a decoder) is too small, a sufficient amount of information may not be delivered, and as a result, the number of nodes in the bottleneck layer may be maintained to be a specific number or more (e.g., half of the input layers or more).

The neural network may be trained in at least one scheme of supervised learning, unsupervised learning, semi supervised learning, or reinforcement learning. The learning of the neural network may be a process in which the neural network applies knowledge for performing a specific operation to the neural network.

The neural network may be trained in a direction to minimize errors of an output. The training of the neural network is a process of repeatedly inputting training data into the neural network and calculating the output of the neural network for the training data and the error of a target and back-propagating the errors of the neural network from the output layer of the neural network toward the input layer in a direction to reduce the errors to update the weight of each node of the neural network. In the case of the supervised learning, the training data labeled with a correct answer is used for each training data (i.e., the labeled training data) and in the case of the unsupervised learning, the correct answer may not be labeled in each training data. That is, for example, the training data in the case of the supervised learning related to the data classification may be data in which category is labeled in each training data. The labeled training data is input to the neural network, and the error may be calculated by comparing the output (category) of the neural network with the label of the training data. As another example, in the case of the unsupervised learning related to the data classification, the training data as the input is compared with the output of the neural network to calculate the error. The calculated error is back-propagated in a reverse direction (i.e., a direction from the output layer toward the input layer) in the neural network and connection weights of respective nodes of each layer of the neural network may be updated according to the back propagation. A variation amount of the updated connection weight of each node may be determined according to a learning rate. Calculation of the neural network for the input data and the back-propagation of the error may constitute a training cycle (epoch). The learning rate may be applied differently according to the number of repetition times of the training cycle of the neural network. For example, in an initial stage of the training of the neural network, the neural network ensures a certain level of performance quickly by using a high learning rate, thereby increasing efficiency and uses a low learning rate in a latter stage of the training, thereby increasing accuracy.

In training of the neural network, the training data may be generally a subset of actual data (i.e., data to be processed using the trained neural network), and as a result, there may be a training cycle in which errors for the training data decrease, but the errors for the actual data increase. Overfitting is a phenomenon in which the errors for the actual data increase due to excessive training of the training data. For example, a phenomenon in which the neural network that trains a cat by showing a yellow cat sees a cat other than the yellow cat and does not recognize the corresponding cat as the cat may be a kind of overfitting. The overfitting may act as a cause which increases the error of the machine learning algorithm. Various optimization methods may be used in order to prevent the overfitting. In order to prevent the overfitting, a method such as increasing the training data, regularization, dropout of omitting a part of the node of the network in the process of training, utilization of a batch normalization layer, etc., may be applied.

FIG. 3 is a flowchart of a method for improving semantic segmentation accuracy at boundary regions according to an exemplary embodiment of the present disclosure.

A computing device 100 according to an exemplary embodiment of the present disclosure may directly obtain or receive, from an external system, “information for performing object segmentation by using a neural network model”. The external system may be a server, database or the like that stores and manages the information for performing the object segmentation by using the neural network model. The computing device 100 may use the information obtained directly or received from the external system as “input data for performing the object segmentation by using the neural network model”.

According to an exemplary embodiment of the present disclosure, the computing device 100 may obtain a training image including an object (S110). In this case, the object included in the training image may refer to an object to be subjected to object segmentation, and examples thereof may illustratively include, but are not limited to, a box, a product, and the like in a manufacturing process. Further, the training image may include an image in which the object is captured to be included and an image in which a region of the object is labeled. Meanwhile, the obtained training image may be used in a process of training the neural network model for performing the object segmentation, which will be described below.

According to an exemplary embodiment of the present disclosure, the computing device 100 may train the neural network model for performing the object segmentation by using the training image obtained through step S110 (S120). For example, the computing device 100 may obtain a predefined training rate for the neural network model, and train, based on the predefined training rate, the neural network model for performing the object segmentation by using the training image. Further, the computing device 100 may stop training when the training rate of the neural network model reaches the predefined training rate. As a specific example, the computing device 100 may set a predefined training rate for the neural network model for performing the object segmentation to be 99%, and thus may stop training when the training rate of the neural network model reaches the predefined training rate of 99%. In this case, as the predefined training rate, in addition to 99%, various examples may be used, but examples thereof are not limited thereto. In addition, the computing device 100 may train, based on at least one of a cross entropy loss and a dice coefficient loss, the neural network model for performing the object segmentation by utilizing the training image, and DeepLabv3+SETR, DETR, or the like may be illustratively utilized as the neural network model for performing the object segmentation, but examples thereof are not limited thereto. Meanwhile, a process of training, by the computing device 100, the neural network model for performing the object segmentation by using the obtained training image will be described below in detail with reference to FIG. 4.

According to an exemplary embodiment of the present disclosure, the computing device 100 may identify a boundary region of the object included in the training image obtained through step S110 (S130). According to an example, the computing device 100 may identify the boundary region of the object based on a predefined size based on a contour of the object included in the training image. For example, the computing device 100 may identify, based on a contour of the object region included in the training image, the region as a boundary region of the object based on a predefined size of 10 pixels. However, as the predefined size, various examples may be used in addition to 10 pixels. As another example, the computing device 100 may identify the boundary region of the object based on a distance up to a boundary of the object from a center of the object included in the training image. Specifically, the computing device 100 may set a horizontal axis and a vertical axis with respect to the center of the object when the object has a rectangular shape, may identify the boundary region of the object based on a distance between the set horizontal axis and an inner boundary or a contour of the object, and may identify the boundary region of the object based on the distance between the set vertical axis and the inner boundary or the contour of the object. Meanwhile, a specific description for the process of identifying the object region of the object included in the training image by the computing device 10 through the exemplary embodiment of the present disclosure will be described below through FIGS. 5A and 5B below.

According to an exemplary embodiment of the present disclosure, the computing device 100 may obtain an additional training image based on the training image obtained through step S110 and the boundary region identified through step S130 (S140). For example, the computing device 100 may set a separate weight for the identified boundary region of the training image to obtain the additional training image. Specifically, the computing device 100 may set a second weight for the identified boundary region of the additional training image to be higher than a first weight of an intra-object region included in the additional training image, and exemplarily, the computing device 110 may set the first weight for the intra-object region included in the additional training image to 1, and set the second weights for the identified boundary region of the additional training image to 2. However, a value of the set first or second weight is not limited to 1 or 2, and various exemplary embodiments applicable to those skilled in the art may be applied. Further, the computing device 100 may train the neural network model trained through the above-described exemplary embodiment through the additional training image, and specific descriptions related thereto will be described below.

According to an exemplary embodiment of the present disclosure, the computing device 100 may additionally train the neural network model trained through step S120 based on the additional training image obtained through step S140 (S150). Specifically, the computing device 100 may additionally train the trained neural network model based on the additional training image for which the weight is set. For example, the computing device 100 may utilize a boundary intersection over union (IOU) to additionally train the trained neural network model based on the additional training image. Additionally the computing device 100 may utilize the additional training image as a validation metric for the additionally trained neural network model. In this regard, the computing device 100 may additionally train the trained neural network model based on the additional training image for which the weight is set, so as to solve a problem in that a score is set at a very high level even if a number of pixels of a boundary line are not accurately predicted due to the measurement of a simple overlapping portion between two sets of the training image and the prediction image, and obtain a technical effect of improving semantic segmentation accuracy of the boundary region. Meanwhile, a specific process in which the computing device 100 additionally trains the neural network model trained based on the additional training image will be described below with reference to FIG. 6.

FIG. 4 is a schematic diagram for describing a process of training a neural network model for performing object segmentation by utilizing a training image according to an exemplary embodiment of the present disclosure.

Referring to FIG. 4, the computing device 100 may obtain training images 10 and 11 including objects. In this case, the object included in the training image 10 may refer to an object to be subjected to object segmentation, and examples thereof may illustratively include, but are not limited to, a box, a product, and the like in a manufacturing process. Further, the training images 10 and 11 may include an image 10 in which the object is captured to be included and an image 11 in which a region of the object is labeled. Meanwhile, the obtained training images 10 and 11 may be used in a process of training a neural network model 12 for performing the object segmentation, which will be described below.

Thereafter, the computing device 100 may utilize the obtained training images 10 and 11 to train the neural network model 12 for performing object segmentation. Specifically, the computing device 100 may input the image 10 captured to include the object into the neural network model 12, and output a region 13-1 determined as the object as a prediction. In addition, the computing device 100 may compare the region 13-1 determined to be the predicted object with a correct answer 11-1 included in the image 11 labeled with the region of the object to calculate a loss function, and train the neural network model 12 for performing the object segmentation to minimize the loss function. In this case, the computing device 100 may train, based on at least one of a cross entropy loss and a dice coefficient loss, the neural network model for performing the object segmentation by utilizing the training image, and DeepLabv3+SETR, DETR, or the like may be illustratively used as the neural network model for performing the object segmentation, but examples thereof are not limited thereto. Meanwhile, the computing device 100 may obtain a predefined training rate for the neural network model 12, and based on the predefined training rate, train the neural network model 12 by utilizing the training images 10 and 11, and stop training when the training rate of the neural network model 12 reaches the predefined training rate. For example, the computing device 100 may set a predefined training rate for the neural network model 12 for performing the object segmentation to be 99%, and thus may stop training when the training rate of the neural network model 12 reaches the predefined training rate of 99%. Through this, the computing device 100 may first train the neural network model 12 to predict an overall region of an object before additionally training the neural network model 12 to accurately distinguish boundaries of the object, thereby increasing efficiency of additional training, which will be described below. In this case, as the predefined training rate, in addition to 99%, various examples may be used, but examples thereof are not limited thereto. Meanwhile, the computing device 100 may obtain additional training images based on the training images 10 and 11, which will be described in detail below with reference to FIGS. 5A and 5B.

FIGS. 5A and 5B are schematic diagrams for describing a process of obtaining an additional training image based on a training image and an identified boundary region according to an embodiment of the present disclosure.

First, referring to FIG. 5A, the computing device 100 may identify, based on a predefined size based on a contour of an object included in each of the training images 10 and 11, a boundary region 11-2 of the object. In this regard, when the object is a rectangular mobile phone, the object may have a rectangular shape, but in reality, the boundary may have a bumpy shape rather than a perfectly straight line. Therefore, in order to accurately distinguish the boundary of the object, the computing device 100 needs a process of identifying the boundary region of the object in addition to distinguishing the overall region of the object, and obtaining the additional training image so that the boundary of the region may be more accurately distinguished. For example, the computing device 100 may identify, based on a contour of the object region included in the training image 11, the region as the boundary region 11-2 of the object based on a predefined size of 10 pixels. However, as the predefined size, various examples may be used in addition to 10 pixels.

As another example, referring to FIG. 5B, the computing device 100 may identify, based on a distance up to a boundary 20-1 or 20-2 of an object based on a center of the object included in each of the training images 10 and 11, a boundary region 21-1 of the object. Specifically, the computing device 100 may set a horizontal axis and a vertical axis with respect to the center of the object when the object has a rectangular shape, may identify the boundary region of the object based on a distance between the set horizontal axis and an inner boundary or a contour of the object, and may identify the boundary region of the object based on the distance between the set vertical axis and the inner boundary or the contour of the object. For example, the computing device 100 may measure a shortest distance 20-1 from a horizontal axis relative to the center of the object to an inner boundary of the object, and may measure a longest distance 20-2 from the horizontal axis relative to the center of the object, to a contour of the object. Similarly, the computing device 100 may also identify a size (illustratively, A px) of a boundary region of the object based on “a longest distance 20-2 from a horizontal axis to a contour of the object”-“a shortest distance 20-1 from the horizontal axis to an inner boundary of the object,” and “a longest distance from a vertical axis to the contour of the object”-“a shortest distance from the vertical axis to the inner boundary of the object,” to obtain the boundary region 21-1 of the object. Meanwhile, in addition to the rectangular object in the example of FIG. 5B, when the object is circular, the computing device 100 may obtain the boundary region 21-1 of the object by identifying the size of the boundary region of the object through a shortest distance from the center to the circle inscribed in the object, a longest distance to the circle circumscribing the object, and the like, and various examples may be utilized. Through this, the computing device 100 may identify a region in which the boundary of the object needs to be accurately distinguished, and the identified boundary region of the object may be utilized in a process of obtaining an additional training image, which will be described below.

Further, the computing device 100 may obtain an additional training image 11′ or 21 based on the training images 10 and 11 and the boundary region 11-2 or 21-1 identified via the exemplary embodiments of FIGS. 5A and 5B. For example, the computing device 100 may obtain the additional training image 11′ or 21 by setting a separate weight for the training images 10 and 11 and the identified boundary region 11-2 or 21-1. Specifically, the computing device 100 may set a second weight for the identified boundary region 11-2 or 21-1 of the additional training image 11′ or 21 to be higher than a first weight of an intra-object region included in the additional training image 11′ or 21, and exemplarily, the computing device 100 may set the first weight for the intra-object region included in the additional training image 11′ or 21 to 1, and set the second weight for the identified boundary region 11-2 or 21-1 of the additional training image 11′ or 21 to 2. However, a value of the set first or second weight is not limited to 1 or 2, and various exemplary embodiments applicable to those skilled in the art may be applied. Meanwhile, the computing device 100 may train the neural network model trained through the above-described exemplary embodiment of FIG. 4 through the additional training image 11′ or 21, and specific descriptions related thereto will be described below through FIG. 6. Meanwhile, expressions such as first, second, etc., disclosed throughout the present disclosure are just a meaning for distinguishing components, and the present disclosure is not limited thereto.

FIG. 6 is a schematic diagram for describing a process of additionally training the neural network model trained based on the additional training image according to an exemplary embodiment of the present disclosure.

Referring to FIG. 6, the computing device 100 may additionally train the neural network model 12′ trained at the training rate of 99% based on the additional training image 11′ or 21 for which the weight is set. For example, the computing device 100 may utilize a boundary intersection over union (IOU) to additionally train the trained neural network model 12′ based on the additional training image 11′ or 21. More specifically, the computing device 100 may input the image 10 captured to include the object into the trained neural network model 12′, and output a region 31-1 determined as the object as a prediction. In addition, the computing device 100 may compare the region 31-1 determined to be the predicted object with a correct answer 11-1, 11-2, or 21-1 included in the additional training image 11′ or 21 labeled with the region and boundary of the object to calculate a loss function, and additionally train the trained neural network model 12′ for performing the object segmentation to minimize the loss function. In this case, in the process of calculating the loss function, the computing device 100 may additionally train the trained neural network model 12′ by reflecting a fact that the second weight for the identified boundary region 11-2 or 21-1 is set to be higher than the first weight of the intra-object region 11-1 included in the additional training image 11′ or 21. Additionally the computing device 100 may utilize the additional training image as a validation metric for the additionally trained neural network model to evaluate a performance of the additionally trained neural network model. Through this, the computing device 100 may additionally train the trained neural network model 12′ based on the additional training image 11′ or 21 for which the weight is set, so as to solve a problem in that a score is set at a very high level even if a number of pixels of a boundary line are not accurately predicted due to the measurement of a simple overlapping portion between two sets of the training image and the prediction image, and obtain a technical effect of improving semantic segmentation accuracy of the boundary region.

Disclosed is a computer readable medium storing the data structure according to an exemplary embodiment of the present disclosure.

The data structure may refer to the organization, management, and storage of data that enables efficient access to and modification of data. The data structure may refer to the organization of data for solving a specific problem (e.g., data search, data storage, data modification in the shortest time). The data structures may be defined as physical or logical relationships between data elements, designed to support specific data processing functions. The logical relationship between data elements may include a connection between data elements that the user defines. The physical relationship between data elements may include an actual relationship between data elements physically stored on a computer-readable storage medium (e.g., persistent storage device). The data structure may specifically include a set of data, a relationship between the data, a function which may be applied to the data, or instructions. Through an availablely designed data structure, a computing device can perform operations while using the resources of the computing device to a minimum. Specifically, the computing device can increase the efficiency of operation, read, insert, delete, compare, exchange, and search through the availablely designed data structure.

The data structure may be divided into a linear data structure and a non-linear data structure according to the type of data structure. The linear data structure may be a structure in which only one data is connected after one data. The linear data structure may include a list, a stack, a queue, and a deque. The list may mean a series of data sets in which an order exists internally. The list may include a linked list. The linked list may be a data structure in which data is connected in a scheme in which each data is linked in a row with a pointer. In the linked list, the pointer may include link information with next or previous data. The linked list may be represented as a single linked list, a double linked list, or a circular linked list depending on the type. The stack may be a data listing structure with limited access to data. The stack may be a linear data structure that may process (e.g., insert or delete) data at only one end of the data structure. The data stored in the stack may be a data structure (LIFO—Last in First Out) in which the data is input last and output first. The queue is a data listing structure that may access data limitedly and unlike a stack, the queue may be a data structure (FIFO—First in First Out) in which late stored data is output late. The deque may be a data structure capable of processing data at both ends of the data structure.

The non-linear data structure may be a structure in which a plurality of data are connected after one data. The non-linear data structure may include a graph data structure. The graph data structure may be defined as a vertex and an edge, and the edge may include a line connecting two different vertices. The graph data structure may include a tree data structure. The tree data structure may be a data structure in which there is one path connecting two different vertices among a plurality of vertices included in the tree. That is, the tree data structure may be a data structure that does not form a loop in the graph data structure.

In the present disclosure, a network function, an artificial neural network, and a neural network may be used to be exchangeable. From here on, it will be described uniformly using neural networks.

The data structure may include the neural network. In addition, the data structures, including the neural network, may be stored in a computer readable medium. The data structure including the neural network may also include data preprocessed for processing by the neural network, data input to the neural network, weights of the neural network, hyper parameters of the neural network, data obtained from the neural network, an active function associated with each node or layer of the neural network, and a loss function for training the neural network. The data structure including the neural network may include predetermined components of the components disclosed above. In other words, the data structure including the neural network may include all of data preprocessed for processing by the neural network, data input to the neural network, weights of the neural network, hyper parameters of the neural network, data obtained from the neural network, an active function associated with each node or layer of the neural network, and a loss function for training the neural network or a combination thereof. In addition to the above-described configurations, the data structure including the neural network may include predetermined other information that determines the characteristics of the neural network. In addition, the data structure may include all types of data used or generated in the calculation process of the neural network, and is not limited to the above. The computer readable medium may include a computer readable recording medium and/or a computer readable transmission medium. The neural network may be generally constituted by an aggregate of calculation units which are mutually connected to each other, which may be called nodes. The nodes may also be called neurons. The neural network is configured to include one or more nodes.

The data structure may include data input into the neural network. The data structure including the data input into the neural network may be stored in the computer readable medium. The data input to the neural network may include training data input in a neural network training process and/or input data input to a neural network in which training is completed. The data input to the neural network may include preprocessed data and/or data to be preprocessed. The preprocessing may include a data processing process for inputting data into the neural network. Therefore, the data structure may include data to be preprocessed and data generated by preprocessing. The data structure is just an example and the present disclosure is not limited thereto.

The data structure may include the weight of the neural network (in the present disclosure, the weight and the parameter may be used as the same meaning). In addition, the data structures, including the weight of the neural network, may be stored in the computer readable medium. The neural network may include a plurality of weights. The weight may be variable and the weight is variable by a user or an algorithm in order for the neural network to perform a desired function. For example, when one or more input nodes are mutually connected to one output node by the respective links, the output node may determine a data value output from an output node based on values input in the input nodes connected with the output node and the weights set in the links corresponding to the respective input nodes. The data structure is just an example and the present disclosure is not limited thereto.

As a non-limiting example, the weight may include a weight which varies in the neural network training process and/or a weight in which neural network training is completed. The weight which varies in the neural network training process may include a weight at a time when a training cycle starts and/or a weight that varies during the training cycle. The weight in which the neural network training is completed may include a weight in which the training cycle is completed. Accordingly, the data structure including the weight of the neural network may include a data structure including the weight which varies in the neural network training process and/or the weight in which neural network training is completed. Accordingly, the above-described weight and/or a combination of each weight are included in a data structure including a weight of a neural network. The data structure is just an example and the present disclosure is not limited thereto.

The data structure including the weight of the neural network may be stored in the computer-readable storage medium (e.g., memory, hard disk) after a serialization process. Serialization may be a process of storing data structures on the same or different computing devices and later reconfiguring the data structure and converting the data structure to a form that may be used. The computing device may serialize the data structure to send and receive data over the network. The data structure including the weight of the serialized neural network may be reconfigured in the same computing device or another computing device through deserialization. The data structure including the weight of the neural network is not limited to the serialization. Furthermore, the data structure including the weight of the neural network may include a data structure (for example, B-Tree, Trie, m-way search tree, AVL tree, and Red-Black Tree in a nonlinear data structure) to increase the efficiency of operation while using resources of the computing device to a minimum. The above-described matter is just an example and the present disclosure is not limited thereto.

The data structure may include hyper-parameters of the neural network. In addition, the data structures, including the hyper-parameters of the neural network, may be stored in the computer readable medium. The hyper-parameter may be a variable which may be varied by the user. The hyper-parameter may include, for example, a learning rate, a cost function, the number of training cycle iterations, weight initialization (for example, setting a range of weight values to be subjected to weight initialization), and Hidden Unit number (e.g., the number of hidden layers and the number of nodes in the hidden layer). The data structure is just an example and the present disclosure is not limited thereto.

FIG. 7 is a normal and schematic view of an exemplary computing environment in which the exemplary embodiments of the present disclosure may be implemented.

It is described above that the present disclosure may be generally implemented by the computing device, but those skilled in the art will well know that the present disclosure may be implemented in association with a computer executable command which may be executed on one or more computers and/or in combination with other program modules and/or a combination of hardware and software.

In general, the program module includes a routine, a program, a component, a data structure, and the like that execute a specific task or implement a specific abstract data type. Further, it will be well appreciated by those skilled in the art that the method of the present disclosure can be implemented by other computer system configurations including a personal computer, a handheld computing device, microprocessor-based or programmable home appliances, and others (the respective devices may operate in connection with one or more associated devices as well as a single-processor or multi-processor computer system, a mini computer, and a main frame computer.