CN110705714B - Deep learning model detection method, deep learning platform and computer equipment - Google Patents
Deep learning model detection method, deep learning platform and computer equipment Download PDFInfo
- Publication number
- CN110705714B CN110705714B CN201910926898.8A CN201910926898A CN110705714B CN 110705714 B CN110705714 B CN 110705714B CN 201910926898 A CN201910926898 A CN 201910926898A CN 110705714 B CN110705714 B CN 110705714B
- Authority
- CN
- China
- Prior art keywords
- file
- data
- deep learning
- binary file
- binary
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Image Analysis (AREA)
Abstract
The application relates to a deep learning model detection method, a deep learning platform and computer equipment. The method comprises the following steps: acquiring data to be detected, and converting the data to be detected into a first binary file for storage; selecting a training frame based on user input, and converting the training frame into a second binary file for storage; converting the first binary file into a model data file corresponding to the selected training frame; and obtaining a deep learning model output file according to the model data file and the second binary file. The training frames under different depth learning frames are converted into binary files, the data to be detected are converted into model data files corresponding to the frames according to the selected training frames, deep learning model detection is carried out, a user only needs to care about the input data to be detected and the output deep learning model output files, the user does not need to learn various deep learning frames, and the cost of developing the deep learning models by the user is reduced.
Description
Technical Field
The application relates to the technical field of information processing, in particular to a deep learning model detection method, a deep learning platform and computer equipment.
Background
In the deep learning process, various deep learning framework construction principles exist in the prior art, such as tensorflow, caffe, keras, pyrrch and the like. The TensorFlow is a symbolic mathematical system based on data flow programming, is widely applied to programming realization of various machine learning algorithms, and the predecessor of the TensorFlow is a neural network algorithm library of Google. Caffe is a deep learning framework with expressiveness, speed and thinking modularization. Keras is an open source artificial neural network library written by Python, and can be used as a high-level application program interface of Tensorflow, Microsoft-CNTK and Theano for designing, debugging, evaluating, applying and visualizing a deep learning model. PyTorch is a Python-based library that provides a flexible deep learning development platform.
Models supported by the deep learning frameworks in the prior art are different, and when the models are applied to products, certain limitations are caused. A deep learning training model file is obtained, the model file needs to be analyzed in the process of prediction, but if a user does not use the framework and does not know the principle of the model file, the user needs to invest time to learn the framework so as to reproduce the deep learning model.
Disclosure of Invention
In view of the above, it is necessary to provide a deep learning model detection method, a deep learning platform, and a computer device that can be applied to various deep learning frame architectures.
A method of deep learning model detection, the method comprising: acquiring data to be detected, and converting the data to be detected into a first binary file for storage; selecting a training frame based on user input, and converting the training frame into a second binary file for storage; converting the first binary file into a model data file corresponding to the selected training frame; and obtaining a deep learning model output file according to the model data file and the second binary file.
In one embodiment, the acquiring data to be detected includes: acquiring to-be-detected data input by a user and a storage path of an output file based on the input parameters; the input parameters include: row parameters, column parameters, data type, image type, and normalization parameters.
In one embodiment, the converting the data to be detected into the first binary file for storage includes: converting the data to be detected into a first binary file; and writing the input parameters and the storage path of the output file into a data head of the first binary file for storage.
In one embodiment, the converting the first binary file into the model data file corresponding to the selected training frame includes: analyzing the first binary file to obtain an input parameter in the data head and a storage path of an output file; and converting the first binary file into a model data file corresponding to the training frame according to the input parameters and the selected training frame.
In one embodiment, the obtaining the deep learning model output file according to the model data file and the second binary file includes: obtaining a deep learning model output file according to the model data file and the second binary file; and storing the deep learning model output file according to the input parameters and the storage path of the output file.
In one embodiment, the method further comprises: a first binary file and a second binary file storage path are preset.
In one embodiment, the second binary file is a static binary file in the form of a protocol buffer.
A deep learning platform, the deep learning platform comprising: the preprocessing module is used for acquiring data to be detected input by a user and a storage path of an output file based on input parameters, converting the data to be detected into a first binary file, and writing the input parameters and the storage path of the output file into a data head of the first binary file for storage; the model processing module is used for selecting a training frame based on user input and converting the training frame into a second binary file for storage; the output module is used for analyzing the first binary file to obtain input parameters in the data head and a storage path of the output file; converting the first binary file into a model data file corresponding to the selected training frame; obtaining a deep learning model output file according to the model data file and the second binary file; and storing the deep learning model output file according to the input parameters and the storage path of the output file.
A computer device comprising a memory storing a computer program and a processor implementing the steps of any of the methods described above when the computer program is executed by the processor.
A computer-readable storage medium, having stored thereon a computer program which, when being executed by a processor, carries out the steps of the method of any of the above.
According to the deep learning model detection method, the deep learning platform and the computer equipment, the data to be detected are obtained, the data to be detected are converted into the first binary file to be stored, the training frame is selected based on user input, the frame is converted into the second binary file to be stored, the first binary file is converted into the model data file corresponding to the selected training frame, and the deep learning model output file is obtained according to the model data file and the second binary file. The training frames under different depth learning frames are converted into a binary file, data to be detected are converted into model data files corresponding to the frames according to the selected training frames, detection of deep learning models is carried out, a user only needs to care about input data to be detected and output deep learning model output files, the user does not need to learn various deep learning frames, and the cost of developing the deep learning models by the user is reduced.
Drawings
FIG. 1 is a schematic flow chart diagram illustrating a deep learning model detection method according to an embodiment;
FIG. 2 is a block diagram showing the structure of a deep learning model detection apparatus according to an embodiment;
FIG. 3 is a diagram of the internal structure of a computer device in one embodiment.
Detailed Description
To facilitate understanding of the present application, the present application will be described in detail with reference to the accompanying drawings. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application, and in order to provide a thorough understanding of the present application, preferred embodiments of the present application are set forth in the accompanying drawings. This application may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete. This application is capable of implementation in many other ways than those herein described and of similar modifications by one of ordinary skill in the art without departing from the spirit and scope of the present application and is therefore not limited to the specific embodiments disclosed below.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the feature. In the description of the present application, "plurality" means at least two, e.g., two, three, etc., unless explicitly specified otherwise. In the description of the present application, "a number" means at least one, such as one, two, etc., unless specifically limited otherwise.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
Deep learning is one of machine learning, and machine learning is a must-pass path for realizing artificial intelligence. The concept of deep learning is derived from the research of artificial neural networks, and a multi-layer perceptron comprising a plurality of hidden layers is a deep learning structure. Deep learning is a general term of a class of pattern analysis methods, and mainly relates to three methods: the first kind of convolutional neural network is a neural network system based on convolutional operation. The second kind of self-coding neural network is a neural network based on multilayer neurons, and mainly includes self-coding and sparse coding. And the third deep confidence network is trained in a mode of a multilayer self-coding neural network, and the weight of the neural network is further optimized by combining the identification information. Deep learning is the intrinsic law and expression level of learning sample data, and information obtained in the learning process is very helpful for interpretation of data such as characters, images and sounds. The final aim of the method is to enable a machine to have analysis and learning capabilities like a human, and to recognize data such as characters, images and sounds. Deep learning is a complex machine learning algorithm, and achieves the effect in speech and image recognition far exceeding the prior related art.
In the initial stage of deep learning, each deep learning researcher needs to write a large amount of repetitive codes. In order to improve the work efficiency, the researchers write the codes into a frame and put the frame on the network for all the researchers to use, and the frame is a deep learning frame. The most popular deep learning frameworks around the world are paddlepaddlefold, tensrflow, Caffe, Theano, MXNet, Torch, PyTorch, and the like.
In one embodiment, as shown in fig. 1, a deep learning model detection method is provided, which includes the following steps:
step S102, data to be detected is obtained, and the data to be detected is converted into a first binary file to be stored.
Specifically, for a deep learning model which is trained, data to be detected needs to be input into the model when the deep learning model is used, so that an output file is obtained, and the output file is a result obtained after the data to be detected is analyzed by the deep learning model. Taking an example of detecting a focus by using a deep learning model for explanation, an original image obtained by scanning of medical imaging equipment is input into the deep learning model to obtain a focus region image. The medical imaging device may be a PET, CT, MR or like scanning device. Wherein, the original image input into the deep learning model is data to be detected, and the obtained focus area image is an output file. The more specific method for acquiring the data to be detected is as follows: and acquiring to-be-detected data input by a user based on the input parameters. The input parameters comprise row, column, data type, rgb/gray, normalization, threshold and the like. row is a row parameter, column is a column parameter, data type is a data type, rgb/gray is an image type, normalization is a normalization parameter, and threshold is a threshold. The input parameters are format limitations of the data to be predicted, and can be understood as related parameters of the data to be predicted. And acquiring the data to be detected in a corresponding input parameter format according to preset input parameters. The storage path of the output file of the deep learning model is also required to be set in advance. After the data to be detected and the storage path of the output file are acquired, the data to be detected need to be processed and stored. And converting the data to be detected into a first binary file, and writing the input parameters and the storage path of the output file into the data head of the first binary file for storage.
And step S104, selecting a training frame based on user input, and converting the training frame into a second binary file for storage.
Specifically, a variety of deep learning frameworks are provided within the system, such as: PaddlePaddle, tensoflow, Caffe, keras, thano, MXNet, Torch, PyTorch, and the like. All the deep learning frames can be uniformly converted into binary files, and then the binary files of the corresponding deep learning frames are called according to the deep learning frames selected by the user. Or receiving the deep learning frame selected by the user, and then converting the binary file for the corresponding deep learning frame. Only the deep learning frame selected by the user is finally converted into the binary file, and the method is not particularly limited in the application. The preferred second binary is a static binary in the form of a physical buffer. Protocol Buffers are a lightweight and efficient structured data storage format, and can be used for structured data serialization or serialization. It is well suited for data storage or RPC data exchange formats. The method can be used for language independence, platform independence and extensible serialization structure data formats in the fields of communication protocols, data storage and the like, and Protocol Buffers are binary data transmission formats with excellent efficiency and compatibility.
Step S106, converting the first binary file into a model data file corresponding to the selected training frame.
Specifically, a first binary file is obtained first, and the first binary file is analyzed to obtain an input parameter included in a data header of the first binary file and a storage path of an output file. And converting the first binary file into a model data file corresponding to the training frame according to the input parameters and the deep learning frame selected by the user. For example, if the user selects the Caffe deep learning framework, the first binary file is converted into an input model array format specific to Caffe; and if the user selects the keras deep learning framework, the first binary file is converted into an input model array format specific to the keras.
And step S108, obtaining a deep learning model output file according to the model data file and the second binary file.
Specifically, according to the input of the user, the storage format and the storage parameters of the output file are obtained. The storage parameter may be directly input by a user, or may be obtained according to the input parameter, that is, the input parameter is used as the storage parameter of the output file. The storage format includes: picture format, binary file format, and dicom format, etc. The storage parameters include: row, column, data type, and rgb/gray, where row is a row parameter, column is a column parameter, data type is a data type, and rgb/gray is an image type. And storing the deep learning model output file according to the input parameters and the storage path of the output file. That is, the output file is stored in the storage path of the output file according to the storage format and the input parameters.
In one embodiment, the first binary file and the second binary file storage path are preset. So as to conveniently call the corresponding file according to the instruction of the user.
According to the deep learning model detection method, the data to be detected are obtained, the data to be detected are converted into the first binary file to be stored, the training frame is selected based on user input, the frame is converted into the second binary file to be stored, the first binary file is converted into the model data file corresponding to the selected training frame, and the deep learning model output file is obtained according to the model data file and the second binary file. The training frames under different depth learning frames are converted into a binary file, data to be detected are converted into model data files corresponding to the frames according to the selected training frames, detection of deep learning models is carried out, a user only needs to care about input data to be detected and output deep learning model output files, the user does not need to learn various deep learning frames, and the cost of developing the deep learning models by the user is reduced.
It should be understood that, although the steps in the flowchart of fig. 1 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a portion of the steps in fig. 1 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.
In one embodiment, as shown in fig. 2, there is provided a deep learning platform comprising: a preprocessing module 100, a model processing module 200, and an output module 300, wherein:
the preprocessing module 100 is configured to obtain data to be detected input by a user and a storage path of an output file based on an input parameter, convert the data to be detected into a first binary file, and write the input parameter and the storage path of the output file into a data header of the first binary file for storage.
And the model processing module 200 is configured to select a training frame based on user input, and convert the training frame into a second binary file for storage.
The output module 300 is configured to parse the first binary file to obtain an input parameter in the data header and a storage path of the output file; converting the first binary file into a model data file corresponding to the selected training frame; obtaining a deep learning model output file according to the model data file and the second binary file; and storing the deep learning model output file according to the input parameters and the storage path of the output file.
More specifically, the preprocessing module 100 obtains input parameters (row, column, data type, rgb/gray, normalization, threshold, etc.) and a storage path of the output formatted file based on the user input. And storing and processing the data to be predicted, storing the data to be predicted into a binary file according to a storage path, and storing the input parameters and the storage path of the output formatted file into a data header of the binary file. The model processing module 200 performs agreement conversion processing on the model file obtained by the existing mainstream deep learning framework. An interface is provided to support the user to select the training frame used by the user and the storage path of the final conversion model file. The model formats and internal structures stored after different frames are trained are different, and static binary files which are internally and uniformly converted into a protocal buffer form are stored according to storage paths. The PB is a light and efficient structured data storage format, can be used for structured data serialization, and is very suitable for being used as a data storage or RPC data exchange format. It can be used in the fields of communication protocol, data storage, etc. and has no language relation, platform relation and extensible serialization structure data format. It is a binary data transmission format excellent in efficiency and compatibility. The output module 300 reads the binary file stored in the preprocessing module 100 into the output module 300, analyzes header information in the binary file, and converts the binary file into a specific input model array format. An interface is provided for the user to select the storage format (picture format, binary file, dicom format, etc.) and storage parameters (row, column, data type, rgb/gray) of the resulting output. And (3) after the results obtained by the preprocessing module 100, the model processing module 200 and the output module 300 are immature in the product environment and deployment is finished, starting prediction data and completing batch data prediction.
The embodiment provides a platform scheme, a man-machine interactive interface is designed for a user to select customization, model files obtained under different learning frames are converted into files in a protocol buffer form, and then a special interface analysis file is provided, so that a universal interface which only needs to care about input and output parameters and shields variables and parameters of an intermediate process is provided for the user.
For specific definition of the deep learning platform, reference may be made to the above definition of the deep learning model detection method, which is not described herein again. The modules in the deep learning platform can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent of a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 3. The computer device comprises a processor, a memory, a network interface, a display screen and an input device which are connected through a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a deep learning model detection method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
It will be appreciated by those skilled in the art that the configuration shown in fig. 3 is a block diagram of only a portion of the configuration associated with the present application, and is not intended to limit the computing device to which the present application may be applied, and that a particular computing device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:
acquiring data to be detected, and converting the data to be detected into a first binary file for storage; selecting a training frame based on user input, and converting the training frame into a second binary file for storage; converting the first binary file into a model data file corresponding to the selected training frame; and obtaining a deep learning model output file according to the model data file and the second binary file.
In one embodiment, the processor when executing the computer program further performs the steps of:
acquiring to-be-detected data input by a user and a storage path of an output file based on the input parameters; the input parameters include: row parameters, column parameters, data type, image type, and normalization parameters.
In one embodiment, the processor when executing the computer program further performs the steps of:
converting the data to be detected into a first binary file; and writing the input parameters and the storage path of the output file into a data head of the first binary file for storage.
In one embodiment, the processor when executing the computer program further performs the steps of:
analyzing the first binary file to obtain input parameters in the data head and a storage path of the output file; and converting the first binary file into a model data file corresponding to the training frame according to the input parameters and the selected training frame.
In one embodiment, the processor when executing the computer program further performs the steps of:
obtaining a deep learning model output file according to the model data file and the second binary file; and storing the deep learning model output file according to the input parameters and the storage path of the output file.
In one embodiment, the processor when executing the computer program further performs the steps of:
a first binary file and a second binary file storage path are preset.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
the second binary file is a static binary file in a protocol buffer form.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored which, when executed by a processor, performs the steps of:
acquiring data to be detected, and converting the data to be detected into a first binary file for storage; selecting a training frame based on user input, and converting the training frame into a second binary file for storage; converting the first binary file into a model data file corresponding to the selected training frame; and obtaining a deep learning model output file according to the model data file and the second binary file.
In one embodiment, the computer program when executed by the processor further performs the steps of:
acquiring to-be-detected data input by a user and a storage path of an output file based on the input parameters; the input parameters include: row parameters, column parameters, data type, image type, and normalization parameters.
In one embodiment, the computer program when executed by the processor further performs the steps of:
converting the data to be detected into a first binary file; and writing the input parameters and the storage path of the output file into a data head of the first binary file for storage.
In one embodiment, the computer program when executed by the processor further performs the steps of:
analyzing the first binary file to obtain input parameters in the data head and a storage path of the output file; and converting the first binary file into a model data file corresponding to the training frame according to the input parameters and the selected training frame.
In one embodiment, the computer program when executed by the processor further performs the steps of:
obtaining a deep learning model output file according to the model data file and the second binary file; and storing the deep learning model output file according to the input parameters and the storage path of the output file.
In one embodiment, the computer program when executed by the processor further performs the steps of:
a first binary file and a second binary file storage path are preset.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), synchronous Link (Synchlink) DRAM (SLDRAM), Rambus (Rambus) direct RAM (RDRAM), direct bused dynamic RAM (DRDRAM), and bused dynamic RAM (RDRAM).
All possible combinations of the technical features in the above embodiments may not be described for the sake of brevity, but should be considered as being within the scope of the present disclosure as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is specific and detailed, but not to be understood as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent application shall be subject to the appended claims.
Claims (10)
1. A deep learning model detection method, the method comprising:
acquiring data to be detected, and converting the data to be detected into a first binary file for storage;
selecting a training frame based on user input, and converting the training frame into a second binary file for storage;
converting the first binary file into a model data file corresponding to the selected training frame;
obtaining a deep learning model output file according to the model data file and the second binary file;
the obtaining of the deep learning model output file according to the model data file and the second binary file comprises:
and inputting the model data file into a deep learning model, and calling the second binary file to obtain a deep learning model output file.
2. The method according to claim 1, wherein the acquiring data to be detected comprises:
acquiring to-be-detected data input by a user and a storage path of an output file based on the input parameters;
the input parameters include: row parameters, column parameters, data type, image type, and normalization parameters.
3. The method according to claim 2, wherein converting the data to be detected into the first binary file for storage comprises:
converting the data to be detected into a first binary file;
and writing the input parameters and the storage path of the output file into a data head of a first binary file for storage.
4. The method of claim 3, wherein converting the first binary file into the model data file corresponding to the selected training frame comprises:
analyzing the first binary file to obtain input parameters in the data head and a storage path of the output file;
and converting the first binary file into a model data file corresponding to the training frame according to the input parameters and the selected training frame.
5. The method of claim 4, wherein after inputting the model data file into the deep learning model and calling the second binary file to obtain the deep learning model output file, further comprising:
and storing the deep learning model output file according to the input parameters and the storage path of the output file.
6. The method of claim 1, further comprising:
a first binary file and a second binary file storage path are preset.
7. The method of claim 1,
the second binary file is a static binary file in a protocol buffer form.
8. A deep learning platform, comprising:
the preprocessing module is used for acquiring data to be detected input by a user and a storage path of an output file based on input parameters, converting the data to be detected into a first binary file, and writing the input parameters and the storage path of the output file into a data head of the first binary file for storage;
the model processing module is used for selecting a training frame based on user input and converting the training frame into a second binary file for storage;
the output module is used for analyzing the first binary file to obtain input parameters in the data head and a storage path of the output file; converting the first binary file into a model data file corresponding to the selected training frame; obtaining a deep learning model output file according to the model data file and the second binary file; storing the output file of the deep learning model according to the input parameters and the storage path of the output file;
and the output module is also used for inputting the model data file into a deep learning model and calling the second binary file to obtain a deep learning model output file.
9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor when executing the computer program performs the steps of the method according to any of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910926898.8A CN110705714B (en) | 2019-09-27 | 2019-09-27 | Deep learning model detection method, deep learning platform and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910926898.8A CN110705714B (en) | 2019-09-27 | 2019-09-27 | Deep learning model detection method, deep learning platform and computer equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110705714A CN110705714A (en) | 2020-01-17 |
| CN110705714B true CN110705714B (en) | 2022-07-22 |
Family
ID=69197117
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910926898.8A Active CN110705714B (en) | 2019-09-27 | 2019-09-27 | Deep learning model detection method, deep learning platform and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110705714B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114356787B (en) * | 2022-03-18 | 2022-08-16 | 江苏清微智能科技有限公司 | Automatic testing method and device for deep learning model compiler and storage medium |
| CN114629812A (en) * | 2022-03-28 | 2022-06-14 | 中国电子科技集团公司第三十八研究所 | Cluster visualization system and method based on autonomous controllable platform |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107423817A (en) * | 2017-04-17 | 2017-12-01 | 星环信息科技（上海）有限公司 | The method and apparatus that a kind of deep learning is realized |
| CN108153896A (en) * | 2018-01-08 | 2018-06-12 | 第四范式（北京）技术有限公司 | For input data, the processing method and processing device of output data |
| CN109146084A (en) * | 2018-09-06 | 2019-01-04 | 郑州云海信息技术有限公司 | A kind of method and device of the machine learning based on cloud computing |
| CN109635918A (en) * | 2018-10-30 | 2019-04-16 | 银河水滴科技（北京）有限公司 | The automatic training method of neural network and device based on cloud platform and preset model |
-
2019
- 2019-09-27 CN CN201910926898.8A patent/CN110705714B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107423817A (en) * | 2017-04-17 | 2017-12-01 | 星环信息科技（上海）有限公司 | The method and apparatus that a kind of deep learning is realized |
| CN108153896A (en) * | 2018-01-08 | 2018-06-12 | 第四范式（北京）技术有限公司 | For input data, the processing method and processing device of output data |
| CN109146084A (en) * | 2018-09-06 | 2019-01-04 | 郑州云海信息技术有限公司 | A kind of method and device of the machine learning based on cloud computing |
| CN109635918A (en) * | 2018-10-30 | 2019-04-16 | 银河水滴科技（北京）有限公司 | The automatic training method of neural network and device based on cloud platform and preset model |
Non-Patent Citations (1)
| Title |
|---|
| 基于深度学习的目标检测框架进展研究;寇大磊 等著;《计算机工程与应用 网络首发》;20190326;第25-34页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110705714A (en) | 2020-01-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020078236A1 (en) | Two-dimensional code generation method and apparatus, storage medium and electronic device | |
| CN110020582B (en) | Face emotion recognition method, device, equipment and medium based on deep learning | |
| US10878570B2 (en) | Knockout autoencoder for detecting anomalies in biomedical images | |
| CN111931865A (en) | Training method and device of image classification model, computer equipment and storage medium | |
| CN110827236B (en) | Brain tissue layering method, device and computer equipment based on neural network | |
| CN110705714B (en) | Deep learning model detection method, deep learning platform and computer equipment | |
| CN116486422A (en) | Data processing method and related equipment | |
| CN109711545A (en) | Creation method, device, system and the computer-readable medium of network model | |
| CN113313199A (en) | Brain CT medical report automatic generation method based on weak supervision attention | |
| US20230245483A1 (en) | Handwriting recognition method and apparatus, and electronic device and storage medium | |
| Soullard et al. | Ctcmodel: a keras model for connectionist temporal classification | |
| CN112613425A (en) | Target identification method and system for small sample underwater image | |
| CN117036933A (en) | Intelligent crop pest control system and method | |
| CN116109980A (en) | Action recognition method based on video text matching | |
| CN113516182B (en) | Visual question-answering model training and visual question-answering method and device | |
| CN116052714A (en) | Data processing method and device | |
| CN113850236B (en) | 3D attitude estimation method and device, computer equipment and storage medium | |
| CN112329744B (en) | Picture character recognition method and device | |
| US20240019931A1 (en) | Methods and systems for eye-gaze location detection and accurate collection of eye-gaze data | |
| CN113011555B (en) | Data processing method, device, equipment and storage medium | |
| Sharma et al. | Cotton Crop Disease Detection System Using Deep Learning Approach | |
| KR20230076075A (en) | Method and apparatus for optimal layer-wise execution policy prediction based on supervised learning method in systolic array based npu environment for inference | |
| CN117392491A (en) | Data processing method and device | |
| Akram et al. | From Data Quality to Model Performance: Navigating the Landscape of Deep Learning Model Evaluation | |
| Ahmed et al. | KERAS DEEP LEARNING PACKAGE IN PYTHON: A REVIEW |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 201807 Shanghai City, north of the city of Jiading District Road No. 2258 Applicant after: Shanghai Lianying Medical Technology Co.,Ltd. Address before: 201807 Shanghai City, north of the city of Jiading District Road No. 2258 Applicant before: SHANGHAI UNITED IMAGING HEALTHCARE Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |