Chatbot for defining a machine learning (ML) solution
The present disclosure relates to systems and methods for an intelligent assistant (e.g., a chatbot) that can be used to enable a user to generate a machine learning system. Techniques can be used to automatically generate a machine learning system to assist a user. In some cases, the user may not be a software developer and may have little or no experience in either machine learning techniques or software programming. In some embodiments, a user can interact with an intelligent assistant. The interaction can be aural, textual, or through a graphical user interface. The chatbot can translate natural language inputs into a structural representation of a machine learning solution using an ontology. In this way, a user can work with artificial intelligence without being a data scientist to develop, train, refine, and compile machine learning models as stand-alone executable code.
1. A computer-implemented method comprising:
receiving, from a user device, a first input;
predicting, using a computational technique, a type of desired result based on the first input;
identifying, using metadata associated with each machine-learning-model framework of a set of machine-learning-model frameworks, one or more machine-learning-model frameworks based on the predicted type of desired result, wherein the one or more machine-learning-model frameworks are of the set of machine-learning-model frameworks;
presenting, for each machine-learning-model framework of the one or more machine-learning-model frameworks, a representation of a corresponding machine-learning-model architecture on a display, wherein one or more machine-learning-model architectures are presented;
receiving a second input identifying a selection of a particular machine-learning-model architecture of the one or more machine-learning-model architectures;
receiving a third input identifying a data source for generating a machine learning architecture;
receiving a fourth input identifying one or more constraints for the machine learning architecture;
generating code for a machine learning model based at least in part on the second input identifying the selection of the particular machine-learning-model architecture, the third input identifying the data source, and the fourth input identifying the one or more constraints; and
storing the generated code in a memory.
2. The computer-implemented method of claim 1 , wherein the particular machine-learning-model architecture includes a neural network.
3. The computer-implemented method of claim 1 , wherein the particular machine-learning-model architecture includes a classifier network.
4. The computer-implemented method of claim 1 , further comprising:
analyzing the one or more constraints to generate a second plurality of code for the particular machine-learning-model architecture based at least in part on optimizing the one or more constraints;
generating an optimized solution; and
displaying the optimized solution.
5. The computer-implemented method of claim 1 , further comprising deploying the particular machine-learning-model architecture via an intelligent assistant interface.
6. The computer-implemented method of claim 1 , wherein the first input includes textual input.
7. The computer-implemented method of claim 1 , wherein the one or more constraints includes a constraint pertaining to at least one of resources, location, security, or privacy.
8. A system comprising:
one or more data processors; and
a non-transitory computer readable storage medium containing instructions which, when executed on the one or more data processors, cause the one or more data processors to perform a set of actions including:
receiving, from a user device, a first input;
predicting, using a computational technique, a type of desired result based on the first input;
identifying, using metadata associated with each machine-learning-model framework of a set of machine-learning-model frameworks, one or more machine-learning-model frameworks based on the predicted type of desired result, wherein the one or more machine-learning-model frameworks are of the set of machine-learning-model frameworks;
presenting, for each machine-learning-model framework of the one or more machine-learning-model frameworks, a representation of a corresponding machine-learning-model architecture on a display, wherein one or more machine-learning-model architectures are presented;
receiving a second input identifying a selection of a particular machine-learning-model architecture of the one or more machine-learning-model architectures;
receiving a third input identifying a data source for generating a machine learning architecture;
receiving a fourth input identifying one or more constraints for the machine learning architecture;
generating code for a machine learning model based at least in part on the second input identifying the selection of the particular machine-learning-model architecture, the third input identifying the data source, and the fourth input identifying the one or more constraints; and
storing the generated code in a memory.
9. The system of claim 8 , wherein the particular machine-learning-model architecture includes a neural network.
10. The system of claim 8 , wherein the particular machine-learning-model architecture includes a classifier network.
11. The system of claim 8 , wherein the set of actions further includes:
analyzing the one or more constraints to generate a second plurality of code for the particular machine-learning-model architecture based at least in part on optimizing the one or more constraints;
generating an optimized solution; and
displaying the optimized solution.
12. The system of claim 8 , wherein the set of actions further includes deploying the particular machine-learning-model architecture via an intelligent assistant interface.
13. The system of claim 8 , wherein the first input includes textual input.
14. The system of claim 8 , wherein the one or more constraints includes a constraint pertaining to at least one of resources, location, security, or privacy.
15. A computer-program product tangibly embodied in a non-transitory machine-readable storage medium, including instructions configured to cause one or more data processors to perform a set of actions including:
receiving, from a user device, a first input;
predicting, using a computational technique, a type of desired result based on the first input;
identifying, using metadata associated with each machine-learning-model framework of a set of machine-learning-model frameworks, one or more machine-learning-model frameworks based on the predicted type of desired result, wherein the one or more machine-learning-model frameworks are of the set of machine-learning-model frameworks;
presenting, for each machine-learning-model framework of the one or more machine-learning-model frameworks, a representation of a corresponding machine-learning-model architecture on a display, wherein one or more machine-learning-model architectures are presented;
receiving a second input identifying a selection of a particular machine-learning-model architecture of the one or more machine-learning-model architectures;
receiving a third input identifying a data source for generating a machine learning architecture;
receiving a fourth input identifying one or more constraints for the machine learning architecture;
generating code for a machine learning model based at least in part on the second input identifying the selection of the particular machine-learning-model architecture, the third input identifying the data source, and the fourth input identifying the one or more constraints; and
storing the generated code in a memory.
16. The computer-program product of claim 15 , wherein the particular machine-learning-model architecture includes a neural network.
17. The computer-program product of claim 15 , wherein the particular machine-learning-model architecture includes a classifier network.
18. The computer-program product of claim 15 , wherein the set of actions further includes:
analyzing the one or more constraints to generate a second plurality of code for the particular machine-learning-model architecture based at least in part on optimizing the one or more constraints;
generating an optimized solution; and
displaying the optimized solution.
19. The computer-program product of claim 15 , wherein the set of actions further includes deploying the particular machine-learning-model architecture via an intelligent assistant interface.
20. The computer-program product of claim 15 , wherein the first input includes textual input.