Systems and methods for automatically serializing and deserializing models
A system serializing and deserializing models configured to (i) store a first model, wherein the first model includes a plurality of functionalities; (ii) generate a human-readable document based on the first model, wherein the human-readable document describes the first model; (iii) generate a second model based on the human-readable document, wherein the second model includes the plurality of functionalities; (iv) train the second model; (v) generate a new human-readable document based on the trained second model; and (vi) generate an updated second model based on the new human-readable document.
1 . A computer system for iteratively generating and updating a model based on a human-readable document, the computer system including at least one processor in communication with at least one memory device, the at least one processor is programmed to:
store a first model, wherein the first model is an original model stored in the at least one memory device, wherein the first model includes a first set of functionalities;
generate a first human-readable document based at least in part upon the first model, the first human-readable document describing the first set of functionalities and embedding (i) a series of hidden identifiers and (ii) one or more tags related to the first set of functionalities in the first human-readable document;
generate a second model, wherein the second model is a deserialized product generated from the first human-readable document by translating data from the first human-readable document using the series of hidden identifiers and the one or more tags, wherein the second model has a second set of functionalities based on the first set of functionalities of the first model;
train the second model, using at least Bayesian Program Learning, to alter one or more functionalities of the second set of functionalities;
determine, based on a comparison of one or more functionalities of the second set of functionalities, as altered, to one or more functionalities of the first set of functionalities, whether the one or more functionalities of the second set of functionalities, as altered, is altered more than one or more pre-determined thresholds, wherein the comparison is performed by:
executing a set of input data (1) on the first model to receive a first set of results and (2) on the second model to receive a second set of results; and
comparing the first set of results with the second set of results to evaluate differences between the one or more functionalities of the second set of functionalities, as altered, and the one or more functionalities of the first set of functionalities;
after determining that the one or more functionalities of the second set of functionalities is altered more than the one or more pre-determined thresholds, convert the first human-readable document into a new human-readable document by updating the series of hidden identifiers and the one or more tags;
generate an updated second model based at least in part upon the new human-readable document, wherein the second model, as updated, includes the one or more functionalities of the second set of functionalities, as altered, in the new human-readable document; and
train the updated second model, using at least Bayesian Program Learning, to further refine the second set of functionalities, as altered, based on the new human-readable document.
2 . The computer system in accordance with claim 1 , wherein the at least one processor is further programmed to:
receive a second set of input data;
perform an execution of the second model for each of the second set of input data; and
train the second model based on the executions of the second model for the second set of input data using at least Bayesian Program Learning to alter one or more functionalities of the second set of functionalities.
3 . The computer system in accordance with claim 1 , wherein the at least one processor is further programmed to:
receive a request by a user to access a website; and
perform an execution of the second model based on the request by the user.
4 . The computer system in accordance with claim 1 , wherein the at least one processor is further programmed to:
store the first human-readable document in a web accessible location;
generate a link to the web accessible location; and
integrate the link into a website associated with the second model.
5 . The computer system in accordance with claim 4 , wherein the at least one processor is further programmed to replace the first human-readable document with the new human-readable document.
6 . The computer system in accordance with claim 1 , wherein the at least one processor is further programmed to:
receive an initial set of input data; and
train the first model based on the initial set of input data.
7 . The computer system in accordance with claim 1 , wherein the at least one processor is further programmed to:
receive a template; and
compare the template to at least one of the first model or the second model to create a comparison,
wherein at least one of:
generating the first human-readable document includes generating the first human-readable document based on the comparison; or
generating the new human-readable document includes generating the new human-readable document based on the comparison.
8 . The computer system in accordance with claim 7 , wherein the template includes one or more tags, and wherein the at least one processor is further programmed to:
compare each tag of the one or more tags to functionalities of the first model to identify a functionality feature of the first model that matches a corresponding tag of the one or more tags; and
apply the functionality to at least one location associated with the corresponding tag.
9 . The computer system in accordance with claim 8 , wherein generating the second model includes generating the second model based on the template and the first human-readable document.
10 . The computer system in accordance with claim 9 , wherein the at least one processor is further programmed to:
identify one or more functionalities associated with one or more tagged inputs based on one or more placeholders in the template; and
generate the second model to include the one or more functionalities associated with the one or more tagged inputs and to exclude text not associated with the one or more tagged inputs.
11 . The computer system in accordance with claim 1 , wherein the at least one processor is further programmed to:
store a set of templates, wherein each template of the set of templates is associated with a different human-readable document; and
receive a selection of a first template of the set of templates by a user,
wherein generating the first human-readable document includes generating the first human-readable document based on the first template and the first set of functionalities of the first model.
12 . A computer-based method for iteratively generating and updating a model based on a human-readable document, the computer-based method is implemented on a serialization/deserialization (“SD”) computer device including at least one processor in communication with at least one memory device, the computer-based method comprising:
storing, in the memory device, a first model, wherein the first model is an original model stored in the at least one memory device, wherein the first model includes a first set of functionalities;
generating, by the processor, a first human-readable document based at least in part upon the first model, the first human-readable document describing the first set of functionalities and embedding (i) a series of hidden identifiers and (ii) one or more tags related to the first set of functionalities in the first human-readable document;
generating, by the processor, a second model, wherein the second model is a deserialized product generated from the first human-readable document by translating data from the first human-readable document using the series of hidden identifiers and the one or more tags, wherein the second model has a second set of functionalities based on the first set of functionalities of the first model;
training the second model, using at least Bayesian Program Learning to alter one or more functionalities of the second set of functionalities;
determining, based on a comparison of one or more functionalities of the second set of functionalities, as altered, to one or more functionalities of the first set of functionalities, whether the one or more functionalities of the second set of functionalities, as altered, is altered-more than one or more pre-determined thresholds, wherein the comparison is performed by:
executing a set of input data (1) on the first model to receive a first set of results and (2) on the second model to receive a second set of results; and
comparing the first set of results with the second set of results to evaluate differences between the one or more functionalities of the second set of functionalities, as altered, and the one or more functionalities of the first set of functionalities;
after determining that the one or more functionalities of the second set of functionalities is altered more than the one or more pre-determined thresholds, converting, by the processor, the first human-readable document into a new human-readable document by updating the series of hidden identifiers and the one or more tags;
generate an updated second model based at least in part upon the new human-readable document, wherein the second model, as updated, includes the one or more functionalities of the second set of functionalities, as altered, in the new human-readable document; and
train the updated second model, using at least Bayesian Program Learning to further refine the second set of functionalities, as altered, based on the new human-readable document.
13 . The computer-based method of claim 12 , wherein training the second model further comprises:
receiving a second set of input data;
performing an execution of the second model for each of the second set of input data; and
training the second model based on the executions of the second model for the second set of input data using at least Bayesian Program Learning to alter one or more functionalities of the second set of functionalities.
14 . The method of claim 12 , further comprising:
receiving a request by a user to access a website; and
performing an execution of the second model based on the request by the user.
15 . The method of claim 12 , further comprising:
storing the first human-readable document in a web accessible location;
generating a link to the web accessible location; and
integrating the link into a website associated with the second model.
16 . The method of claim 15 , further comprising replacing the first human-readable document with the new human-readable document.
17 . The method of claim 12 , further comprising:
receiving an initial set of input data; and
training the first model based on the initial set of input data.
18 . The method of claim 12 , further comprising:
receiving a template; and
comparing the template to at least one of the first model or the second model to create a comparison,
wherein at least one of:
generating the first human-readable document includes generating the first human-readable document based on the comparison; or
generating the new human-readable document includes generating the new human-readable document based on the comparison.
19 . A non-transitory computer-readable medium having computer-executable instructions embodied thereon, wherein when executed by at least one processor, the computer-executable instructions cause the at least one processor to:
store a first model, wherein the first model is an original model stored in at least one memory device, wherein the first model includes a first set of functionalities;
generate a first human-readable document based at least in part upon the first model, the first human-readable document describing the first set of functionalities and embedding (i) a series of hidden identifiers and (ii) one or more tags related to the first set of functionalities in the first human-readable document;
generate a second model, wherein the second model is a deserialized product generated from the first human-readable document by translating data from the first human-readable document using the series of hidden identifiers and the one or more tags, wherein the second model has a second set of functionalities based on the first set of functionalities of the first model;
train the second model, using at least Bayesian Program Learning to alter one or more functionalities of the second set of functionalities;
determine, based on a comparison of one or more functionalities of the second set of functionalities, as altered, to one or more functionalities of the first set of functionalities, whether the one or more functionalities of the second set of functionalities, as altered, is altered more than one or more pre-determined thresholds, wherein the comparison is performed by:
executing a set of input data (1) on the first model to receive a first set of results and (2) on the second model to receive a second set of results; and
comparing the first set of results with the second set of results to evaluate differences between the one or more functionalities of the second set of functionalities, as altered, and the one or more functionalities of the first set of functionalities;
after determining that the one or more functionalities of the second set of functionalities is altered more than the one or more pre-determined thresholds, convert the first human-readable document into a new human-readable document by updating the series of hidden identifiers and the one or more tags;
generate an updated second model based at least in part upon the new human-readable document, wherein the second model, as updated, includes the one or more functionalities of the second set of functionalities, as altered, in the new human-readable document; and
train the updated second model, using at least Bayesian Program Learning, to further refine the second set of functionalities, as altered, based on the new human-readable document.
20 . The non-transitory computer-readable medium of claim 19 , wherein the at least
one processor is further programmed to:
receive a request by a user to access a website; and
perform an execution of the second model based on the request by the user.