GUI for interacting with analytics provided by machine-learning services
View Patent ↗A data pipeline tool provides a machine-learning design interface that a user can utilize (e.g., via an electronic device such as a personal computer, tablet, or smart phone) to design or configure data pipelines or workflows defining the manner in which ML models are developed, trained, tested, validated, or deployed. Once deployed, a designed ML model may generate predictive results based on input data fed to the ML model. The tool may present the predictive results via a GUI, and may enable a user to mark-up or otherwise interact with those predictive results. The tool may enable the user to share the results (which may include a mark-up or annotation provided by a user).
1. A system for visualizing and sharing machine-learning results, the system comprising:
(A) a display;
(B) a user input component;
(C) one or more processors coupled to the display and to the user input component;
(D) one or more memories, coupled to the one or more processors, storing a set of instructions that, when executed, cause the one or more processors to:
(i) retrieve, from the one or more memories, a machine-learning (ML) model trained using training data, the training data comprising a first plurality of instances, each instance of the first plurality of instances including:
first known values for a set of input features, and
an actual value of a target feature, the actual value corresponding to the first known values;
(ii) retrieve, from the one or more memories, a second plurality of instances, each instance of the second plurality of instances including:
second known values for the set of input features;
(iii) generate a predicted value of the target feature, for each instance of the second plurality of instances, by applying the ML model to the second known values, wherein each predicted value corresponds to a particular instance in the second plurality of instances;
(iv) display, via the display, a diagram including a visual representation of the predicted values;
(v) receive, via the user input component, user input representing a selection of a first subset of the predicted values;
(vi) update the diagram to include an annotation that visually distinguishes the first subset from a second subset of the predicted values exclusive of the first subset; and
(vii) provide data representing the first subset to a second computing device.
2. The system of claim 1 ,
wherein the data representing the first subset includes: the diagram including the visual representation of the predicted values, including the annotation visually distinguishing the first subset from the second subset,
wherein the second computing device analyzes the data to generate and display the visual representation of the first subset by: analyzing the data to generate and display (a) the visual representation of the predicted values, and (b) the annotation visually distinguishing the first subset from the second subset.
3. The system of claim 1 , wherein the data is stored as an image file.
4. The system of claim 1 , wherein the diagram is a two axis graph, wherein a first axis represents a one of the set of input features and wherein a second axis represents the target feature.
5. The system of claim 1 , wherein the diagram is a three axis graph, wherein: (i) a first axis represents a first feature from the set of input features, (ii) a second axis represents a second feature from the set of input features, and (iii) a third axis represents the target feature.
6. The system of claim 1 , wherein the diagram is a multi-axis graph including more than three axes, wherein: (i) a first axis represents the target feature; and (ii) three or more axes represent three or more features from the set of input features.
7. The system of claim 1 , wherein the annotation comprises a shape configured by a user to encompass the first subset.
8. The system of claim 1 , wherein the annotation comprises text identifying the first subset.
9. The system of claim 1 , wherein the annotation comprises a graphic indicating the first subset.
10. A method for visualizing and sharing machine-learning results, the method comprising:
(A) implementing a first set of operations by a first one or more computing devices, including:
(i) retrieving from memory a machine-learning (ML) model trained using training data, the training data comprising a first plurality of instances, each instance of the first plurality of instances including:
first known values for a set of input features, and
an actual value of a target feature, the actual value corresponding to the first known values;
(ii) retrieving from memory a second plurality of instances each instance of the second plurality of instances including:
second known values for the set of input features;
(iii) generating a predicted value of the target feature, for each instance of the second plurality of instances, by applying the ML model to the second known values, wherein predicted value corresponds to a particular instance in the second plurality of instances;
(iv) displaying a diagram including a visual representation of the predicted values;
(v) receiving user input representing a selection of a first subset of the predicted values;
(vi) updating the diagram to include an annotation that visually distinguishes the first subset from a second subset of the predicted values exclusive of the first subset; and
(vii) providing data representing the first subset to a second computing device; and
(B) implementing a second set of operations by the second computing device, including:
(i) receiving the data; and
(ii) analyzing the data to generate and display a visual representation of the first subset.
11. The method of claim 10 ,
wherein the data representing the first subset includes: the diagram including the visual representation of the predicted values, including the annotation visually distinguishing the first subset from the second subset,
wherein analyzing the data to generate and display the visual representation of the first subset includes: analyzing the data to generate and display (a) the visual representation of the predicted values, and (b) the annotation visually distinguishing the first subset from the second subset.
12. The method of claim 10 , wherein the data is stored as an image file.
13. The method of claim 10 , wherein the diagram is a two axis graph, wherein a first axis represents a one of the set of input features and wherein a second axis represents the target feature.
14. The method of claim 10 , wherein the diagram is a three axis graph, wherein: (i) a first axis represents a first feature from the set of input features, (ii) a second axis represents a second feature from the set of input features, and (iii) a third axis represents the target feature.
15. The method of claim 10 , wherein the diagram is a multi-axis graph including more than three axes, wherein: (i) a first axis represents the target feature; and (ii) three or more axes represent three or more features from the set of input features.
16. The method of claim 10 , wherein the annotation comprises a shape configured by a user to encompass the first subset.
17. The method of claim 10 , wherein the annotation comprises text identifying the first subset.
18. The method of claim 10 , wherein the annotation comprises a graphic indicating the first subset.
19. The method of claim 16 , wherein an outline of the shape includes one or more of a set of straight lines and curved lines.
20. The method of claim 16 , wherein the shape further includes a filling of a level of transparency configured by the user.