-
Notifications
You must be signed in to change notification settings - Fork 1.8k
update doc: NNI core capacities in Overview.md, standalone mode in Trials.md #1821
update doc: NNI core capacities in Overview.md, standalone mode in Trials.md #1821
Conversation
@@ -127,6 +127,23 @@ In the YAML configure file, you need to set *useAnnotation* to true to enable NN | |||
useAnnotation: true | |||
``` | |||
|
|||
## Standalone mode for debug |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Add this to "Overview>Key Concepts" section too?
docs/en_US/TrialExample/Trials.md
Outdated
@@ -127,6 +127,23 @@ In the YAML configure file, you need to set *useAnnotation* to true to enable NN | |||
useAnnotation: true | |||
``` | |||
|
|||
## Standalone mode for debug | |||
|
|||
NNI supports standalone mode for trial code to run without starting an NNI experiment. This is for finding out bugs in trial code more conveniently. NNI annotation natively supports standalone mode, as the added NNI related lines are comments. For NNI trial APIs, the APIs have changed behaviors in standalone mode, some APIs return fake/empty values, and some APIs do not really report values. Details are shown below. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
"standalone mode" sounds confusing... may we call it "single trial debug mode" or something more clear connect to "trial".
docs/en_US/TrialExample/Trials.md
Outdated
@@ -127,6 +127,23 @@ In the YAML configure file, you need to set *useAnnotation* to true to enable NN | |||
useAnnotation: true | |||
``` | |||
|
|||
## Standalone mode for debug | |||
|
|||
NNI supports standalone mode for trial code to run without starting an NNI experiment. This is for finding out bugs in trial code more conveniently. NNI annotation natively supports standalone mode, as the added NNI related lines are comments. For NNI trial APIs, the APIs have changed behaviors in standalone mode, some APIs return fake/empty values, and some APIs do not really report values. Details are shown below. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
For NNI trial APIs, the APIs have changed behaviors in standalone mode, some APIs return fake/empty values, and some APIs do not really report values. Details are shown below.
In trial debug model, experiment related fields are filled with dummy values, please refer to the following table for the full lists of these fields.
@@ -46,6 +46,33 @@ For each experiment, user only needs to define a search space and update a few l | |||
|
|||
More details about how to run an experiment, please refer to [Get Started](Tutorial/QuickStart.md). | |||
|
|||
## Core Features |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm thinking to replace this or put this ahead of the current "easy-to-use", "scalability", etc. section. let's discuss more in our scrum today.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Approved for quick iterative update.
describe core capacities of NNI, including hyperparameter tuning, neural architecture search, model compression, and feature engineering.
describe standalone mode, in which trial code can run without NNI experiment for easy debug.