Difference between revisions of "Machine Learning Services"

From DAVE Developer's Wiki
Jump to: navigation, search
(2 intermediate revisions by the same user not shown)
Line 3: Line 3:
 
!colspan="3" style="width:100%; border-left:solid 2px #ededed; border-right:solid 0px #ededed; border-top:solid 2px #ededed;border-bottom:solid 0px #ededed; background-color:#ededed; font-size:14px; color:#000000; text-align:center" | Machine Learning Services
 
!colspan="3" style="width:100%; border-left:solid 2px #ededed; border-right:solid 0px #ededed; border-top:solid 2px #ededed;border-bottom:solid 0px #ededed; background-color:#ededed; font-size:14px; color:#000000; text-align:center" | Machine Learning Services
 
|-
 
|-
|style="width:33%; border-left:solid 2px #ededed;border-right:solid 0px #ededed;border-top:solid 0px #ededed;border-bottom:solid 2px #ededed; text-align:left; vertical-align:top; background-color:#ffffff"| <br />[[File:TBD.png|none|200px|center]]<br />
+
|style="width:33%; border-left:solid 2px #ededed;border-right:solid 0px #ededed;border-top:solid 0px #ededed;border-bottom:solid 2px #ededed; text-align:left; vertical-align:top; background-color:#ffffff"| <br />[[File:NeuralNetwork.png|none|x150px|center]]<br />
'''Machine Learning TBD'''
+
'''Model design'''
* TBD
+
* PoC design for dataset acquisition
* TBD
+
* Dataset recording
|style="width:33%; border-left:solid 0px #ededed;border-right:solid 0px #ededed;border-top:solid 0px #ededed;border-bottom:solid 2px #ededed; text-align:left; vertical-align:top; background-color:#ffffff"| <br />[[File:TBD.png|none|200px|center|]]<br />
+
* Model design and definition (using the most popular techniques depending on the Scope of Supply)
'''Machine Learning TBD'''
+
* Dataset analysis and preparation for Learning and Training phase
* TBD
+
|style="width:33%; border-left:solid 0px #ededed;border-right:solid 0px #ededed;border-top:solid 0px #ededed;border-bottom:solid 2px #ededed; text-align:left; vertical-align:top; background-color:#ffffff"| <br />[[File:002425_%2B0.003355520.png|none|x150px|center|]]<br />
* TBD
+
'''Model Learning and Training'''
|style="width:33%; border-left:solid 0px #ededed;border-right:solid 2px #ededed;border-top:solid 0px #ededed;border-bottom:solid 2px #ededed; text-align:left; vertical-align:top; background-color:#ffffff"|<br /> [[File:TBD.png|none|200px|center|]]<br />
+
* Model deployment on target
'''Machine Learning TBD'''
+
* Design of Learning machine on local or cloud solution
* TBD
+
* test bench based solution for accuracy evaluation
* TBD
+
* Deployment on the field and fine tuning
 +
|style="width:33%; border-left:solid 0px #ededed;border-right:solid 2px #ededed;border-top:solid 0px #ededed;border-bottom:solid 2px #ededed; text-align:left; vertical-align:top; background-color:#ffffff"|<br /> [[File:ML-TN-001-MPSoC-PL1.png|none|x150px|center|]]<br />
 +
'''Continuous Learning and Update'''
 +
* IoT based continuous recording of new dataset information
 +
* New dataset information review and integration
 +
* Continuous model Training and Learning
 +
* Deployment OTA on the field of new model updates
 
|-
 
|-
 
|}
 
|}
Line 45: Line 51:
 
| style="width:1%; border-left:solid 0px #ededed;border-right:solid 0px #ededed;border-top:solid 0px #ededed;border-bottom:solid 2px #ffffff; background-color:#ffffff" |
 
| style="width:1%; border-left:solid 0px #ededed;border-right:solid 0px #ededed;border-top:solid 0px #ededed;border-bottom:solid 2px #ffffff; background-color:#ffffff" |
 
| style="width:25%; border-left:solid 2px #ededed;border-right:solid 2px #ededed;border-top:solid 2px #ededed;border-bottom:solid 2px #ededed; background-color:#ffffff; vertical-align:top"|
 
| style="width:25%; border-left:solid 2px #ededed;border-right:solid 2px #ededed;border-top:solid 2px #ededed;border-bottom:solid 2px #ededed; background-color:#ffffff; vertical-align:top"|
Depending on the requested task, the project may require a regular service for updates and improvements (such as the CVE monitoring and resolution via updated version of the Linux kernel). In this case, this service is released regularly via sw updates in our Gitlab server with release candidates and official releases
+
Depending on the requested task, the project may require a regular service for updates and improvements (such as the improvement s on ML model thanks to an increased/modified dataset). In this case, this service is released regularly via sw updates in our Gitlab server with release candidates and official releases
 
|-
 
|-
 
|}
 
|}
Line 71: Line 77:
 
| style="width:30%; border-left:solid 2px #ededed;border-right:solid 2px #ededed;border-top:solid 2px #ededed;border-bottom:solid 2px #ededed; background-color:#ffffff; vertical-align:top"|
 
| style="width:30%; border-left:solid 2px #ededed;border-right:solid 2px #ededed;border-top:solid 2px #ededed;border-bottom:solid 2px #ededed; background-color:#ffffff; vertical-align:top"|
 
* [[:Category:Machine Learning | Category Machine Learning Services]]
 
* [[:Category:Machine Learning | Category Machine Learning Services]]
* [[TBD | Machine Learning Application Notes]]
+
* [[:Category:Machine Learning AN| Machine Learning Application Notes]]
* [[TBD | Machine Learning Technical Notes]]
+
* [[:Category:Machine Learning TN| Machine Learning Technical Notes]]
* [[TBD | Machine Learning White Papers]]
+
* [[:Category:Machine Learning WP| Machine Learning White Papers]]
* [[TBD | Machine Learning Case Histories]]
+
* [[:Category:Machine Learning CH| Machine Learning Case Histories]]
 
|-
 
|-
 
|}
 
|}
  
 
[[Category:Machine Learning]]
 
[[Category:Machine Learning]]

Revision as of 13:03, 5 January 2021

Machine Learning Services

NeuralNetwork.png

Model design

  • PoC design for dataset acquisition
  • Dataset recording
  • Model design and definition (using the most popular techniques depending on the Scope of Supply)
  • Dataset analysis and preparation for Learning and Training phase

002425 +0.003355520.png

Model Learning and Training

  • Model deployment on target
  • Design of Learning machine on local or cloud solution
  • test bench based solution for accuracy evaluation
  • Deployment on the field and fine tuning

ML-TN-001-MPSoC-PL1.png

Continuous Learning and Update

  • IoT based continuous recording of new dataset information
  • New dataset information review and integration
  • Continuous model Training and Learning
  • Deployment OTA on the field of new model updates


Machine Learning Services: what we offer in short Machine Learning Services: the on board mechanism Machine Learning Services: the deployment Machine Learning Services: the maintenance/upgrade

DAVE Embedded Systems offers the above services with two possible approaches:

  • based on binding formal quotation after discussions and effort evaluation
  • based on Time & Material approach with an initial estimation of the effort

The standard approach requires an initial contact with the technical team. Customers together the team defines:

  • The technical specification
  • The scope of Supply
  • The Acceptance Criterias jointly used for evaluate the task and approve it

Depending on the requested task, the project is managed via gantt approach with multiple tasks with peer review (both internals and shared with customers). The typical deployment is then shared with customers via a dedicated wiki structure (like this but private between DAVE Embedded Systems and customers) and a dedicated project branch on our Gitlab server.

Depending on the requested task, the project may require a regular service for updates and improvements (such as the improvement s on ML model thanks to an increased/modified dataset). In this case, this service is released regularly via sw updates in our Gitlab server with release candidates and official releases


General Information Pricing Insights
  • [link_TBD {{{nome-som}}} SOM Brochure PDF ]
Customer-service.png Online Technical Helpdesk
Sale-tag.png Machine Learning Services RFQ Request For Quotation