[IMGREC-40] Train custom neural network for image recognition Created: 26/Oct/18 Updated: 22/Aug/19 Resolved: 09/Jan/19 |
|
| Status: | Closed |
| Project: | Image Recognition |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Improvement | Priority: | Neutral |
| Reporter: | Cedric Reichenbach | Assignee: | Cedric Reichenbach |
| Resolution: | Done | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||
| Template: |
|
||||||||
| Acceptance criteria: |
Empty
|
||||||||
| Task DoD: |
[ ]*
Doc/release notes changes? Comment present?
[ ]*
Downstream builds green?
[ ]*
Solution information and context easily available?
[ ]*
Tests
[ ]*
FixVersion filled and not yet released
[ ] 
Architecture Decision Record (ADR)
|
||||||||
| Date of First Response: | |||||||||
| Epic Link: | Periscope improvements | ||||||||
| Sprint: | Basel 161, Foundation 1 | ||||||||
| Story Points: | 13 | ||||||||
| Description |
|
See So far, we're using a provided model from the dl4j zoo with pre-trained weights for the ImageNet-1000 dataset, which is not a good fit for general-purpose image recognition (see Since there seem to be no pre-trained networks available that are a better fit, we should train our own. However, complete training from scratch should not be necessary; just doing transfer learning by "fine-tuning" an existing pre-trained model should be enough: Only replace the output layer with one that fits our new number of classes, then freeze all other layers and train on e.g. ImageNet data. Documentation about transfer learning with dl4j TBD: Classes (labels) we want to support |
| Comments |
| Comment by Antti Hietala [ 11/Dec/18 ] |
I propose Core Wordnet 5000, see |
| Comment by Cedric Reichenbach [ 11/Dec/18 ] |
|
ahietala sounds like a good idea. However, there are a couple potential issues:
|