-
Fix loading weights in n-dim dense -> 1x1 conv 7 of 7 checklist items completed!821
- Merged
- 2
updated -
Support for parsing nested models 7 of 7 checklist items completed!794
- Merged
- 2
updated -
- Merged
- 2
updated -
Improved parsing of pytorch models using torch.FX - Clean 8 of 8 checklist items completed!799 v0.8.0
- Merged
- 3
updated -
Fix Vitis Conv1D/2D latency strategy 1 of 1 checklist item completed!815
- Merged
- 3
updated -
Propagate zeros from Conv layers to multiplication config 7 of 7 checklist items completed!797
- Merged
- 2
updated -
Fix over-allocation of bits for quantised po2 7 of 7 checklist items completed!806
- Merged
- 2
updated -
- Merged
- 2
updated -
Discussion - Inlined Conv slows down latency significantly (up to x15 - x20) 9 of 9 checklist items completed!800
- Merged
- 5
updated -
- Merged
- 2
updated -
Fix parsing of QConv2DBatchnorm weights 7 of 7 checklist items completed!802
- Merged
- 2
updated -
- Merged
- 8
updated -
Remove tf_to_hls 7 of 7 checklist items completed!795
- Merged
- 3
updated -
Don't use reader in ModelGraph and layers 6 of 6 checklist items completed!770 v0.8.0
- Merged
- 10
updated -
Decouple pipeline style from strategy 7 of 8 checklist items completed!781 v0.8.0
- Merged
-
- 7
updated -
bump 0.7.1 2 of 2 checklist items completed!791
- Merged
-
- 1
- Approved
updated -
Minor docs fixes for 0.7.1 1 of 1 checklist item completed!788
- Merged
- 3
updated -
Explicitly set strategy for pointwise conv 7 of 7 checklist items completed!785
- Merged
- 2
updated -
Update Jenkins test environment to avoid dependency hell 7 of 7 checklist items completed!786
- Merged
- 3
updated -
Fix RNN layers when strategy=resource 7 of 7 checklist items completed!780
- Merged
- 2
updated
