Skip to content

Tags: TarteelAI/NeMo

Tags

v1.8.2

Toggle v1.8.2's commit message
update branch

Signed-off-by: ericharper <[email protected]>

v1.8.1

Toggle v1.8.1's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
[TTS] Restore_buffer bug fix and update NeMo checkpoint URL (NVIDIA#4041

)

* restore_buffer bug fix and update NeMo checkpoint URL

Signed-off-by: Subhankar Ghosh <[email protected]>

* skip test

Signed-off-by: ericharper <[email protected]>

* use tacotron2 until new fastpitch model is on ngc

Signed-off-by: ericharper <[email protected]>

* use tacotron2 until new fastpitch model is on ngc

Signed-off-by: ericharper <[email protected]>

Co-authored-by: ericharper <[email protected]>

v1.8.0

Toggle v1.8.0's commit message
update package info

Signed-off-by: ericharper <[email protected]>

v1.7.2

Toggle v1.7.2's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
yuya/megatron t5 xnli eval (NVIDIA#3848)

* Add eval only script for xnli

Signed-off-by: Yu Yao <[email protected]>

* Remove unused imports

Signed-off-by: Yu Yao <[email protected]>

* Revert "Remove unused imports"

This reverts commit ee6c556.

* Remove unused imports

Signed-off-by: Yu Yao <[email protected]>

Co-authored-by: Yu Yao <[email protected]>

v1.7.1

Toggle v1.7.1's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
Raise error if FP16 training is tried with O2 recipe. (NVIDIA#3806)

* raise error

Signed-off-by: ericharper <[email protected]>

* update assert

Signed-off-by: ericharper <[email protected]>

* update error message

Signed-off-by: ericharper <[email protected]>

* update error message

Signed-off-by: ericharper <[email protected]>

v1.7.0

Toggle v1.7.0's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
Prompt tuning bug fix (NVIDIA#3780)

* Making updated code backwards compatible with previous prompt tuned models

Signed-off-by: Virginia Adams <[email protected]>

* Fixed backward compatiablity bug

Signed-off-by: Virginia Adams <[email protected]>

* Removed random import

Signed-off-by: Virginia Adams <[email protected]>

Co-authored-by: Eric Harper <[email protected]>

v1.6.2

Toggle v1.6.2's commit message
Change Apex error for warning in NLP collection

Signed-off-by: Oleksii Kuchaiev <[email protected]>

v1.6.1

Toggle v1.6.1's commit message
Merge branch 'r1.6.1' of github.com:NVIDIA/NeMo into r1.6.1

v1.6.0

Toggle v1.6.0's commit message
Update changelog to add authors

Signed-off-by: smajumdar <[email protected]>

v1.5.1

Toggle v1.5.1's commit message
v1.5.1 Release