Reinvent 2024 early#4946

Merged

pintaoz-aws merged 79 commits intomasterfrom

reinvent-2024-early

Dec 4, 2024

Commits on Dec 4, 2024

Base model trainer (#1521)

benieric
authored and
pintaoz-aws
committed
feature: support script mode with local train.sh (#1523)

benieric
authored and
pintaoz-aws
committed
Image Spec refactoring and updates (#1525)

nargokul
authored and
pintaoz-aws
committed
Add unit tests for ModelTrainer (#1527)

benieric
authored and
pintaoz-aws
committed
Add example notebook (#1528)

benieric
authored and
pintaoz-aws
committed
Add enviornment variable bootstrapping script (#1530)

benieric
authored and
pintaoz-aws
committed
feature: add utility function to capture local snapshot (#1524)
pintaoz-aws
committed
Support intelligent parameters (#1540)
pintaoz-aws
committed
Revert Image Spec (#1541)

nargokul
authored and
pintaoz-aws
committed
Cleanup ModelTrainer (#1542)

benieric
authored and
pintaoz-aws
committed
General image builder (#1546)
pintaoz-aws
committed
Latest Container Image (#1545)

nargokul
authored and
pintaoz-aws
committed
Cleanup ModelTrainer code (#1552)

benieric
authored and
pintaoz-aws
committed
feat: add pre-processing and post-processing logic to inference_spec (#1560)

pravali96
authored and
pintaoz-aws
committed
Add Distributed Training Support Model Trainer (#1536)

benieric
authored and
pintaoz-aws
committed
Add path to set Additional Settings in ModelTrainer (#1555)

benieric
authored and
pintaoz-aws
committed
Mask Sensitive Env Logs in Container (#1568)

benieric
authored and
pintaoz-aws
committed
Fix bug in script mode setup ModelTrainer (#1575)

benieric
authored and
pintaoz-aws
committed
Feature: ModelBuilder supports HuggingFace Models with benchmark data and deployment configs (#1572)

xiongz945
authored and
pintaoz-aws
committed
Simplify Config Class Names and DistributedRunner structures (#1573)

benieric
authored and
pintaoz-aws
committed
Remove ignored files

benieric
authored and
pintaoz-aws
committed
Add in_process mode support for DJL and TorchServe servers (#1570)

pravali96
authored and
pintaoz-aws
committed
Pass hyperparameters as CLI args (#1577)

benieric
authored and
pintaoz-aws
committed
Trainer handshake (#1535)

committed
Add Support for Training Recipes (#1565)

authored and
pintaoz-aws
committed
Support building image from Dockerfile (#1571)

committed
Use exact python path in trainer template (#1584)

benieric
authored and
pintaoz-aws
committed
Unified Deployment interface in Model Builder (#1549)

committed
Add recipes examples (#1582)

benieric
authored and
pintaoz-aws
committed
update notebooks (#1588)

benieric
authored and
pintaoz-aws
committed
update notebooks (#1592)

benieric
authored and
pintaoz-aws
committed
Single container local training (#1556)

committed
Bug fixes (#1596)
pintaoz-aws
committed
Update ModelTrainer Notebooks (#1597)

benieric
authored and
pintaoz-aws
committed
add inference morpheus nbs (#1594)

gwang111
authored and
pintaoz-aws
committed
Fix: move the functionality from latest_container_image to retrieve (#1583)

authored and
pintaoz-aws
committed
Add bugbash bootstrapping (#1598)

benieric
authored and
pintaoz-aws
committed
Fix: remove the special condition and fix the unit test (#1601)

authored and
pintaoz-aws
committed
Notebooks update for Bugbash (#1595)

nargokul
authored and
pintaoz-aws
committed
Add Rich Logging to Model Builder (#1604)

nargokul
authored and
pintaoz-aws
committed
Fix: codestyles (#1606)

benieric
authored and
pintaoz-aws
committed
add modelID support to model builder InProcess model (#1580)

pravali96
authored and
pintaoz-aws
committed
Update kandinsky in ModelTrainer and allow setting requirements (#1587)

benieric
authored and
pintaoz-aws
committed
[Updated] Add telemetry to ModelTrainer, Estimator and ModelBuilder (#1608)

zhaoqizqwang
authored and
pintaoz-aws
committed
Integration tests for Model Builder Handshake (#1610)

nargokul
authored and
pintaoz-aws
committed
Use sagemaker core Session (#1607)

authored and
pintaoz-aws
committed
Skip JS model mapping with env vars or image URI provided (#1599)

committed
pin xgboost dlc to 1.7.1 to fix test (#1616)

gwang111
authored and
pintaoz-aws
committed
Revert image builder (#1614)

pintaoz-aws
and
benieric
committed
add integ test for base_model_builder_deploy and remove print statement (#1612)

authored and
pintaoz-aws
committed
Fix tests and codestyle (#1619)

authored and
pintaoz-aws
committed
Fix: Correctly serialize SM_HPS env var (#1611)

benieric
authored and
pintaoz-aws
committed
Intelligent defaults for Model Trainer (#1586)

nargokul
authored and
pintaoz-aws
committed
add in-process mode definition to docs (#1622)

authored and
pintaoz-aws
committed
Update ModelTrainer Interface Parameters (#1617)

benieric
authored and
pintaoz-aws
committed
Model Trainer Bucket improvements (#1618)

nargokul
authored and
pintaoz-aws
committed
Add interface units for ModelTrainer (#1631)

benieric
authored and
pintaoz-aws
committed
Update hyperpod recipe uris (#1629)

benieric
authored and
pintaoz-aws
committed
Integ tests for local mode model trainer (#1623)
pintaoz-aws
committed
Morpheus tests (#1633)

nargokul
authored and
pintaoz-aws
committed
remove example notebooks artifacts (#1634)

benieric
authored and
pintaoz-aws
committed
feat: Partner App Auth Provider for SDK support (#1548)

authored and
pintaoz-aws
committed
change: fix the file uploading signature verification error (#1551)

authored and
pintaoz-aws
committed
Feature: Support GPU training recipes with Sagemaker Python SDK (#1516)

schinmayee
authored and
pintaoz-aws
committed
Feature: Support Neuron training recipes. (#1526)

schinmayee
authored and
pintaoz-aws
committed
Feature: Resolve recipes correctly before launching (#1529)

schinmayee
authored and
pintaoz-aws
committed
Feature: Add unit tests for recipes and minor bug fixes. (#1532)

schinmayee
authored and
pintaoz-aws
committed
Feature: Move image uris and git repos for training recipes to json (#1547)

schinmayee
authored and
pintaoz-aws
committed
Update MANIFEST.in so that wheel builds correctly (#1563)

schinmayee
authored and
pintaoz-aws
committed
Remove default values for fields in recipe_overrides and fix recipe path. (#1566)

schinmayee
authored and
pintaoz-aws
committed
Change default source directory to current, add option to specify source dir (#1593)

schinmayee
authored and
pintaoz-aws
committed
Changes for SMP v2.7.0 (#1609)

authored and
pintaoz-aws
committed
Update URIs to public for training recipes (#1621)

schinmayee
authored and
pintaoz-aws
committed
Neuron URIs update (#1626)

schinmayee
authored and
pintaoz-aws
committed
Usage docs for training recipes (#1630)

schinmayee
authored and
pintaoz-aws
committed
Add model trainer documentation (#1639)

benieric
authored and
pintaoz-aws
committed
Enable the Recipe tests marked with @pytest.mark.skip(reason="Hyperpod recipe code unavailable" (#1642)

nargokul
authored and
pintaoz-aws
committed
Add graphne to the doc requirements
pintaoz-aws
committed
Add graphene to doc requirements
pintaoz-aws
committed