Commit 018018b
committed
Merge branch 'release-v0.99'
============================== Release Notes: v0.99 ==============================
Support for new training algorithms:
- Improvements to LTFB infrastructure (including transfer of SGD and Adam hyperparameters)
Support for new network structures:
- Support for Wide ResNets
Support for new layers:
Python front-end:
- Python front-end for generating neural network architectures (lbann namespace):
including layers, objective functions, callbacks, metrics, and optimizers.
- Python interface for launching (SLURM or LSF) jobs on HPC systems
- Support for running LBANN experiments and capturing experimental output
- Network templates for AlexNet, LeNet, arbitrary ResNet models, and Wide ResNet models
- Python scripts for LeNet, AlexNet, and (Wide) ResNets in model zoo.
Performance optimizations:
- GPU implementation of RMSprop optimizer.
- cuDNN convolution algorithms are determined by empirically measuring
performance rather than using heuristics.
- Avoid setting up unused bias weights.
- Perform gradient accumulations in-place when possible.
Model portability & usability:
Internal features:
- Weight gradient allreduces are in-place rather than on a staging buffer.
- Fully connected and convolution layers only create bias weights when
needed.
- Optimizer exposes gradient buffers so they can be updated in-place.
- Added callback support to explicitly save model
- Min-max metric for reporting on multiple LTFB trainers
- Cleanup of Hydrogen interface to match Hydrogen v1.2.0
- Added type-erased matrix class for internal refactoring
- Make CUB always log performance critical events
I/O & data readers:
- Python data reader that interacts with an embedded Python session.
- Optimized data store to provide preload option
- Extended data store to operate with Cosmoflow-numpy data reader
Build system:
- Added documentation for how users can use Spack to install LBANN
either directly or via environments.
- Conduit is a required dependency.
- Provided Spack environment for installing LBANN as a user
- Improved documentation on lbann.readthedocs.io
- CMake installs a module file in the installation directory that
sets up PATH and PYTHONPATH variables appropriately
Bug fixes:
- Models can now be copied or setup multiple times.
- Fixed incorrect weight initialization with multiple trainers.
- Updated I/O random number generators to be C++ thread safe (rather than OpenMP)
- Added an I/O random number generator for preprocessing that is independent
of the data sequence RNG.
- Fixed initialization order of RNGs and multiple models / trainers.
- General fixes for I/O and LTFB interaction.
Retired features:
- "Zero" layer (hack for early GAN implementation).
- Removed data reader specific implementations of data store (in favor of Conduit-based
data store)File tree
850 files changed
+25225
-18065
lines changed- bamboo
- common_python
- compiler_tests
- integration_tests
- expected_values
- catalyst
- clang4
- gcc4
- gcc7
- intel18
- pascal/gcc7
- ray/gcc4
- surface/gcc4
- unit_tests
- prototext
- cmake
- configure_files
- modules
- containers
- docs
- experiments
- external
- TBinf
- cub
- include/lbann
- callbacks
- data_readers
- patchworks
- data_store
- io
- data_buffers
- layers
- activations
- image
- io
- input
- learning
- loss
- math
- misc
- regularizers
- transform
- metrics
- models
- objective_functions
- weight_regularization
- optimizers
- proto
- utils
- impl
- threads
- weights
- model_zoo
- data_readers
- jag_utils
- models
- alexnet
- autoencoder_candle_pilot1
- autoencoder_cifar10
- autoencoder_imagenet
- autoencoder_mnist
- candle/pilot1
- cosmoflow
- gan
- jags/cycle_gan
- mnist
- jag
- ae_cycle_gan
- 3models
- cycle_gan
- gan
- cyclic
- vanilla
- wae_cycle_gan
- lenet_mnist
- molecular_autoencoder_candle_pilot2
- python/keras
- resnet50
- siamese
- finetune-cub
- triplet
- simple_mnist
- vram
- optimizers
- tests
- layer_tests
- vision
- python
- docs
- onnx
- viz/l2o
- alexnet
- mnist
- resnet50
- plot
- examples
- onnx
- plot
- data
- lbann
- contrib
- lc
- models
- launcher
- models
- onnx
- l2o
- layers
- o2l
- layers
- tests
- plot
- util
- scripts
- onnx
- plot
- scripts
- proto
- lbann
- contrib
- models
- spack_environments
- users/llnl_lc
- ppc64le_cuda
- x86_64_cuda
- src
- callbacks
- data_readers
- patchworks
- data_store
- io
- data_buffers
- layers
- activations
- image
- learning
- loss
- math
- misc
- regularizers
- transform
- metrics
- models
- objective_functions
- weight_regularization
- optimizers
- proto
- factories
- utils
- threads
- unit_test
- weights
- superbuild
- conduit
- cub
- lbann
- tests
- test_stack_tracing
- tools
- compute_mean
- lbann
- utils
- mnist
- unit_test
- viz
- examples
Some content is hidden
Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.
850 files changed
+25225
-18065
lines changed| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
| 1 | + | |
| 2 | + | |
| 3 | + | |
| 4 | + | |
| 5 | + | |
| 6 | + | |
| 7 | + | |
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
1 | | - | |
| 1 | + | |
2 | 2 | | |
3 | 3 | | |
4 | 4 | | |
| |||
48 | 48 | | |
49 | 49 | | |
50 | 50 | | |
51 | | - | |
52 | | - | |
| 51 | + | |
| 52 | + | |
53 | 53 | | |
54 | 54 | | |
55 | 55 | | |
| |||
100 | 100 | | |
101 | 101 | | |
102 | 102 | | |
103 | | - | |
| 103 | + | |
104 | 104 | | |
105 | 105 | | |
106 | 106 | | |
| |||
110 | 110 | | |
111 | 111 | | |
112 | 112 | | |
113 | | - | |
| 113 | + | |
| 114 | + | |
114 | 115 | | |
| 116 | + | |
115 | 117 | | |
116 | 118 | | |
117 | 119 | | |
118 | 120 | | |
| 121 | + | |
| 122 | + | |
| 123 | + | |
119 | 124 | | |
120 | 125 | | |
121 | 126 | | |
| |||
167 | 172 | | |
168 | 173 | | |
169 | 174 | | |
170 | | - | |
| 175 | + | |
171 | 176 | | |
172 | 177 | | |
173 | 178 | | |
174 | 179 | | |
175 | | - | |
| 180 | + | |
176 | 181 | | |
177 | 182 | | |
178 | 183 | | |
| |||
209 | 214 | | |
210 | 215 | | |
211 | 216 | | |
212 | | - | |
| 217 | + | |
213 | 218 | | |
214 | 219 | | |
215 | 220 | | |
216 | 221 | | |
217 | 222 | | |
218 | | - | |
| 223 | + | |
219 | 224 | | |
220 | 225 | | |
221 | 226 | | |
| |||
287 | 292 | | |
288 | 293 | | |
289 | 294 | | |
| 295 | + | |
| 296 | + | |
| 297 | + | |
| 298 | + | |
| 299 | + | |
| 300 | + | |
| 301 | + | |
| 302 | + | |
| 303 | + | |
| 304 | + | |
| 305 | + | |
| 306 | + | |
| 307 | + | |
| 308 | + | |
| 309 | + | |
| 310 | + | |
| 311 | + | |
| 312 | + | |
| 313 | + | |
| 314 | + | |
| 315 | + | |
| 316 | + | |
| 317 | + | |
290 | 318 | | |
291 | 319 | | |
292 | 320 | | |
| |||
305 | 333 | | |
306 | 334 | | |
307 | 335 | | |
308 | | - | |
| 336 | + | |
309 | 337 | | |
310 | 338 | | |
311 | 339 | | |
| |||
336 | 364 | | |
337 | 365 | | |
338 | 366 | | |
339 | | - | |
340 | | - | |
| 367 | + | |
| 368 | + | |
341 | 369 | | |
342 | 370 | | |
343 | | - | |
344 | | - | |
| 371 | + | |
| 372 | + | |
345 | 373 | | |
346 | 374 | | |
347 | | - | |
| 375 | + | |
348 | 376 | | |
349 | 377 | | |
350 | 378 | | |
| |||
402 | 430 | | |
403 | 431 | | |
404 | 432 | | |
405 | | - | |
| 433 | + | |
406 | 434 | | |
407 | 435 | | |
| 436 | + | |
| 437 | + | |
| 438 | + | |
| 439 | + | |
| 440 | + | |
| 441 | + | |
| 442 | + | |
| 443 | + | |
| 444 | + | |
| 445 | + | |
| 446 | + | |
| 447 | + | |
| 448 | + | |
| 449 | + | |
| 450 | + | |
| 451 | + | |
| 452 | + | |
| 453 | + | |
| 454 | + | |
408 | 455 | | |
409 | 456 | | |
410 | 457 | | |
| |||
430 | 477 | | |
431 | 478 | | |
432 | 479 | | |
| 480 | + | |
| 481 | + | |
| 482 | + | |
| 483 | + | |
433 | 484 | | |
434 | 485 | | |
435 | 486 | | |
| |||
460 | 511 | | |
461 | 512 | | |
462 | 513 | | |
| 514 | + | |
| 515 | + | |
| 516 | + | |
| 517 | + | |
463 | 518 | | |
464 | 519 | | |
465 | 520 | | |
| |||
516 | 571 | | |
517 | 572 | | |
518 | 573 | | |
519 | | - | |
520 | | - | |
| 574 | + | |
| 575 | + | |
521 | 576 | | |
522 | 577 | | |
523 | 578 | | |
| |||
559 | 614 | | |
560 | 615 | | |
561 | 616 | | |
| 617 | + | |
| 618 | + | |
| 619 | + | |
| 620 | + | |
| 621 | + | |
| 622 | + | |
| 623 | + | |
| 624 | + | |
| 625 | + | |
| 626 | + | |
| 627 | + | |
| 628 | + | |
| 629 | + | |
| 630 | + | |
| 631 | + | |
| 632 | + | |
| 633 | + | |
| 634 | + | |
| 635 | + | |
| 636 | + | |
| 637 | + | |
| 638 | + | |
| 639 | + | |
| 640 | + | |
| 641 | + | |
| 642 | + | |
| 643 | + | |
| 644 | + | |
| 645 | + | |
| 646 | + | |
| 647 | + | |
| 648 | + | |
| 649 | + | |
| 650 | + | |
| 651 | + | |
| 652 | + | |
| 653 | + | |
| 654 | + | |
| 655 | + | |
| 656 | + | |
| 657 | + | |
| 658 | + | |
| 659 | + | |
| 660 | + | |
| 661 | + | |
| 662 | + | |
| 663 | + | |
| 664 | + | |
| 665 | + | |
| 666 | + | |
| 667 | + | |
| 668 | + | |
| 669 | + | |
| 670 | + | |
| 671 | + | |
| 672 | + | |
| 673 | + | |
| 674 | + | |
562 | 675 | | |
563 | 676 | | |
564 | 677 | | |
| |||
583 | 696 | | |
584 | 697 | | |
585 | 698 | | |
| 699 | + | |
586 | 700 | | |
587 | 701 | | |
| 702 | + | |
588 | 703 | | |
589 | 704 | | |
590 | 705 | | |
| |||
632 | 747 | | |
633 | 748 | | |
634 | 749 | | |
635 | | - | |
| 750 | + | |
| 751 | + | |
636 | 752 | | |
637 | 753 | | |
638 | 754 | | |
639 | 755 | | |
640 | 756 | | |
641 | 757 | | |
| 758 | + | |
| 759 | + | |
| 760 | + | |
| 761 | + | |
| 762 | + | |
| 763 | + | |
| 764 | + | |
| 765 | + | |
| 766 | + | |
| 767 | + | |
| 768 | + | |
| 769 | + | |
| 770 | + | |
| 771 | + | |
| 772 | + | |
| 773 | + | |
| 774 | + | |
| 775 | + | |
| 776 | + | |
| 777 | + | |
| 778 | + | |
| 779 | + | |
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
1 | | - | |
2 | | - | |
| 1 | + | |
| 2 | + | |
3 | 3 | | |
4 | 4 | | |
5 | 5 | | |
| |||
8 | 8 | | |
9 | 9 | | |
10 | 10 | | |
11 | | - | |
| 11 | + | |
12 | 12 | | |
13 | 13 | | |
14 | 14 | | |
| |||
21 | 21 | | |
22 | 22 | | |
23 | 23 | | |
24 | | - | |
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
21 | 21 | | |
22 | 22 | | |
23 | 23 | | |
24 | | - | |
25 | | - | |
| 24 | + | |
| 25 | + | |
| 26 | + | |
26 | 27 | | |
| 28 | + | |
| 29 | + | |
| 30 | + | |
| 31 | + | |
| 32 | + | |
| 33 | + | |
| 34 | + | |
27 | 35 | | |
28 | 36 | | |
29 | 37 | | |
| |||
42 | 50 | | |
43 | 51 | | |
44 | 52 | | |
45 | | - | |
| 53 | + | |
| 54 | + | |
| 55 | + | |
46 | 56 | | |
| 57 | + | |
| 58 | + | |
47 | 59 | | |
48 | 60 | | |
49 | 61 | | |
| |||
0 commit comments