llnl/lbann issues and pull requests

#2479 - error: floating point classification requires argument of floating point type (passed in 'half_float::half')

Issue - State: open - Opened by yurivict about 1 month ago

#2478 - comm_nb_request.hpp:80:3: error: unknown type name 'MPI_Request'

Issue - State: open - Opened by yurivict about 1 month ago

#2477 - Further IO improvements for CosmoFlow

Pull Request - State: closed - Opened by fiedorowicz1 4 months ago

#2476 - Cosmoflow IO performance optimizations

Pull Request - State: closed - Opened by fiedorowicz1 4 months ago

#2475 - Integrated Stephen Fleischmans changes to the benchmarking branch. Ev…

Pull Request - State: closed - Opened by sfleisch 5 months ago

#2474 - Add new LC systems

Pull Request - State: closed - Opened by fiedorowicz1 5 months ago

#2473 - added template for data reader to pass conduit node from driver

Pull Request - State: open - Opened by bvanessen 5 months ago

#2472 - fix broken mnist link (fixes #2466)

Pull Request - State: open - Opened by jvwilliams23 5 months ago

#2471 - Fix remaining CI errors

Pull Request - State: closed - Opened by bvanessen 6 months ago

#2470 - Fix a mkdir -p issue

Pull Request - State: closed - Opened by benson31 6 months ago
Labels: bug, review requested

#2469 - Remove some logic from the superbuild

Pull Request - State: closed - Opened by benson31 6 months ago
Labels: review requested, build, CI

#2468 - Support escaped file names with catch v3 multiple reporters

Pull Request - State: closed - Opened by benson31 7 months ago
Labels: review requested, CI

#2467 - Fix Adam

Pull Request - State: closed - Opened by fiedorowicz1 7 months ago

#2466 - Expired link in the readthedocs

Issue - State: open - Opened by kf-cuanschutz 7 months ago

#2465 - Skip running Tessellate backpropagation if the matrix is empty

Pull Request - State: closed - Opened by tbennun 7 months ago - 2 comments

#2464 - fix missing header for dump_model_graph

Pull Request - State: closed - Opened by jvwilliams23 7 months ago

#2463 - cuTENSOR error (CUTENSOR_STATUS_NOT_SUPPORTED)

Issue - State: closed - Opened by jvwilliams23 7 months ago - 1 comment

#2462 - Superbuild example script updates

Pull Request - State: open - Opened by tbennun 8 months ago

#2461 - More CosmoFlow Fixes

Pull Request - State: closed - Opened by fiedorowicz1 8 months ago

#2460 - Use CUTENSOR for Tessellate backpropagation in supported cases

Pull Request - State: closed - Opened by tbennun 8 months ago

#2459 - Fix a couple issues with the auto-detection of the NVCC_GENCODE in NCCL

Pull Request - State: closed - Opened by benson31 8 months ago
Labels: bug, build

#2458 - Memory profiler: add support for distconv-enabled layers

Pull Request - State: closed - Opened by tbennun 8 months ago - 1 comment

#2457 - Fix Hang in Python Dataset Reader with DistConv

Pull Request - State: closed - Opened by fiedorowicz1 8 months ago

#2456 - Added a post install message to indicate how to load the installed module

Pull Request - State: closed - Opened by bvanessen 8 months ago

#2455 - Update Spack revision

Pull Request - State: closed - Opened by tbennun 8 months ago

#2454 - Add CMAKE_CUDA_ARCHITECTURES support for NCCL's gencode

Pull Request - State: closed - Opened by benson31 8 months ago
Labels: build

#2453 - Fix AppleClang warnings

Pull Request - State: closed - Opened by benson31 8 months ago

#2452 - Quick NVSHMEM tweaks

Pull Request - State: closed - Opened by benson31 9 months ago

#2451 - Cleanup old submodules

Pull Request - State: closed - Opened by bvanessen 9 months ago

#2450 - Fix a compilation error when compiling with half and cuTENSOR

Pull Request - State: closed - Opened by benson31 9 months ago
Labels: bug, build

#2449 - CosmoFlow Updates

Pull Request - State: closed - Opened by fiedorowicz1 9 months ago

#2448 - Add hwloc to the SuperBuild

Pull Request - State: closed - Opened by benson31 9 months ago

#2446 - Fix import issue in Periodic Padding test

Pull Request - State: closed - Opened by tbennun 10 months ago

#2445 - 'flux mini' -> 'flux'

Pull Request - State: closed - Opened by benson31 10 months ago
Labels: python, CI

#2444 - Accept any H2 version

Pull Request - State: closed - Opened by benson31 10 months ago
Labels: build

#2443 - Fix existing-source builds of Clara

Pull Request - State: closed - Opened by benson31 10 months ago
Labels: bug, build

#2442 - Fixes In-Place Activation Reference Counting

Pull Request - State: closed - Opened by fiedorowicz1 10 months ago

#2441 - Add NCCL to the superbuild

Pull Request - State: closed - Opened by benson31 10 months ago
Labels: enhancement, build

#2440 - Add aws-ofi-rccl plugin to the superbuild

Pull Request - State: closed - Opened by benson31 10 months ago
Labels: enhancement, build

#2439 - Make dataset reader faster

Pull Request - State: closed - Opened by tbennun 10 months ago

#2436 - Use Hydrogen for MPI Calls

Pull Request - State: closed - Opened by fiedorowicz1 10 months ago

#2435 - Fix to PeriodicPadding to match pytorch circular padding

Pull Request - State: closed - Opened by jvwilliams23 10 months ago

#2434 - Masked language modeling example

Pull Request - State: closed - Opened by tbennun 11 months ago

#2433 - Potential bug in lbann.Scale

Issue - State: closed - Opened by jvwilliams23 12 months ago - 8 comments
Labels: bug

#2432 - Adding HDF5 compression plugins to superbuild

Pull Request - State: open - Opened by bvanessen 12 months ago

#2430 - Implement multi-dimensional reduction and refactor cuTENSOR support

Pull Request - State: closed - Opened by tbennun 12 months ago - 1 comment

#2429 - Weight demodulation

Issue - State: closed - Opened by jvwilliams23 almost 1 year ago - 9 comments

#2428 - Enable Asymmetric DistConv Kernels

Pull Request - State: closed - Opened by fiedorowicz1 about 1 year ago

#2427 - Minor improvements to the documentation

Pull Request - State: closed - Opened by nmnobre about 1 year ago

#2426 - Extensible configuration of the superbuild location

Pull Request - State: closed - Opened by bvanessen about 1 year ago

#2425 - Fix CosmoFlow Double-Reading

Pull Request - State: closed - Opened by fiedorowicz1 about 1 year ago

#2424 - FSDP: Enable limiting scope using trainer grid rows/columns

Pull Request - State: closed - Opened by tbennun about 1 year ago - 1 comment

#2423 - Do not backpropagate through layers without gradient sources

Pull Request - State: closed - Opened by tbennun about 1 year ago - 3 comments

#2422 - Fix typo in example superbuild script

Pull Request - State: closed - Opened by benson31 about 1 year ago

#2421 - Restore the superbuild

Pull Request - State: closed - Opened by benson31 about 1 year ago
Labels: build

#2420 - Support layer parallelism in transformer application

Pull Request - State: closed - Opened by tbennun about 1 year ago

#2419 - Fix model-parallel layer normalization

Pull Request - State: closed - Opened by tbennun about 1 year ago - 1 comment

#2418 - Add GitHub Actions action that builds LBANN

Pull Request - State: closed - Opened by tbennun about 1 year ago

#2417 - Fix build error in Layer Normalization

Pull Request - State: closed - Opened by tbennun about 1 year ago

#2416 - Removed the specific version of the PrgEnv module used on Cray systems.

Pull Request - State: closed - Opened by bvanessen about 1 year ago

#2415 - Fix compiler warning

Pull Request - State: closed - Opened by benson31 about 1 year ago
Labels: review requested, CI

#2414 - Python Dataset Reader

Pull Request - State: closed - Opened by fiedorowicz1 about 1 year ago

#2413 - Configurable scientific notation in progress bar

Pull Request - State: closed - Opened by tbennun about 1 year ago

#2412 - External metrics

Pull Request - State: closed - Opened by tbennun about 1 year ago - 2 comments

#2411 - Transformer extensions and refactoring

Pull Request - State: closed - Opened by tbennun about 1 year ago - 1 comment

#2410 - Fix TSE LTFB test when top-k generates multiple sends

Pull Request - State: closed - Opened by tbennun about 1 year ago

#2409 - Updated libraries and externals to build on Perlmutter

Pull Request - State: open - Opened by szaman19 about 1 year ago

#2408 - Enable reconfiguring progress bar parameters and random seed for transformers

Pull Request - State: closed - Opened by tbennun about 1 year ago

#2407 - Multiple build errors: error: static_cast from 'const lbann::l2_weight_regularization ' to 'const lbann::objective_function_term ', which are not related by inheritance, is not allowed, etc

Issue - State: closed - Opened by yurivict about 1 year ago - 5 comments

#2406 - error: invalid operands to binary expression ('const lbann::callback::(anonymous namespace)::MemUsage' and 'const lbann::callback::(anonymous namespace)::MemUsage')

Issue - State: closed - Opened by yurivict about 1 year ago - 5 comments
Labels: bug

#2405 - Cleanup data ingestion types

Pull Request - State: closed - Opened by bvanessen about 1 year ago
Labels: bug

#2404 - Bugfix cleanup rng omp

Pull Request - State: open - Opened by bvanessen about 1 year ago

#2403 - Fix pooling workspace behavior on MIOpen

Pull Request - State: closed - Opened by tbennun about 1 year ago - 1 comment

#2401 - Force convolutions to be deterministic when LBANN_DETERMINISTIC

Pull Request - State: closed - Opened by benson31 about 1 year ago
Labels: enhancement, review requested

#2400 - Enabled the CosmoFlow integration test to run in the distconv CI

Pull Request - State: closed - Opened by bvanessen about 1 year ago

#2399 - Simplified Molecular transformer example

Pull Request - State: open - Opened by szaman19 about 1 year ago - 2 comments

#2398 - Bugfix flux hangs on Corona and Lassen CI jobs.

Pull Request - State: closed - Opened by bvanessen about 1 year ago

#2397 - Python updates/cleanup for benchmarking

Pull Request - State: closed - Opened by ndryden about 1 year ago
Labels: enhancement, python

#2396 - Fixes for half and CUDA 12.2

Pull Request - State: closed - Opened by ndryden about 1 year ago
Labels: bug

#2395 - Fix gradient clipping with half

Pull Request - State: closed - Opened by ndryden about 1 year ago
Labels: bug

#2393 - Update summary.cpp

Pull Request - State: closed - Opened by benson31 about 1 year ago
Labels: bug, build

#2390 - Add memory pool report to memory profiler

Pull Request - State: closed - Opened by tbennun over 1 year ago

#2387 - Spack development branch build issue

Issue - State: closed - Opened by jvwilliams23 over 1 year ago - 1 comment
Labels: bug, build

#2382 - Remove unused drivers

Pull Request - State: closed - Opened by benson31 over 1 year ago - 1 comment
Labels: review requested

#2381 - Feature support new spack

Pull Request - State: closed - Opened by bvanessen over 1 year ago

#2377 - Refactor async io to use uint64_t

Pull Request - State: closed - Opened by bvanessen over 1 year ago

#2375 - Update CosmoGAN MultiGAN Spectral Loss

Pull Request - State: closed - Opened by fiedorowicz1 over 1 year ago

#2374 - Clean up gradient checker

Pull Request - State: closed - Opened by tbennun over 1 year ago

#2369 - test_unit_algo_ltfb_trunc_selection.py - fails intermittently

Issue - State: closed - Opened by bvanessen over 1 year ago - 1 comment

#2364 - Node Property Prediction Models on LBANN

Pull Request - State: open - Opened by szaman19 over 1 year ago

#2361 - LBANN Implementation of DPMNN on CSD-10K Data

Pull Request - State: open - Opened by szaman19 over 1 year ago

#2356 - Use DiHydrogen for memory profiler information

Pull Request - State: open - Opened by tbennun over 1 year ago
Labels: do_not_merge

#2355 - Double data type optional

Pull Request - State: closed - Opened by bvanessen over 1 year ago - 1 comment

#2342 - Initial layer parallelism

Pull Request - State: closed - Opened by benson31 over 1 year ago

#2328 - HDF5 Data reader enhancements

Pull Request - State: open - Opened by bvanessen over 1 year ago

#2314 - Add dace kernels for cosmoflow

Pull Request - State: closed - Opened by bvanessen over 1 year ago

#2255 - Add Upsample Layer

Pull Request - State: closed - Opened by fiedorowicz1 almost 2 years ago
Labels: enhancement, in progress

#2248 - Add support for changing distconv overlaps in Identity layer.

Pull Request - State: closed - Opened by fiedorowicz1 almost 2 years ago - 1 comment
Labels: enhancement

#2229 - Adds Distconv support for Layer Normalization layer

Pull Request - State: open - Opened by szaman19 almost 2 years ago - 2 comments

#2194 - Add Alternating Generator/Discriminator Updates to ExaGAN

Pull Request - State: closed - Opened by fiedorowicz1 about 2 years ago

GitHub / llnl/lbann issues and pull requests