.coderabbit.yaml
.dockerignore
.gitattributes
.gitignore
.pre-commit-config.yaml
CHANGELOG.md
CODE_OF_CONDUCT.md
CONTRIBUTING.md
Dockerfile
LICENSE
README.md
SECURITY.md
mkdocs.yml
pyproject.toml
.devcontainer/devcontainer.json
.git-hooks/block-workflow-artifacts.sh
.github/PULL_REQUEST_TEMPLATE.md
.github/ISSUE_TEMPLATE/bug_report.yml
.github/ISSUE_TEMPLATE/config.yml
.github/ISSUE_TEMPLATE/feature_request.yml
.github/ISSUE_TEMPLATE/i_found_a_bottleneck.yml
.github/workflows/ci.yml
.github/workflows/codeql.yml
.github/workflows/docs.yml
.github/workflows/publish.yml
.github/workflows/release.yml
docs/index.md
docs/assets/Architecture_excali_b.png
docs/assets/cli_demo.gif
docs/assets/dashboard_live.gif
docs/assets/end-of-run-summary.png
docs/assets/local_ui.png
docs/developer_guide/architecture.md
docs/developer_guide/contributing.md
docs/developer_guide/extending.md
docs/developer_guide/rank-straggler-policy.md
docs/guides/ddp-slow-training-rank-straggler.md
docs/guides/low-gpu-utilization-pytorch.md
docs/guides/pytorch-input-pipeline-bottleneck.md
docs/guides/pytorch-memory-creep.md
docs/guides/slow-pytorch-training.md
docs/stylesheets/traceopt.css
docs/user_guide/compare.md
docs/user_guide/distributed-training.md
docs/user_guide/faq.md
docs/user_guide/integrations.md
docs/user_guide/public-api.md
docs/user_guide/quickstart.md
docs/user_guide/reading-output.md
docs/user_guide/slurm.md
docs/user_guide/integrations/deepspeed.md
docs/user_guide/integrations/huggingface.md
docs/user_guide/integrations/lightning.md
docs/user_guide/integrations/ray.md
docs/user_guide/integrations/wandb-mlflow.md
examples/README.md
examples/manual_custom_minimal.py
examples/quickstart.py
examples/summary_logging_minimal.py
examples/advanced/bert_gradient_accum.py
examples/advanced/bert_single_gpu_compare.py
examples/advanced/fsdp_minimal_cuda.py
examples/advanced/huggingface_vision_vit.py
examples/diagnosis/dataloader_bottleneck_demo.py
examples/diagnosis/h2d_timing_demo.py
examples/distributed/ddp_minimal.py
examples/distributed/ddp_rank_straggler_demo.py
examples/integrations/deepspeed_config_minimal.json
examples/integrations/deepspeed_minimal.py
examples/integrations/huggingface_trainer_minimal.py
examples/integrations/lightning_minimal.py
examples/ray/lightning_text_classifier.py
examples/ray/torchtrainer_minimal.py
examples/slurm/README.md
examples/slurm/launch.sh
examples/slurm/traceml_ddp.sbatch
notebooks/README.md
notebooks/data_loading_bottleneck.ipynb
notebooks/huggingface_dataloading_bottleneck.ipynb
src/dev/demo/mlp_ddp_compute_straggler.py
src/dev/demo/mlp_ddp_healthy_baseline.py
src/dev/demo/mlp_ddp_input_bound.py
src/dev/demo/mlp_ddp_input_straggler.py
src/dev/demo/mlp_ddp_memory_creep.py
src/dev/scenarios/bert_ddp.py
src/dev/scenarios/bert_memory_creep.py
src/dev/scenarios/bert_straggler.py
src/dev/scenarios/vit_ddp_stress.py
src/traceml/__init__.py
src/traceml_ai/__init__.py
src/traceml_ai/_version.py
src/traceml_ai/api.py
src/traceml_ai.egg-info/PKG-INFO
src/traceml_ai.egg-info/SOURCES.txt
src/traceml_ai.egg-info/dependency_links.txt
src/traceml_ai.egg-info/entry_points.txt
src/traceml_ai.egg-info/requires.txt
src/traceml_ai.egg-info/scm_file_list.json
src/traceml_ai.egg-info/scm_version.json
src/traceml_ai.egg-info/top_level.txt
src/traceml_ai/aggregator/__init__.py
src/traceml_ai/aggregator/aggregator_main.py
src/traceml_ai/aggregator/sqlite_writer.py
src/traceml_ai/aggregator/summary_service.py
src/traceml_ai/aggregator/trace_aggregator.py
src/traceml_ai/aggregator/display_drivers/__init__.py
src/traceml_ai/aggregator/display_drivers/base.py
src/traceml_ai/aggregator/display_drivers/cli.py
src/traceml_ai/aggregator/display_drivers/layout.py
src/traceml_ai/aggregator/display_drivers/nicegui.py
src/traceml_ai/aggregator/display_drivers/server_readiness.py
src/traceml_ai/aggregator/display_drivers/staleness.py
src/traceml_ai/aggregator/display_drivers/summary.py
src/traceml_ai/aggregator/display_drivers/assets/fonts/Geist.woff2
src/traceml_ai/aggregator/display_drivers/assets/fonts/GeistMono.woff2
src/traceml_ai/aggregator/display_drivers/nicegui_sections/__init__.py
src/traceml_ai/aggregator/display_drivers/nicegui_sections/model_combined_section.py
src/traceml_ai/aggregator/display_drivers/nicegui_sections/model_diagnostics_section.py
src/traceml_ai/aggregator/display_drivers/nicegui_sections/pages.py
src/traceml_ai/aggregator/display_drivers/nicegui_sections/process_section.py
src/traceml_ai/aggregator/display_drivers/nicegui_sections/step_memory_section.py
src/traceml_ai/aggregator/display_drivers/nicegui_sections/system_section.py
src/traceml_ai/aggregator/display_drivers/nicegui_sections/theme.py
src/traceml_ai/aggregator/display_drivers/nicegui_sections/ui_shell.py
src/traceml_ai/aggregator/sqlite_writers/__init__.py
src/traceml_ai/aggregator/sqlite_writers/process.py
src/traceml_ai/aggregator/sqlite_writers/runtime_environment.py
src/traceml_ai/aggregator/sqlite_writers/stdout_stderr.py
src/traceml_ai/aggregator/sqlite_writers/step_memory.py
src/traceml_ai/aggregator/sqlite_writers/step_time.py
src/traceml_ai/aggregator/sqlite_writers/system.py
src/traceml_ai/analytics/__init__.py
src/traceml_ai/analytics/trends/__init__.py
src/traceml_ai/analytics/trends/core.py
src/traceml_ai/analytics/trends/schema.py
src/traceml_ai/config/__init__.py
src/traceml_ai/config/yaml_loader.py
src/traceml_ai/core/__init__.py
src/traceml_ai/core/lifecycle.py
src/traceml_ai/core/registry.py
src/traceml_ai/core/rendering.py
src/traceml_ai/core/summaries.py
src/traceml_ai/core/telemetry.py
src/traceml_ai/database/__init__.py
src/traceml_ai/database/database.py
src/traceml_ai/database/database_sender.py
src/traceml_ai/database/database_writer.py
src/traceml_ai/diagnostics/DIAGNOSIS.md
src/traceml_ai/diagnostics/__init__.py
src/traceml_ai/diagnostics/bands.py
src/traceml_ai/diagnostics/common.py
src/traceml_ai/diagnostics/model_diagnostics.py
src/traceml_ai/diagnostics/registry.py
src/traceml_ai/diagnostics/step_memory_formatters.py
src/traceml_ai/diagnostics/trends.py
src/traceml_ai/diagnostics/process/__init__.py
src/traceml_ai/diagnostics/process/api.py
src/traceml_ai/diagnostics/process/context.py
src/traceml_ai/diagnostics/process/policy.py
src/traceml_ai/diagnostics/process/rules.py
src/traceml_ai/diagnostics/step_memory/__init__.py
src/traceml_ai/diagnostics/step_memory/adapters.py
src/traceml_ai/diagnostics/step_memory/api.py
src/traceml_ai/diagnostics/step_memory/context.py
src/traceml_ai/diagnostics/step_memory/policy.py
src/traceml_ai/diagnostics/step_memory/rules.py
src/traceml_ai/diagnostics/step_memory/trend.py
src/traceml_ai/diagnostics/step_time/__init__.py
src/traceml_ai/diagnostics/step_time/adapters.py
src/traceml_ai/diagnostics/step_time/api.py
src/traceml_ai/diagnostics/step_time/context.py
src/traceml_ai/diagnostics/step_time/formatters.py
src/traceml_ai/diagnostics/step_time/policy.py
src/traceml_ai/diagnostics/step_time/rules.py
src/traceml_ai/diagnostics/step_time/trend.py
src/traceml_ai/diagnostics/system/__init__.py
src/traceml_ai/diagnostics/system/api.py
src/traceml_ai/diagnostics/system/context.py
src/traceml_ai/diagnostics/system/policy.py
src/traceml_ai/diagnostics/system/rules.py
src/traceml_ai/instrumentation/__init__.py
src/traceml_ai/instrumentation/h2d.py
src/traceml_ai/instrumentation/hooks/__init__.py
src/traceml_ai/instrumentation/hooks/optimizer_hooks.py
src/traceml_ai/instrumentation/patches/__init__.py
src/traceml_ai/instrumentation/patches/backward_auto_timer_patch.py
src/traceml_ai/instrumentation/patches/dataloader_patch.py
src/traceml_ai/instrumentation/patches/forward_auto_timer_patch.py
src/traceml_ai/instrumentation/patches/h2d_auto_timer_patch.py
src/traceml_ai/integrations/__init__.py
src/traceml_ai/integrations/huggingface.py
src/traceml_ai/integrations/lightning.py
src/traceml_ai/integrations/ray.py
src/traceml_ai/launcher/__init__.py
src/traceml_ai/launcher/cli.py
src/traceml_ai/launcher/commands.py
src/traceml_ai/launcher/launch_config.py
src/traceml_ai/launcher/manifest.py
src/traceml_ai/launcher/process.py
src/traceml_ai/loggers/__init__.py
src/traceml_ai/loggers/error_log.py
src/traceml_ai/renderers/__init__.py
src/traceml_ai/renderers/base_renderer.py
src/traceml_ai/renderers/stdout_stderr_renderer.py
src/traceml_ai/renderers/utils.py
src/traceml_ai/renderers/model_diagnostics/renderer.py
src/traceml_ai/renderers/process/__init__.py
src/traceml_ai/renderers/process/cli_compute.py
src/traceml_ai/renderers/process/common.py
src/traceml_ai/renderers/process/computer.py
src/traceml_ai/renderers/process/dashboard_compute.py
src/traceml_ai/renderers/process/renderer.py
src/traceml_ai/renderers/stdout_stderr/common.py
src/traceml_ai/renderers/step_memory/cli_compute.py
src/traceml_ai/renderers/step_memory/common.py
src/traceml_ai/renderers/step_memory/computer.py
src/traceml_ai/renderers/step_memory/dashboard_compute.py
src/traceml_ai/renderers/step_memory/formatter.py
src/traceml_ai/renderers/step_memory/renderer.py
src/traceml_ai/renderers/step_memory/schema.py
src/traceml_ai/renderers/step_time/__init__.py
src/traceml_ai/renderers/step_time/compute.py
src/traceml_ai/renderers/step_time/renderer.py
src/traceml_ai/renderers/step_time/schema.py
src/traceml_ai/renderers/system/__init__.py
src/traceml_ai/renderers/system/cli_cluster.py
src/traceml_ai/renderers/system/cli_compute.py
src/traceml_ai/renderers/system/common.py
src/traceml_ai/renderers/system/computer.py
src/traceml_ai/renderers/system/dashboard_compute.py
src/traceml_ai/renderers/system/renderer.py
src/traceml_ai/reporting/SCHEMA.md
src/traceml_ai/reporting/__init__.py
src/traceml_ai/reporting/config.py
src/traceml_ai/reporting/final.py
src/traceml_ai/reporting/primary_diagnosis.py
src/traceml_ai/reporting/schema.py
src/traceml_ai/reporting/summary_artifact.py
src/traceml_ai/reporting/topology.py
src/traceml_ai/reporting/compare/__init__.py
src/traceml_ai/reporting/compare/command.py
src/traceml_ai/reporting/compare/core.py
src/traceml_ai/reporting/compare/formatters.py
src/traceml_ai/reporting/compare/io.py
src/traceml_ai/reporting/compare/model.py
src/traceml_ai/reporting/compare/policy.py
src/traceml_ai/reporting/compare/render.py
src/traceml_ai/reporting/compare/verdict.py
src/traceml_ai/reporting/compare/sections/__init__.py
src/traceml_ai/reporting/compare/sections/base.py
src/traceml_ai/reporting/compare/sections/process.py
src/traceml_ai/reporting/compare/sections/step_memory.py
src/traceml_ai/reporting/compare/sections/step_time.py
src/traceml_ai/reporting/compare/sections/system.py
src/traceml_ai/reporting/html/__init__.py
src/traceml_ai/reporting/html/banner.py
src/traceml_ai/reporting/html/document.py
src/traceml_ai/reporting/html/sections.py
src/traceml_ai/reporting/html/sections_helpers.py
src/traceml_ai/reporting/html/style.py
src/traceml_ai/reporting/html/svg.py
src/traceml_ai/reporting/html/textutils.py
src/traceml_ai/reporting/html/writer.py
src/traceml_ai/reporting/sections/__init__.py
src/traceml_ai/reporting/sections/base.py
src/traceml_ai/reporting/sections/output.py
src/traceml_ai/reporting/sections/process/__init__.py
src/traceml_ai/reporting/sections/process/builder.py
src/traceml_ai/reporting/sections/process/formatter.py
src/traceml_ai/reporting/sections/process/loader.py
src/traceml_ai/reporting/sections/process/model.py
src/traceml_ai/reporting/sections/step_memory/__init__.py
src/traceml_ai/reporting/sections/step_memory/builder.py
src/traceml_ai/reporting/sections/step_memory/formatter.py
src/traceml_ai/reporting/sections/step_memory/loader.py
src/traceml_ai/reporting/sections/step_memory/model.py
src/traceml_ai/reporting/sections/step_time/__init__.py
src/traceml_ai/reporting/sections/step_time/builder.py
src/traceml_ai/reporting/sections/step_time/formatter.py
src/traceml_ai/reporting/sections/step_time/loader.py
src/traceml_ai/reporting/sections/step_time/model.py
src/traceml_ai/reporting/sections/system/__init__.py
src/traceml_ai/reporting/sections/system/builder.py
src/traceml_ai/reporting/sections/system/formatter.py
src/traceml_ai/reporting/sections/system/loader.py
src/traceml_ai/reporting/sections/system/model.py
src/traceml_ai/reporting/summaries/__init__.py
src/traceml_ai/reporting/summaries/issue_summary.py
src/traceml_ai/reporting/summaries/process.py
src/traceml_ai/reporting/summaries/step_memory.py
src/traceml_ai/reporting/summaries/step_time.py
src/traceml_ai/reporting/summaries/summary_formatting.py
src/traceml_ai/reporting/summaries/summary_io.py
src/traceml_ai/reporting/summaries/summary_layout.py
src/traceml_ai/reporting/summaries/system.py
src/traceml_ai/reporting/view/__init__.py
src/traceml_ai/reporting/view/command.py
src/traceml_ai/runtime/__init__.py
src/traceml_ai/runtime/arming.py
src/traceml_ai/runtime/config.py
src/traceml_ai/runtime/environment.py
src/traceml_ai/runtime/environment_state.py
src/traceml_ai/runtime/executor.py
src/traceml_ai/runtime/exporter.py
src/traceml_ai/runtime/identity.py
src/traceml_ai/runtime/launch_context.py
src/traceml_ai/runtime/lifecycle.py
src/traceml_ai/runtime/runtime.py
src/traceml_ai/runtime/sampler_registry.py
src/traceml_ai/runtime/sender.py
src/traceml_ai/runtime/session.py
src/traceml_ai/runtime/settings.py
src/traceml_ai/runtime/state.py
src/traceml_ai/runtime/stdout_stderr_capture.py
src/traceml_ai/samplers/__init__.py
src/traceml_ai/samplers/base_sampler.py
src/traceml_ai/samplers/process_sampler.py
src/traceml_ai/samplers/runtime_context.py
src/traceml_ai/samplers/runtime_environment_sampler.py
src/traceml_ai/samplers/stdout_stderr_sampler.py
src/traceml_ai/samplers/step_memory_sampler.py
src/traceml_ai/samplers/step_time_sampler.py
src/traceml_ai/samplers/system_manifest.py
src/traceml_ai/samplers/system_sampler.py
src/traceml_ai/samplers/utils.py
src/traceml_ai/samplers/schema/process.py
src/traceml_ai/samplers/schema/step_memory.py
src/traceml_ai/samplers/schema/step_time_schema.py
src/traceml_ai/samplers/schema/system.py
src/traceml_ai/sdk/__init__.py
src/traceml_ai/sdk/initial.py
src/traceml_ai/sdk/instrumentation.py
src/traceml_ai/sdk/protocol.py
src/traceml_ai/sdk/summary_client.py
src/traceml_ai/sdk/summary_projection.py
src/traceml_ai/sdk/wrappers.py
src/traceml_ai/telemetry/__init__.py
src/traceml_ai/telemetry/control.py
src/traceml_ai/telemetry/envelope.py
src/traceml_ai/transport/__init__.py
src/traceml_ai/transport/distributed.py
src/traceml_ai/transport/tcp_transport.py
src/traceml_ai/utils/__init__.py
src/traceml_ai/utils/atomic_io.py
src/traceml_ai/utils/base_trace_event.py
src/traceml_ai/utils/cuda_event_pool.py
src/traceml_ai/utils/flush_buffers.py
src/traceml_ai/utils/formatting.py
src/traceml_ai/utils/msgpack_codec.py
src/traceml_ai/utils/step_memory.py
src/traceml_ai/utils/step_time_sqlite.py
src/traceml_ai/utils/step_time_window.py
src/traceml_ai/utils/step_windows.py
src/traceml_ai/utils/timing.py
src/traceml_ai/utils/training_strategy.py
src/traceml_ai/utils/ast_analysis/__init__.py
src/traceml_ai/utils/ast_analysis/code_manifest.py
src/traceml_ai/utils/ast_analysis/helpers.py
src/traceml_ai/utils/ast_analysis/models.py
src/traceml_ai/utils/ast_analysis/scanner.py
src/traceml_ai/utils/ast_analysis/visitor.py
tests/__init__.py
tests/conftest.py
tests/test_h2d_timing.py
tests/aggregator/test_aggregator_main_env.py
tests/aggregator/test_finalization.py
tests/aggregator/test_step_time_sqlite_writer.py
tests/aggregator/test_summary_service.py
tests/benchmarks/bench_tcp_drain.py
tests/config/__init__.py
tests/config/test_yaml_loader.py
tests/core/__init__.py
tests/core/test_packaging_constraints.py
tests/core/test_protocols.py
tests/core/test_registry.py
tests/diagnostics/__init__.py
tests/diagnostics/test_bands.py
tests/diagnostics/test_common.py
tests/diagnostics/test_process.py
tests/diagnostics/test_registry.py
tests/diagnostics/test_step_memory.py
tests/diagnostics/test_step_memory_package.py
tests/diagnostics/test_step_time.py
tests/diagnostics/test_system.py
tests/diagnostics/test_trend_core.py
tests/display/__init__.py
tests/display/test_model_combined_section.py
tests/display/test_nicegui_driver.py
tests/display/test_nicegui_errors.py
tests/display/test_nicegui_subscribers.py
tests/display/test_renderer_contracts.py
tests/display/test_server_readiness.py
tests/display/test_staleness.py
tests/display/test_step_memory_formatter.py
tests/integrations/__init__.py
tests/integrations/test_deepspeed.py
tests/integrations/test_hf_trainer.py
tests/integrations/test_lightning.py
tests/integrations/test_ray.py
tests/renderers/test_step_time_compute.py
tests/renderers/test_step_time_renderer.py
tests/reporting/__init__.py
tests/reporting/test_view.py
tests/reporting/compare/__init__.py
tests/reporting/compare/test_compare.py
tests/reporting/html/__init__.py
tests/reporting/html/conftest.py
tests/reporting/html/test_artifacts.py
tests/reporting/html/test_banner.py
tests/reporting/html/test_cli.py
tests/reporting/html/test_file_api.py
tests/reporting/html/test_plumbing.py
tests/reporting/html/test_protocol_path.py
tests/reporting/html/test_render.py
tests/reporting/html/test_sections.py
tests/reporting/html/test_svg.py
tests/reporting/html/test_textutils.py
tests/reporting/summary/__init__.py
tests/reporting/summary/test_final.py
tests/reporting/summary/test_fixtures.py
tests/reporting/summary/test_primary_diagnosis.py
tests/reporting/summary/test_step_memory.py
tests/reporting/summary/test_step_time.py
tests/reporting/summary/test_step_time_card.py
tests/reporting/summary/test_system_process.py
tests/runtime/__init__.py
tests/runtime/test_environment.py
tests/runtime/test_environment_state.py
tests/runtime/test_executor.py
tests/runtime/test_exporter.py
tests/runtime/test_final_summary_smoke.py
tests/runtime/test_identity.py
tests/runtime/test_launcher.py
tests/runtime/test_lifecycle.py
tests/runtime/test_runtime_environment_sampler.py
tests/runtime/test_runtime_identity_wiring.py
tests/runtime/test_sampler_registry.py
tests/runtime/test_stderr_capture.py
tests/runtime/test_step_time_sampler.py
tests/runtime/test_tcp_transport.py
tests/runtime/test_telemetry_publisher.py
tests/runtime/test_trace_session_state.py
tests/sdk/__init__.py
tests/sdk/test_grad_accum.py
tests/sdk/test_import_compat.py
tests/sdk/test_init_and_wrappers.py
tests/sdk/test_init_runtime.py
tests/sdk/test_instrumentation_namespace.py
tests/sdk/test_summary_client.py
tests/telemetry/__init__.py
tests/telemetry/test_distributed.py
tests/telemetry/test_msgpack.py
tests/telemetry/test_runtime_environment_projection.py
tests/telemetry/test_sender_sequence.py
tests/telemetry/test_system_projection_identity.py