Easily fine-tune (SFT/RL), evaluate, and deploy Qwen, Gemma, or any open agentic LLM/VLM!
This report summarizes scanner evidence for oumi-ai/oumi, a Python project with 9,391 GitHub stars. SynthScan examined 250,274 lines of code across 1450 source files, recording 7062 pattern matches distributed across 24 syntactic categories. The legacy adjusted pattern signal is 37.0 (High pattern signal); it is not an authorship estimate.
Use the findings to ask whether maintainers can explain intent, verify dependencies, simplify unnecessary scaffolding, and focus review effort. Risk Payload v1 adds ownership, supply-chain, over-engineering, and review-effort vectors when available. These are prompts for human verification, not probabilities, security findings, or measured maintenance costs.
Use each vector to guide manual verification of the underlying findings.
Can maintainers explain intent, assumptions, and trade-offs? 5778 finding(s).
Are imports declared and package/API names verified? 580 finding(s).
Are wrappers, guards, and repeated structures necessary? 4452 finding(s).
Which changes need focused human review? 7062 finding(s).
This chart tracks the legacy adjusted pattern score across scan runs. Changes can reflect source changes, scan coverage, or detector/scoring versions; they do not measure AI adoption or prove that a team refactored code. Compare runs only when their scoring versions are compatible. The dashed line tracks the raw pattern-hit count.
Groups findings by configured severity weight. Severity is not a confidence estimate, security rating, or proof that a package is missing. Verify imports against the project's manifests and inspect each finding in its local design context.
This horizontal bar chart shows where configured pattern scores accumulate across directories. Treat the bars as locations to inspect, not as evidence of authorship or code quality. Use the file-level findings to decide whether a directory needs a focused ownership, dependency, complexity, or review check.
The scanner identified 7062 distinct pattern matches across 24 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a detector match to inspect in context. Expand any category row to inspect the individual file paths, line numbers, code snippets, and the lexical context (CODE, COMMENT, or STRING) in which each match was detected.
Reading the findings table: The Severity column indicates the configured finding weight (CRITICAL / HIGH / MEDIUM / LOW). The Context column identifies whether the match occurred inside executable code, an inline comment, or a string literal — legacy comment-context matches receive a ×1.5 score weight. The ⚡ bolt icon marks clustered matches: three or more patterns within a 10-line window, each receiving an additional ×1.5 density multiplier under the legacy scoring model.
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | platform/banking77-classifier/preprocess_banking77.py | 26 | def _load_in_context_examples() -> list[str]: | CODE |
| LOW | platform/banking77-classifier/preprocess_banking77.py | 37 | def _build_classifier_instruction( | CODE |
| LOW | tests/markers.py | 47 | def requires_cuda_initialized() -> pytest.MarkDecorator: | CODE |
| LOW | tests/markers.py | 59 | def requires_cuda_not_available() -> pytest.MarkDecorator: | CODE |
| LOW | tests/unit/test_train.py | 78 | def mock_build_dataset_mixture(): | CODE |
| LOW | tests/unit/test_train.py | 98 | def test_train_function_passes_debug_flag_correctly( | CODE |
| LOW | tests/unit/test_train.py | 141 | def _config_with_collator_kwargs(collator_kwargs): | CODE |
| LOW | tests/unit/test_train.py | 167 | def test_recompile_limit_unset_without_pad_to_multiple_of( | CODE |
| LOW | tests/unit/test_train.py | 182 | def test_recompile_limit_self_set_with_pad_to_multiple_of( | CODE |
| LOW | tests/unit/test_train.py | 200 | def test_recompile_limit_env_override_wins(monkeypatch, restore_dynamo_limits): | CODE |
| LOW⚡ | tests/unit/quantize/test_awq_quantizer.py | 48 | def test_supports_method_valid(self): | CODE |
| LOW⚡ | tests/unit/quantize/test_awq_quantizer.py | 55 | def test_supports_method_invalid(self): | CODE |
| LOW⚡ | tests/unit/quantize/test_awq_quantizer.py | 61 | def test_validate_config_valid(self): | CODE |
| LOW⚡ | tests/unit/quantize/test_awq_quantizer.py | 66 | def test_validate_config_invalid_method(self): | CODE |
| LOW | tests/unit/quantize/test_awq_quantizer.py | 77 | def test_validate_config_invalid_format(self): | CODE |
| LOW⚡ | tests/unit/quantize/test_awq_quantizer.py | 92 | def test_raise_if_requirements_not_met_missing_awq(self): | CODE |
| LOW⚡ | tests/unit/quantize/test_awq_quantizer.py | 102 | def test_raise_if_requirements_not_met_no_gpu(self): | CODE |
| LOW⚡ | tests/unit/quantize/test_awq_quantizer.py | 127 | def test_validate_config_valid_simple(self): | CODE |
| LOW⚡ | tests/unit/quantize/test_awq_quantizer.py | 137 | def test_raise_if_requirements_not_met_no_awq_simple(self): | CODE |
| LOW⚡ | tests/unit/quantize/test_awq_quantizer.py | 144 | def test_raise_if_requirements_not_met_no_gpu_simple(self, mock_cuda): | CODE |
| LOW | tests/unit/quantize/test_base_quantization.py | 39 | def raise_if_requirements_not_met(self) -> None: | CODE |
| LOW⚡ | tests/unit/quantize/test_base_quantization.py | 57 | def test_supports_method_true(self): | CODE |
| LOW⚡ | tests/unit/quantize/test_base_quantization.py | 61 | def test_supports_method_false(self): | CODE |
| LOW⚡ | tests/unit/quantize/test_base_quantization.py | 65 | def test_supports_format_true(self): | CODE |
| LOW⚡ | tests/unit/quantize/test_base_quantization.py | 70 | def test_supports_format_false(self): | CODE |
| LOW⚡ | tests/unit/quantize/test_base_quantization.py | 74 | def test_validate_config_valid(self): | CODE |
| LOW⚡ | tests/unit/quantize/test_base_quantization.py | 79 | def test_validate_config_invalid_method(self): | CODE |
| LOW | tests/unit/quantize/test_base_quantization.py | 91 | def test_quantize_implementation(self): | CODE |
| LOW | tests/unit/quantize/test_base_quantization.py | 105 | def test_quantization_result_creation(self): | CODE |
| LOW | tests/unit/quantize/test_base_quantization.py | 121 | def test_quantization_result_default_additional_info(self): | CODE |
| LOW | tests/unit/quantize/test_quantize_module.py | 39 | def test_quantize_awq_success(self, mock_build_quantizer): | CODE |
| LOW | tests/unit/quantize/test_quantize_module.py | 69 | def test_quantize_bnb_success(self, mock_build_quantizer): | CODE |
| LOW⚡ | tests/unit/quantize/test_quantize_module.py | 100 | def test_quantize_invalid_config_type(self): | CODE |
| LOW⚡ | tests/unit/quantize/test_quantize_module.py | 105 | def test_quantize_unsupported_method(self): | CODE |
| LOW⚡ | tests/unit/quantize/test_quantize_module.py | 115 | def test_quantize_requirements_not_met(self, mock_build_quantizer): | CODE |
| LOW | tests/unit/quantize/test_quantize_module.py | 128 | def test_quantize_quantizer_failure(self, mock_build_quantizer): | CODE |
| LOW | tests/unit/quantize/test_quantize_module.py | 140 | def test_quantize_different_output_formats(self, mock_build_quantizer): | CODE |
| LOW | tests/unit/quantize/test_quantize_module.py | 168 | def test_quantize_return_type(self, mock_build_quantizer): | CODE |
| LOW | tests/unit/quantize/test_bnb_quantizer.py | 59 | def test_supports_method_invalid(self): | CODE |
| LOW | tests/unit/quantize/test_bnb_quantizer.py | 74 | def test_validate_config_invalid_method(self): | CODE |
| LOW | tests/unit/quantize/test_bnb_quantizer.py | 88 | def test_raise_if_requirements_not_met_missing_bnb(self): | CODE |
| LOW | tests/unit/quantize/test_bnb_quantizer.py | 113 | def test_validate_config_valid(self): | CODE |
| LOW | tests/unit/launcher/test_launcher.py | 85 | def test_launcher_get_cloud_missing_value(mock_registry): | CODE |
| LOW | tests/unit/launcher/test_launcher.py | 106 | def test_launcher_get_cloud_empty(mock_registry): | CODE |
| LOW | tests/unit/launcher/test_launcher.py | 115 | def test_launcher_get_cloud_by_name_missing_value(mock_registry): | CODE |
| LOW | tests/unit/launcher/test_launcher.py | 136 | def test_launcher_get_cloud_by_name_empty(mock_registry): | CODE |
| LOW | tests/unit/launcher/test_launcher.py | 145 | def test_launcher_up_succeeds(mock_registry): | CODE |
| LOW | tests/unit/launcher/test_launcher.py | 175 | def test_launcher_up_succeeds_kwargs(mock_registry): | CODE |
| LOW | tests/unit/launcher/test_launcher.py | 205 | def test_launcher_up_succeeds_no_name(mock_registry): | CODE |
| LOW | tests/unit/launcher/test_launcher.py | 235 | def test_launcher_up_inavlid_cluster(mock_registry): | CODE |
| LOW | tests/unit/launcher/test_launcher.py | 263 | def test_launcher_run_succeeds(mock_registry): | CODE |
| LOW | tests/unit/launcher/test_launcher.py | 311 | def test_launcher_cancel_succeeds(mock_registry): | CODE |
| LOW | tests/unit/launcher/test_launcher.py | 340 | def test_launcher_cancel_fails(mock_registry): | CODE |
| LOW | tests/unit/launcher/test_launcher.py | 357 | def test_launcher_down_succeeds(mock_registry): | CODE |
| LOW | tests/unit/launcher/test_launcher.py | 392 | def test_launcher_status_multiple_clouds(mock_registry): | CODE |
| LOW | tests/unit/launcher/test_launcher.py | 533 | def test_launcher_status_filters_clusters(mock_registry): | CODE |
| LOW | tests/unit/launcher/test_launcher.py | 651 | def test_launcher_status_filters_jobs(mock_registry): | CODE |
| LOW | tests/unit/launcher/test_launcher.py | 783 | def test_launcher_status_filters_clouds(mock_registry): | CODE |
| LOW | tests/unit/launcher/test_launcher.py | 903 | def test_launcher_status_all_filters(mock_registry): | CODE |
| LOW | tests/unit/launcher/test_launcher.py | 1010 | def test_launcher_status_inits_new_clouds(mock_registry): | CODE |
| 3670 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | CONTRIBUTING.md | 81 | COMMENT | |
| LOW | platform/banking77-classifier/preprocess_banking77.py | 141 | print("\n 2-shot sample (system prompt, last 200 chars):") | COMMENT |
| LOW | platform/banking77-classifier/preprocess_banking77.py | 161 | # transform_row( | COMMENT |
| LOW | platform/banking77-classifier/preprocess_banking77.py | 181 | # SCRIPT_DIR / "banking77-test-3-shot.jsonl", | COMMENT |
| LOW | tests/unit/quantize/test_awq_quantizer.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/quantize/test_base_quantization.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/quantize/__init__.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/quantize/test_quantize_module.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/quantize/test_bnb_quantizer.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/core/test_length_analyzer.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/core/tokenizers/test_special_tokens.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | …ts/unit/core/callbacks/test_metrics_logger_callback.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/core/synthesis/test_synthesis_pipeline.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/core/synthesis/test_tool_router.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/core/synthesis/test_dataset_planner.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/core/synthesis/test_data_synthesizer.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/core/synthesis/test_attribute_formatter.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/core/synthesis/test_attribute_synthesizer.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | …s/unit/core/synthesis/test_attribute_transformation.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/core/synthesis/test_dataset_ingestion.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/core/synthesis/test_document_ingestion.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | …s/unit/core/synthesis/test_conversation_synthesizer.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/core/trainers/test_trl_dpo_trainer.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/core/configs/test_analyze_config.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/core/configs/test_synthesis_config.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/core/configs/params/test_deepspeed_params.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/core/configs/params/test_gkd_params.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/core/configs/params/test_training_params.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | …ts/unit/core/configs/params/test_environment_params.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/core/configs/params/test_gold_params.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/core/configs/params/test_synthesis_params.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/core/configs/params/test_tool_params.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/core/configs/params/test_grounding_params.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/builders/test_quantizers.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/builders/test_environments.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/builders/test_adapter_trainable.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/deploy/test_base_client.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/deploy/test_utils.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/deploy/test_fireworks_client.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/deploy/test_parasail_client.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/deploy/__init__.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/deploy/test_deploy_config.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/deploy/test_fireworks_errors.py | 1 | # Copyright 2026 - Oumi | COMMENT |
| LOW | tests/unit/environments/test_database_session.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/environments/test_executor_resolution.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/environments/test_executable_environment.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/environments/test_executable_tool.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | …t/environments/test_database_executable_environment.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/environments/test_base_environment.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/environments/test_simulated_environment.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/environments/test_lookup_environment.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/environments/test_endpoint_environment.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | …sts/unit/datasets/test_vision_dpo_jsonlines_dataset.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/datasets/test_base_dpo_dataset.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/datasets/test_text_dpo_jsonlines_dataset.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/datasets/grpo/test_letter_count.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/utils/test_placeholders.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/utils/test_analysis_utils.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/utils/test_grpo_utils.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| LOW | tests/unit/utils/test_http.py | 1 | # Copyright 2025 - Oumi | COMMENT |
| 953 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | platform/banking77-classifier/preprocess_banking77.py | 115 | # ── Step 1: Download and read dataset ────────────────────────────── | COMMENT |
| MEDIUM | platform/banking77-classifier/preprocess_banking77.py | 126 | # ── Step 2: Verify transform function ────────────────────────────── | COMMENT |
| MEDIUM | platform/banking77-classifier/preprocess_banking77.py | 146 | # ── Step 3: Create JSONL datasets ─────────────────────────────────── | COMMENT |
| MEDIUM⚡ | tests/unit/core/types/test_conversation.py | 1304 | # ----------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | tests/unit/core/types/test_conversation.py | 1306 | # ----------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | tests/unit/core/types/test_conversation.py | 1442 | # ----------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | tests/unit/core/types/test_conversation.py | 1444 | # ----------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tests/unit/core/types/test_conversation.py | 1045 | # ----------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tests/unit/core/types/test_conversation.py | 1047 | # ----------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tests/unit/core/types/test_conversation.py | 1215 | # ----------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tests/unit/core/types/test_conversation.py | 1217 | # ----------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | tests/unit/core/synthesis/test_dataset_planner.py | 1195 | # =========================== | COMMENT |
| MEDIUM⚡ | tests/unit/core/synthesis/test_dataset_planner.py | 1197 | # =========================== | COMMENT |
| MEDIUM⚡ | …s/unit/core/synthesis/test_conversation_synthesizer.py | 1770 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | …s/unit/core/synthesis/test_conversation_synthesizer.py | 1772 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | …s/unit/core/synthesis/test_conversation_synthesizer.py | 3002 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | …s/unit/core/synthesis/test_conversation_synthesizer.py | 3004 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …s/unit/core/synthesis/test_conversation_synthesizer.py | 2118 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …s/unit/core/synthesis/test_conversation_synthesizer.py | 2120 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | …llators/test_text_completions_collator_with_padding.py | 325 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | …llators/test_text_completions_collator_with_padding.py | 327 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | …llators/test_text_completions_collator_with_padding.py | 345 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | …llators/test_text_completions_collator_with_padding.py | 347 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | …llators/test_text_completions_collator_with_padding.py | 393 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | …llators/test_text_completions_collator_with_padding.py | 395 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | …llators/test_text_completions_collator_with_padding.py | 456 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | …llators/test_text_completions_collator_with_padding.py | 458 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | …llators/test_text_completions_collator_with_padding.py | 510 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | …llators/test_text_completions_collator_with_padding.py | 512 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | …llators/test_text_completions_collator_with_padding.py | 670 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | …llators/test_text_completions_collator_with_padding.py | 672 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | …llators/test_text_completions_collator_with_padding.py | 859 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | …llators/test_text_completions_collator_with_padding.py | 861 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | …llators/test_text_completions_collator_with_padding.py | 919 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | …llators/test_text_completions_collator_with_padding.py | 921 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …llators/test_text_completions_collator_with_padding.py | 279 | # =========================================================================== | COMMENT |
| MEDIUM | …llators/test_text_completions_collator_with_padding.py | 281 | # =========================================================================== | COMMENT |
| MEDIUM⚡ | tests/unit/builders/test_collators.py | 451 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | tests/unit/builders/test_collators.py | 453 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | tests/unit/builders/test_collators.py | 510 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | tests/unit/builders/test_collators.py | 512 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tests/unit/builders/test_collators.py | 264 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tests/unit/builders/test_collators.py | 267 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tests/unit/builders/test_collators.py | 554 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tests/unit/builders/test_collators.py | 556 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | tests/unit/deploy/test_parasail_client.py | 648 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | tests/unit/deploy/test_parasail_client.py | 650 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | tests/unit/deploy/test_parasail_client.py | 683 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | tests/unit/deploy/test_parasail_client.py | 685 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | tests/unit/deploy/test_parasail_client.py | 723 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | tests/unit/deploy/test_parasail_client.py | 725 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | tests/unit/deploy/test_parasail_client.py | 757 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | tests/unit/deploy/test_parasail_client.py | 759 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | tests/unit/deploy/test_parasail_client.py | 557 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | tests/unit/deploy/test_parasail_client.py | 559 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | tests/unit/deploy/test_parasail_client.py | 627 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | tests/unit/deploy/test_parasail_client.py | 629 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | tests/unit/deploy/test_deploy_config.py | 150 | # =================================================================== | COMMENT |
| MEDIUM⚡ | tests/unit/deploy/test_deploy_config.py | 152 | # =================================================================== | COMMENT |
| MEDIUM⚡ | tests/unit/deploy/test_deploy_config.py | 185 | # =================================================================== | COMMENT |
| 193 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | install.sh | 3 | # Usage: | COMMENT |
| LOW | configs/apis/gemini/infer_gemini_2_5_pro.yaml | 6 | # Usage: | COMMENT |
| LOW | configs/apis/openrouter/infer_gpt_5_2.yaml | 6 | # Usage: | COMMENT |
| LOW | configs/apis/openrouter/infer_llama4_maverick.yaml | 6 | # Usage: | COMMENT |
| LOW | configs/apis/openrouter/infer_claude_4_5_sonnet.yaml | 6 | # Usage: | COMMENT |
| LOW | configs/apis/atlascloud/infer_deepseek_v3.yaml | 7 | # Usage: | COMMENT |
| LOW | configs/apis/anthropic/infer_claude_opus_4_1.yaml | 6 | # Usage: | COMMENT |
| LOW | …s/apis/anthropic/infer_claude_sonnet_tool_calling.yaml | 12 | # Usage: | COMMENT |
| LOW | configs/apis/huggingface/infer_llama_3_1_8b.yaml | 15 | # Usage: | COMMENT |
| LOW | configs/apis/openai/infer_gpt_4_1.yaml | 6 | # Usage: | COMMENT |
| LOW | configs/apis/openai/infer_gpt_4o_mini.yaml | 6 | # Usage: | COMMENT |
| LOW | configs/apis/openai/infer_o1_mini.yaml | 6 | # Usage: | COMMENT |
| LOW | configs/apis/openai/infer_gpt_5.yaml | 6 | # Usage: | COMMENT |
| LOW | configs/apis/openai/infer_gpt_5_chat_latest.yaml | 6 | # Usage: | COMMENT |
| LOW | configs/apis/openai/infer_gpt_4o_tool_calling.yaml | 10 | # Usage: | COMMENT |
| LOW | configs/apis/openai/infer_gpt_5_nano.yaml | 6 | # Usage: | COMMENT |
| LOW | configs/apis/openai/infer_o1.yaml | 6 | # Usage: | COMMENT |
| LOW | configs/apis/openai/infer_o3_mini.yaml | 6 | # Usage: | COMMENT |
| LOW | configs/apis/openai/infer_gpt_5_mini.yaml | 6 | # Usage: | COMMENT |
| LOW | configs/apis/openai/infer_chatgpt_4o_latest.yaml | 6 | # Usage: | COMMENT |
| LOW | configs/apis/openai/infer_gpt_4_1_mini.yaml | 6 | # Usage: | COMMENT |
| LOW | …nfigs/projects/limo/qwen2.5_7b_fft_yarn_deepspeed.yaml | 10 | # Usage: | COMMENT |
| LOW | …n2.5_7b_fft_yarn_deepspeed_memory_optimized_train.yaml | 15 | # Usage: | COMMENT |
| LOW | configs/projects/limo/qwen2.5_7b_fft.yaml | 9 | # Usage: | COMMENT |
| LOW | configs/projects/limo/qwen2.5_7b_fft_yarn.yaml | 10 | # Usage: | COMMENT |
| LOW | …s/projects/dcvlr/starter_kit/molmo-d-train-openr1.yaml | 9 | # Usage: | COMMENT |
| LOW | …s/projects/dcvlr/starter_kit/molmo-o-train-openr1.yaml | 9 | # Usage: | COMMENT |
| LOW | …figs/projects/judges/rule_based/regex_match_phone.yaml | 8 | # Usage: | COMMENT |
| LOW | …rojects/judges/rule_based/regex_no_error_keywords.yaml | 8 | # Usage: | COMMENT |
| LOW | configs/projects/chatqa/chatqa_stage2_train.yaml | 3 | # Usage: | COMMENT |
| LOW | configs/projects/chatqa/gcp_job.yaml | 9 | # Usage: | COMMENT |
| LOW | configs/projects/chatqa/chatqa_stage1_train.yaml | 3 | # Usage: | COMMENT |
| LOW | configs/projects/coalm/70b_infer.yaml | 3 | # Usage: | COMMENT |
| LOW | configs/projects/coalm/8b_infer.yaml | 3 | # Usage: | COMMENT |
| LOW | configs/projects/aya/evaluation/gcp_job.yaml | 9 | # Usage: | COMMENT |
| LOW | configs/projects/aya/evaluation/eval.yaml | 8 | # Usage: | COMMENT |
| LOW | configs/projects/aya/sft/gcp_job.yaml | 9 | # Usage: | COMMENT |
| LOW | configs/projects/aya/sft/train.yaml | 8 | # Usage: | COMMENT |
| LOW | configs/projects/halloumi/gcp_job.yaml | 9 | # Usage: | COMMENT |
| LOW | configs/projects/halloumi/8b_train.yaml | 8 | # Usage: | COMMENT |
| LOW | …igs/recipes/llama4/inference/scout_instruct_infer.yaml | 7 | # Usage: | COMMENT |
| LOW | …es/llama4/inference/scout_instruct_together_infer.yaml | 6 | # Usage: | COMMENT |
| LOW | …ecipes/llama4/inference/scout_instruct_vllm_infer.yaml | 8 | # Usage: | COMMENT |
| LOW | …llama4/inference/maverick_instruct_together_infer.yaml | 6 | # Usage: | COMMENT |
| LOW | …/llama4/inference/scout_instruct_gguf_macos_infer.yaml | 6 | # Usage: | COMMENT |
| LOW | …ecipes/llama4/inference/scout_instruct_gguf_infer.yaml | 6 | # Usage: | COMMENT |
| LOW | …lama4/inference/maverick_instruct_fireworks_infer.yaml | 6 | # Usage: | COMMENT |
| LOW | …igs/recipes/llama4/evaluation/scout_instruct_eval.yaml | 8 | # Usage: | COMMENT |
| LOW | …figs/recipes/llama4/sft/scout_instruct_lora/train.yaml | 8 | # Usage: | COMMENT |
| LOW | …igs/recipes/llama4/sft/scout_instruct_qlora/train.yaml | 8 | # Usage: | COMMENT |
| LOW | configs/recipes/llama4/sft/scout_base_full/gcp_job.yaml | 9 | # Usage: | COMMENT |
| LOW | configs/recipes/llama4/sft/scout_base_full/train.yaml | 8 | # Usage: | COMMENT |
| LOW | …gs/recipes/llama4/sft/scout_instruct_full/gcp_job.yaml | 9 | # Usage: | COMMENT |
| LOW | …figs/recipes/llama4/sft/scout_instruct_full/train.yaml | 8 | # Usage: | COMMENT |
| LOW | configs/recipes/olmo3/inference/7b_infer.yaml | 13 | # Usage: | COMMENT |
| LOW | configs/recipes/olmo3/inference/32b_infer.yaml | 13 | # Usage: | COMMENT |
| LOW | configs/recipes/olmo3/evaluation/7b/eval.yaml | 8 | # Usage: | COMMENT |
| LOW | configs/recipes/olmo3/evaluation/32b/eval.yaml | 8 | # Usage: | COMMENT |
| LOW | configs/recipes/olmo3/sft/7b_full/train.yaml | 11 | # Usage: | COMMENT |
| LOW | configs/recipes/olmo3/sft/32b_lora/train.yaml | 11 | # Usage: | COMMENT |
| 427 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | docs/_summarize_module.py | 29 | Generate a markdown table of objects defined in a Python module. Args: module_name: The name of the module | STRING |
| HIGH | docs/_summarize_module.py | 130 | Generate a markdown table summarizing config files in a folder. Args: config_folder: The folder containing | STRING |
| HIGH | scripts/benchmarks/benchmark_dataloader.py | 179 | Measures the time taken to initialize a dataset using the given dataset function. Parameters: dataset_fn (c | STRING |
| HIGH | src/oumi/__init__.py | 279 | Quantizes a model using the provided configuration. Args: config: Quantization configuration containing mod | STRING |
| HIGH | src/oumi/judge.py | 33 | Create the appropriate judge based on the configuration. Args: judge_config: JudgeConfig object or path to | STRING |
| HIGH | src/oumi/judge.py | 133 | Judge a dataset from a JSONL file using Oumi's Judge framework. This is a convenience wrapper around judge_dataset. | STRING |
| HIGH | src/oumi/judge.py | 167 | Judge a list of conversations from a JSONL file using Oumi's Judge framework. This is a convenience wrapper around | STRING |
| HIGH | src/oumi/quantize/__init__.py | 32 | Main quantization function that routes to appropriate quantizer. Args: config: Quantization configuration c | STRING |
| HIGH | src/oumi/quantize/base.py | 57 | Main quantization method - must be implemented by subclasses. Args: config: Quantization configurat | STRING |
| HIGH | src/oumi/launcher/clients/sky_client.py | 266 | Wait for a submitted launch and return its job information. Args: request_id: ID returned by :meth: | STRING |
| HIGH | src/oumi/launcher/clients/sky_client.py | 342 | Get the current status of a launch request. Args: request_id: ID returned by :meth:`submit`. | STRING |
| HIGH | src/oumi/launcher/clusters/slurm_cluster.py | 102 | Translates an oumi accelerator string into a Slurm ``--gres`` value. Slurm's typed-GRES requires the type string to | STRING |
| HIGH | src/oumi/core/distributed.py | 319 | Wrap the model for distributed training (DDP, FSDP, or DeepSpeed). Args: model: The model to be wrapped. | STRING |
| HIGH | src/oumi/core/datasets/base_sft_dataset.py | 173 | Applies the chat template carried by the tokenizer to the input example. Args: sample (Dict): Mappi | STRING |
| HIGH | src/oumi/core/datasets/base_kto_dataset.py | 125 | Validate and transform a sample to the KTO format. This method processes raw dataset samples and converts them | STRING |
| HIGH | src/oumi/core/datasets/base_kto_dataset.py | 187 | Extract the assistant's response from a chat-formatted completion. This method handles completions that are pro | STRING |
| HIGH | src/oumi/core/models/base_model.py | 124 | Loads a model from a directory saved with `save_pretrained()`. This classmethod instantiates a model and loads | STRING |
| HIGH | src/oumi/core/synthesis/conversation_synthesizer.py | 294 | Synthesize a multi-turn conversation. Order will be identical to the order of the samples. Args: | STRING |
| HIGH | src/oumi/core/synthesis/conversation_synthesizer.py | 531 | Plan the conversation samples with retry logic for failed parses. Args: samples: The conversation s | STRING |
| HIGH | src/oumi/core/synthesis/attribute_synthesizer.py | 228 | Submit a batch inference job for attribute synthesis. Args: samples: The samples to synthesize valu | STRING |
| HIGH | src/oumi/core/synthesis/attribute_synthesizer.py | 255 | Get the status of a batch inference job. Args: batch_id: The batch ID returned from synthesize_batc | STRING |
| HIGH | src/oumi/core/synthesis/attribute_synthesizer.py | 274 | Cancel a batch inference job. Args: batch_id: The batch ID returned from synthesize_batch(). | STRING |
| HIGH | src/oumi/core/synthesis/attribute_synthesizer.py | 298 | Get results from a completed batch inference job. Args: batch_id: The batch ID returned from synthe | STRING |
| HIGH | src/oumi/core/synthesis/attribute_synthesizer.py | 330 | Get partial results from a completed batch inference job. This method returns successful results alongside fail | STRING |
| HIGH | src/oumi/core/synthesis/synthesis_pipeline.py | 179 | Extract a value from a nested structure using bracket notation. Args: sample: The sample dictionary | STRING |
| HIGH | src/oumi/core/trainers/verl_grpo_trainer.py | 276 | Finds prompt, answer, and optional images in a conversation. Supports both single-turn and multi-turn conversat | STRING |
| HIGH | src/oumi/core/inference/base_inference_engine.py | 795 | Gets partial results of a completed batch job. Engines that support batch inference should override this method | STRING |
| HIGH | src/oumi/core/analyze/dataset_analyzer.py | 420 | Query the analysis results using pandas query syntax. Args: query_expression: Pandas query expressi | STRING |
| HIGH | src/oumi/core/analyze/dataset_analyzer.py | 521 | Query conversation-level analysis results using pandas query expression. Args: query_expression: Pa | STRING |
| HIGH | src/oumi/core/analyze/dataset_analyzer.py | 561 | Filter the original dataset based on analysis results. This method uses analysis results to filter the original | STRING |
| HIGH | src/oumi/core/analyze/dataset_analyzer.py | 615 | Create a new dataset containing only the specified conversations. Args: conversation_indices: List | STRING |
| HIGH | src/oumi/core/collators/vision_language_sft_collator.py | 209 | Process a batch of conversation data into model-ready features. This method converts serialized conversations i | STRING |
| HIGH | src/oumi/core/collators/vision_language_sft_collator.py | 300 | Collate individually processed results by padding to max dimensions. Args: results: List of feature | STRING |
| HIGH | …/collators/trl_data_collator_for_completion_only_lm.py | 195 | Re-masks tool results that the chat template nested in an unmasked span. Some templates (e.g. gemma-4) render t | STRING |
| HIGH | …/collators/trl_data_collator_for_completion_only_lm.py | 280 | Apply span-based masking for multi-turn conversations. Masks all labels, then unmarks assistant response spans | STRING |
| HIGH | …/collators/trl_data_collator_for_completion_only_lm.py | 363 | Collates a list of examples into a batch. Args: examples: Examples to collate, each a list of token | STRING |
| HIGH | src/oumi/builders/models.py | 574 | Builds a chat template based on code name. Args: template_name: the code name describing the chat-template. | STRING |
| HIGH | src/oumi/builders/inference_engines.py | 82 | Returns the inference engine based on the provided config. Args: engine_type: Type of inference engine to c | STRING |
| HIGH | src/oumi/builders/quantizers.py | 26 | Create appropriate quantization instance based on method. Args: method: Quantization method name (e.g., "aw | STRING |
| HIGH | src/oumi/builders/tuning.py | 20 | Build a tuner based on the configuration. Args: tuning_params: Tuning configuration parameters. Return | STRING |
| HIGH | src/oumi/builders/collators.py | 91 | Detect end-of-turn token IDs and template string. Compares token-ID prefixes of the text after the last assistant t | STRING |
| HIGH | src/oumi/builders/collators.py | 166 | Detect the assistant response header from the user-to-assistant boundary. Strips the leading end-of-turn prefix (wh | STRING |
| HIGH | src/oumi/builders/collators.py | 239 | Clamp the response template to the generation-prompt boundary. Some chat templates render assistant headers that ex | STRING |
| HIGH | src/oumi/builders/collators.py | 321 | Auto-detect response_template and end_of_turn_template. Applies the chat template to a known test conversation, the | STRING |
| HIGH | src/oumi/builders/collators.py | 427 | The bracket around tool results this model nests inside the assistant turn. Span masking unmasks everything between | STRING |
| HIGH | src/oumi/builders/collators.py | 506 | Builds a data collator based on the given collator name. Args: collator_name: The name of the collator to b | STRING |
| HIGH | src/oumi/builders/collators.py | 620 | Creates data collator if specified in config. Resolves everything the collator needs from the config: the label ign | STRING |
| HIGH | src/oumi/builders/training.py | 44 | Builds a trainer creator functor based on the provided configuration. Args: trainer_type (TrainerType): Enu | STRING |
| HIGH | src/oumi/deploy/deploy_config.py | 137 | Loads and validates a deployment config from a YAML file. Args: config_path: Path to the YAML file. | STRING |
| HIGH | src/oumi/deploy/base_client.py | 302 | Gets the status of an asynchronous upload/processing job. Override in providers that use job-based workflows. | STRING |
| HIGH | src/oumi/deploy/base_client.py | 329 | Starts a stopped endpoint (scales up from 0 replicas). Args: endpoint_id: Provider-specific endpoin | STRING |
| HIGH | src/oumi/deploy/base_client.py | 346 | Stops an endpoint by scaling to 0 replicas (cost savings). Args: endpoint_id: Provider-specific end | STRING |
| HIGH | src/oumi/deploy/fireworks_client.py | 528 | Uploads a model using a pre-computed file inventory and resolver. Unlike ``upload_model()`` which requires all | STRING |
| HIGH | src/oumi/deploy/fireworks_client.py | 1315 | Gets the status of an uploaded model. Args: model_id: Fireworks model ID (short ID or full path) | STRING |
| HIGH | src/oumi/deploy/fireworks_client.py | 1388 | Creates an inference endpoint (deployment) for a model. Deploys either by raw ``hardware`` or by a validated | STRING |
| HIGH | src/oumi/deploy/parasail_client.py | 208 | Validates the HuggingFace model source and checks compatibility. Args: model_source: HuggingFace re | STRING |
| HIGH | src/oumi/deploy/parasail_client.py | 322 | Creates a dedicated Parasail endpoint for a HuggingFace model. Fetches compatible device configurations from `` | STRING |
| HIGH | src/oumi/deploy/parasail_client.py | 392 | Gets details of a dedicated endpoint. Args: endpoint_id: Numeric deployment ID (as a string). | STRING |
| HIGH | src/oumi/deploy/parasail_client.py | 423 | Updates a dedicated endpoint's replica count. Parasail's update flow requires fetching the current deployment a | STRING |
| HIGH | src/oumi/deploy/parasail_client.py | 499 | Lists available hardware configurations for Parasail. Queries the ``/dedicated/devices`` API for hardware compa | STRING |
| 48 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | docs/user_guides/launch/custom_cluster.md | 0 | gets the jobs on this cluster if it exists, else returns none. | STRING |
| HIGH | src/oumi/launcher/clusters/sky_cluster.py | 0 | gets the jobs on this cluster if it exists, else returns none. | STRING |
| HIGH | src/oumi/launcher/clusters/local_cluster.py | 0 | gets the jobs on this cluster if it exists, else returns none. | STRING |
| HIGH | src/oumi/launcher/clusters/polaris_cluster.py | 0 | gets the jobs on this cluster if it exists, else returns none. | STRING |
| HIGH | src/oumi/launcher/clusters/frontier_cluster.py | 0 | gets the jobs on this cluster if it exists, else returns none. | STRING |
| HIGH | src/oumi/launcher/clusters/perlmutter_cluster.py | 0 | gets the jobs on this cluster if it exists, else returns none. | STRING |
| HIGH | docs/user_guides/launch/custom_cluster.md | 0 | gets the cluster with the specified name, or creates one if it doesn't exist. args: name: the name of the cluster. retur | STRING |
| HIGH | src/oumi/launcher/clouds/local_cloud.py | 0 | gets the cluster with the specified name, or creates one if it doesn't exist. args: name: the name of the cluster. retur | STRING |
| HIGH | src/oumi/launcher/clouds/slurm_cloud.py | 0 | gets the cluster with the specified name, or creates one if it doesn't exist. args: name: the name of the cluster. retur | STRING |
| HIGH | src/oumi/launcher/clouds/frontier_cloud.py | 0 | gets the cluster with the specified name, or creates one if it doesn't exist. args: name: the name of the cluster. retur | STRING |
| HIGH | src/oumi/launcher/clouds/perlmutter_cloud.py | 0 | gets the cluster with the specified name, or creates one if it doesn't exist. args: name: the name of the cluster. retur | STRING |
| HIGH | src/oumi/launcher/clouds/polaris_cloud.py | 0 | gets the cluster with the specified name, or creates one if it doesn't exist. args: name: the name of the cluster. retur | STRING |
| HIGH | docs/user_guides/launch/custom_cluster.md | 0 | gets the cluster with the specified name, or none if not found. | STRING |
| HIGH | src/oumi/launcher/clouds/slurm_cloud.py | 0 | gets the cluster with the specified name, or none if not found. | STRING |
| HIGH | src/oumi/launcher/clouds/local_cloud.py | 0 | gets the cluster with the specified name, or none if not found. | STRING |
| HIGH | src/oumi/launcher/clouds/frontier_cloud.py | 0 | gets the cluster with the specified name, or none if not found. | STRING |
| HIGH | src/oumi/launcher/clouds/perlmutter_cloud.py | 0 | gets the cluster with the specified name, or none if not found. | STRING |
| HIGH | src/oumi/launcher/clouds/polaris_cloud.py | 0 | gets the cluster with the specified name, or none if not found. | STRING |
| HIGH | src/oumi/launcher/clouds/sky_cloud.py | 0 | gets the cluster with the specified name, or none if not found. | STRING |
| HIGH | src/oumi/core/launcher/base_cloud.py | 0 | gets the cluster with the specified name, or none if not found. | STRING |
| HIGH | src/oumi/launcher/clients/slurm_client.py | 0 | gets a stream that tails the logs of the target job. args: job_id: the id of the job to tail the logs of. cluster_name: | STRING |
| HIGH | src/oumi/launcher/clusters/sky_cluster.py | 0 | gets a stream that tails the logs of the target job. args: job_id: the id of the job to tail the logs of. cluster_name: | STRING |
| HIGH | src/oumi/launcher/clusters/local_cluster.py | 0 | gets a stream that tails the logs of the target job. args: job_id: the id of the job to tail the logs of. cluster_name: | STRING |
| HIGH | src/oumi/launcher/clusters/polaris_cluster.py | 0 | gets a stream that tails the logs of the target job. args: job_id: the id of the job to tail the logs of. cluster_name: | STRING |
| HIGH | src/oumi/launcher/clusters/frontier_cluster.py | 0 | gets a stream that tails the logs of the target job. args: job_id: the id of the job to tail the logs of. cluster_name: | STRING |
| HIGH | src/oumi/launcher/clusters/perlmutter_cluster.py | 0 | gets a stream that tails the logs of the target job. args: job_id: the id of the job to tail the logs of. cluster_name: | STRING |
| HIGH | src/oumi/launcher/clusters/slurm_cluster.py | 0 | formats the provided date as a string. args: date: the date to format. returns: the formatted date. | STRING |
| HIGH | src/oumi/launcher/clusters/polaris_cluster.py | 0 | formats the provided date as a string. args: date: the date to format. returns: the formatted date. | STRING |
| HIGH | src/oumi/launcher/clusters/frontier_cluster.py | 0 | formats the provided date as a string. args: date: the date to format. returns: the formatted date. | STRING |
| HIGH | src/oumi/launcher/clusters/perlmutter_cluster.py | 0 | formats the provided date as a string. args: date: the date to format. returns: the formatted date. | STRING |
| HIGH | src/oumi/launcher/clusters/slurm_cluster.py | 0 | finds the last pbs instruction line in the script. args: script: the lines of the script. returns: the index of the last | STRING |
| HIGH | src/oumi/launcher/clusters/frontier_cluster.py | 0 | finds the last pbs instruction line in the script. args: script: the lines of the script. returns: the index of the last | STRING |
| HIGH | src/oumi/launcher/clusters/perlmutter_cluster.py | 0 | finds the last pbs instruction line in the script. args: script: the lines of the script. returns: the index of the last | STRING |
| HIGH | src/oumi/launcher/clusters/polaris_cluster.py | 0 | finds the last pbs instruction line in the script. args: script: the lines of the script. returns: the index of the last | STRING |
| HIGH | src/oumi/launcher/clusters/slurm_cluster.py | 0 | creates a job script for the specified job. args: job: the job to create a script for. returns: the script as a string. | STRING |
| HIGH | src/oumi/launcher/clusters/polaris_cluster.py | 0 | creates a job script for the specified job. args: job: the job to create a script for. returns: the script as a string. | STRING |
| HIGH | src/oumi/launcher/clusters/frontier_cluster.py | 0 | creates a job script for the specified job. args: job: the job to create a script for. returns: the script as a string. | STRING |
| HIGH | src/oumi/launcher/clusters/perlmutter_cluster.py | 0 | creates a job script for the specified job. args: job: the job to create a script for. returns: the script as a string. | STRING |
| HIGH | src/oumi/launcher/clusters/slurm_cluster.py | 0 | validates the provided job configuration. args: job: the job to validate. | STRING |
| HIGH | src/oumi/launcher/clusters/local_cluster.py | 0 | validates the provided job configuration. args: job: the job to validate. | STRING |
| HIGH | src/oumi/launcher/clusters/polaris_cluster.py | 0 | validates the provided job configuration. args: job: the job to validate. | STRING |
| HIGH | src/oumi/launcher/clusters/frontier_cluster.py | 0 | validates the provided job configuration. args: job: the job to validate. | STRING |
| HIGH | src/oumi/launcher/clusters/perlmutter_cluster.py | 0 | validates the provided job configuration. args: job: the job to validate. | STRING |
| HIGH | src/oumi/launcher/clusters/slurm_cluster.py | 0 | parses the cluster name into queue and user components. args: name: the name of the cluster. returns: _clusterinfo: the | STRING |
| HIGH | src/oumi/launcher/clouds/frontier_cloud.py | 0 | parses the cluster name into queue and user components. args: name: the name of the cluster. returns: _clusterinfo: the | STRING |
| HIGH | src/oumi/launcher/clouds/perlmutter_cloud.py | 0 | parses the cluster name into queue and user components. args: name: the name of the cluster. returns: _clusterinfo: the | STRING |
| HIGH | src/oumi/launcher/clouds/polaris_cloud.py | 0 | parses the cluster name into queue and user components. args: name: the name of the cluster. returns: _clusterinfo: the | STRING |
| HIGH | src/oumi/launcher/clusters/polaris_cluster.py | 0 | runs the specified job on this cluster. for perlmutter this method consists of 5 parts: 1. copy the working directory to | STRING |
| HIGH | src/oumi/launcher/clusters/frontier_cluster.py | 0 | runs the specified job on this cluster. for perlmutter this method consists of 5 parts: 1. copy the working directory to | STRING |
| HIGH | src/oumi/launcher/clusters/perlmutter_cluster.py | 0 | runs the specified job on this cluster. for perlmutter this method consists of 5 parts: 1. copy the working directory to | STRING |
| HIGH | src/oumi/core/callbacks/hf_mfu_callback.py | 0 | event called at the end of each train step. note that this will be called after all gradient accumulation substeps. | STRING |
| HIGH | src/oumi/core/callbacks/telemetry_callback.py | 0 | event called at the end of each train step. note that this will be called after all gradient accumulation substeps. | STRING |
| HIGH | src/oumi/core/callbacks/mfu_callback.py | 0 | event called at the end of each train step. note that this will be called after all gradient accumulation substeps. | STRING |
| HIGH | src/oumi/core/callbacks/profiler_step_callback.py | 0 | event called at the end of each train step. note that this will be called after all gradient accumulation substeps. | STRING |
| HIGH | src/oumi/core/datasets/base_grpo_dataset.py | 0 | validate and transform the sample into python `dict`. | STRING |
| HIGH | src/oumi/datasets/grpo/gsm8k.py | 0 | validate and transform the sample into python `dict`. | STRING |
| HIGH | src/oumi/datasets/grpo/letter_count.py | 0 | validate and transform the sample into python `dict`. | STRING |
| HIGH | src/oumi/datasets/grpo/countdown.py | 0 | validate and transform the sample into python `dict`. | STRING |
| HIGH | src/oumi/core/synthesis/attribute_synthesizer.py | 0 | mapping of failed index to structured failure info. | STRING |
| HIGH | src/oumi/core/inference/base_inference_engine.py | 0 | mapping of failed index to structured failure info. | STRING |
| 48 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tests/unit/launcher/test_launcher.py | 1347 | CODE | |
| LOW | tests/unit/launcher/clients/test_modal_client.py | 12 | CODE | |
| LOW | tests/unit/launcher/clusters/test_modal_cluster.py | 8 | CODE | |
| LOW | tests/unit/launcher/clouds/test_modal_cloud.py | 8 | CODE | |
| LOW | …sts/unit/core/datasets/test_vision_language_dataset.py | 9 | CODE | |
| LOW | tests/unit/core/trainers/test_verl_grpo_trainer.py | 36 | CODE | |
| LOW | tests/unit/core/configs/test_synthesis_config.py | 17 | CODE | |
| LOW | …ts/unit/core/configs/params/test_environment_params.py | 20 | CODE | |
| LOW | tests/unit/deploy/__init__.py | 17 | CODE | |
| LOW | tests/unit/deploy/__init__.py | 18 | CODE | |
| LOW | tests/unit/deploy/__init__.py | 19 | CODE | |
| LOW | tests/unit/deploy/__init__.py | 20 | CODE | |
| LOW | tests/unit/deploy/__init__.py | 21 | CODE | |
| LOW | tests/unit/deploy/__init__.py | 22 | CODE | |
| LOW | tests/unit/deploy/__init__.py | 23 | CODE | |
| LOW | tests/unit/deploy/__init__.py | 24 | CODE | |
| LOW | tests/unit/deploy/__init__.py | 25 | CODE | |
| LOW | tests/unit/environments/test_database_session.py | 15 | CODE | |
| LOW | tests/unit/environments/test_executor_resolution.py | 17 | CODE | |
| LOW | tests/unit/environments/test_executable_environment.py | 17 | CODE | |
| LOW | tests/unit/environments/test_executable_tool.py | 17 | CODE | |
| LOW | …t/environments/test_database_executable_environment.py | 17 | CODE | |
| LOW | …nit/datasets/test_vision_language_jsonlines_dataset.py | 7 | CODE | |
| LOW | tests/unit/utils/test_io_utils.py | 15 | CODE | |
| LOW | tests/unit/cli/test_cli_speed_regression.py | 8 | CODE | |
| LOW | tests/unit/examples/test_ehr_database_executors.py | 17 | CODE | |
| LOW | tests/unit/examples/test_ehr_database_executors.py | 23 | CODE | |
| LOW | tests/unit/examples/test_ehr_executors.py | 17 | CODE | |
| LOW | tests/unit/inference/test_inference_engine_init.py | 37 | CODE | |
| LOW | tests/unit/inference/test_inference_engine_init.py | 44 | CODE | |
| LOW | tests/unit/inference/test_inference_engine_init.py | 51 | CODE | |
| LOW | tests/unit/inference/test_remote_inference_engine.py | 4848 | CODE | |
| LOW | …igs/examples/synthesis/ehr_stateful_agent/executors.py | 32 | CODE | |
| LOW | …igs/examples/synthesis/ehr_database_agent/executors.py | 26 | CODE | |
| LOW | src/oumi/__init__.py | 84 | CODE | |
| LOW | src/oumi/infer.py | 30 | CODE | |
| LOW | src/oumi/exceptions.py | 24 | CODE | |
| LOW | src/oumi/quantize/__init__.py | 23 | CODE | |
| LOW | src/oumi/quantize/__init__.py | 24 | CODE | |
| LOW | src/oumi/quantize/__init__.py | 25 | CODE | |
| LOW | src/oumi/launcher/__init__.py | 34 | CODE | |
| LOW | src/oumi/launcher/__init__.py | 35 | CODE | |
| LOW | src/oumi/launcher/__init__.py | 35 | CODE | |
| LOW | src/oumi/launcher/__init__.py | 35 | CODE | |
| LOW | src/oumi/launcher/__init__.py | 36 | CODE | |
| LOW | src/oumi/launcher/__init__.py | 36 | CODE | |
| LOW | src/oumi/launcher/__init__.py | 36 | CODE | |
| LOW | src/oumi/launcher/__init__.py | 36 | CODE | |
| LOW | src/oumi/launcher/__init__.py | 36 | CODE | |
| LOW | src/oumi/launcher/__init__.py | 36 | CODE | |
| LOW | src/oumi/launcher/__init__.py | 36 | CODE | |
| LOW | src/oumi/launcher/__init__.py | 36 | CODE | |
| LOW | src/oumi/launcher/__init__.py | 36 | CODE | |
| LOW | src/oumi/launcher/clients/modal_client.py | 32 | CODE | |
| LOW | src/oumi/launcher/clusters/modal_cluster.py | 31 | CODE | |
| LOW | src/oumi/launcher/clouds/modal_cloud.py | 25 | CODE | |
| LOW | src/oumi/launcher/clouds/__init__.py | 35 | CODE | |
| LOW | src/oumi/launcher/clouds/__init__.py | 36 | CODE | |
| LOW | src/oumi/launcher/clouds/__init__.py | 37 | CODE | |
| LOW | src/oumi/launcher/clouds/__init__.py | 38 | CODE | |
| 480 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tests/conftest.py | 79 | except Exception: | CODE |
| LOW | tests/conftest.py | 92 | except Exception: | CODE |
| LOW | tests/unit/core/analyze/test_dataset_analyzer.py | 645 | except Exception as e: | CODE |
| LOW | tests/unit/mcp/test_job_recovery_and_control.py | 73 | except Exception: | CODE |
| LOW | tests/unit/inference/test_base_inference_engine.py | 765 | except Exception as e: # noqa: BLE001 - surface any thread failure | CODE |
| LOW | tests/unit/inference/test_base_inference_engine.py | 807 | except Exception as e: # noqa: BLE001 - surface any thread failure | CODE |
| LOW | tests/unit/inference/test_adaptive_semaphore.py | 503 | except Exception as e: | CODE |
| LOW | tests/e2e/test_sambanova_inference.py | 59 | except Exception as e: | CODE |
| MEDIUM | tests/e2e/test_sambanova_inference.py | 60 | print("An error occurred during inference:", str(e)) | CODE |
| LOW | tests/e2e/test_notebooks.py | 53 | except Exception as e: | CODE |
| LOW | tests/e2e/test_train_e2e.py | 208 | except Exception as e: | CODE |
| LOW | tests/e2e/test_train_e2e.py | 368 | except Exception as e: | CODE |
| LOW | tests/e2e/test_tune_e2e.py | 65 | except Exception as e: | CODE |
| LOW | tests/e2e/test_tune_e2e.py | 153 | except Exception as e: | CODE |
| LOW | tests/e2e/test_eval_e2e.py | 75 | except Exception as e: | CODE |
| LOW | tests/e2e/test_eval_e2e.py | 162 | except Exception as e: | CODE |
| LOW | docs/_summarize_module.py | 167 | except Exception: | CODE |
| LOW | configs/projects/dcvlr/run_image_synthesis.py | 55 | except Exception as e: | CODE |
| MEDIUM | configs/projects/dcvlr/run_image_synthesis.py | 56 | print(f"Error processing image: {str(e)}") | CODE |
| LOW | configs/projects/dcvlr/run_image_synthesis.py | 161 | except Exception as e: | CODE |
| MEDIUM | configs/projects/dcvlr/run_image_synthesis.py | 260 | print(f"Error: Dataset JSON not found at {json_path}") | CODE |
| MEDIUM | configs/projects/dcvlr/run_image_synthesis.py | 345 | print("Error: --repo-id is required when using --push-to-hub") | CODE |
| LOW | configs/projects/dcvlr/run_image_synthesis.py | 363 | except Exception as e: | CODE |
| MEDIUM | configs/projects/dcvlr/run_image_synthesis.py | 364 | print(f"Error pushing to hub: {str(e)}") | CODE |
| LOW | configs/projects/dcvlr/synthesize_images_vllm.py | 113 | except Exception as e: | CODE |
| LOW | configs/projects/dcvlr/synthesize_images_vllm.py | 365 | except Exception as e: | CODE |
| LOW | configs/projects/dcvlr/synthesize_images_vllm.py | 516 | except Exception as e: | CODE |
| LOW | configs/projects/dcvlr/synthesize_images_vllm.py | 526 | except Exception as e: | CODE |
| LOW | configs/projects/dcvlr/synthesize_images_vllm.py | 625 | except Exception as e2: | CODE |
| LOW | configs/projects/dcvlr/synthesize_images_vllm.py | 637 | except Exception as e: | CODE |
| LOW | configs/projects/dcvlr/synthesize_images_vllm.py | 681 | except Exception as e: | CODE |
| LOW | configs/projects/dcvlr/synthesize_images_vllm.py | 794 | except Exception as e: | CODE |
| LOW | scripts/demo.py | 614 | except Exception as e: | CODE |
| LOW | scripts/polaris/jobs/python/vllm_parallel_inference.py | 138 | except Exception as e: | CODE |
| LOW | src/oumi/quantize/base.py | 154 | except Exception: | CODE |
| LOW | src/oumi/launcher/clients/polaris_client.py | 273 | except Exception: | CODE |
| LOW | src/oumi/launcher/clients/slurm_client.py | 360 | except Exception as e: | CODE |
| LOW | src/oumi/launcher/clients/slurm_client.py | 437 | except Exception as e: | CODE |
| LOW⚡ | src/oumi/launcher/clients/slurm_client.py | 461 | except Exception as e: | CODE |
| LOW⚡ | src/oumi/launcher/clients/slurm_client.py | 467 | except Exception: | CODE |
| LOW⚡ | src/oumi/launcher/clients/slurm_client.py | 474 | except Exception as e: | CODE |
| LOW | src/oumi/launcher/clients/slurm_client.py | 615 | except Exception: | CODE |
| LOW | src/oumi/launcher/clients/sky_client.py | 249 | except Exception: | CODE |
| LOW | src/oumi/launcher/clients/sky_client.py | 289 | except Exception: | CODE |
| LOW | src/oumi/launcher/clients/sky_client.py | 385 | except Exception: | CODE |
| LOW | src/oumi/launcher/clients/modal_client.py | 253 | except Exception as e: # noqa: BLE001 | CODE |
| LOW | src/oumi/launcher/clients/modal_client.py | 295 | except Exception as e: # noqa: BLE001 | CODE |
| LOW | src/oumi/launcher/clients/modal_client.py | 309 | except Exception as e: # noqa: BLE001 | CODE |
| LOW | src/oumi/launcher/clients/modal_client.py | 344 | except Exception as e: # noqa: BLE001 | CODE |
| LOW | src/oumi/launcher/clients/modal_client.py | 364 | except Exception as e: # noqa: BLE001 | CODE |
| LOW | src/oumi/core/tokenizers/utils.py | 116 | except Exception: | CODE |
| LOW | src/oumi/core/tokenizers/utils.py | 193 | except Exception as e: | CODE |
| LOW | src/oumi/core/models/base_model.py | 235 | except Exception: | CODE |
| LOW | src/oumi/core/synthesis/conversation_synthesizer.py | 248 | except Exception: | CODE |
| LOW | src/oumi/core/synthesis/conversation_synthesizer.py | 257 | except Exception as exc: | CODE |
| LOW | src/oumi/core/synthesis/document_ingestion.py | 198 | except Exception: | CODE |
| LOW | src/oumi/core/synthesis/attribute_synthesizer.py | 200 | except Exception as e: | CODE |
| LOW | src/oumi/core/synthesis/attribute_synthesizer.py | 366 | except Exception as e: | CODE |
| LOW | src/oumi/core/trainers/oumi_trainer.py | 704 | except Exception as e: | CODE |
| LOW | src/oumi/core/trainers/oumi_trainer.py | 713 | except Exception as e: | CODE |
| 161 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | tests/conftest.py | 92 | except Exception: | CODE |
| HIGH | tests/unit/mcp/test_job_recovery_and_control.py | 73 | except Exception: | CODE |
| HIGH | tests/unit/inference/test_adaptive_semaphore.py | 373 | except asyncio.CancelledError: | CODE |
| HIGH | tests/unit/inference/test_adaptive_semaphore.py | 575 | except asyncio.CancelledError: | CODE |
| HIGH | tests/unit/inference/test_adaptive_semaphore.py | 600 | except ValueError: | CODE |
| HIGH | docs/_summarize_module.py | 167 | except Exception: | CODE |
| HIGH | scripts/demo.py | 39 | except (ImportError, AttributeError): | CODE |
| HIGH | src/oumi/launcher/clients/slurm_client.py | 358 | except ProcessLookupError: | CODE |
| HIGH | src/oumi/launcher/clients/slurm_client.py | 459 | except ProcessLookupError: | CODE |
| HIGH | src/oumi/launcher/clients/slurm_client.py | 472 | except ProcessLookupError: | CODE |
| HIGH | src/oumi/launcher/clients/sky_client.py | 385 | except Exception: | CODE |
| HIGH | src/oumi/core/synthesis/synthesis_pipeline.py | 170 | except (KeyError, IndexError, ValueError): | CODE |
| HIGH | src/oumi/core/configs/internal/supported_models.py | 687 | except (json.JSONDecodeError, OSError): | CODE |
| HIGH | src/oumi/core/configs/params/tuning_params.py | 373 | except Exception: | CODE |
| HIGH | src/oumi/environments/utils.py | 91 | except Exception: | CODE |
| HIGH | src/oumi/environments/database_session.py | 83 | except OSError: | CODE |
| HIGH | src/oumi/utils/str_utils.py | 163 | except json.JSONDecodeError: | CODE |
| HIGH | src/oumi/utils/str_utils.py | 181 | except json.JSONDecodeError: | CODE |
| HIGH | src/oumi/utils/http.py | 97 | except (ValueError, TypeError): | CODE |
| HIGH | src/oumi/mcp/job_launcher.py | 138 | except Exception: | CODE |
| HIGH | src/oumi/mcp/job_launcher.py | 307 | except ProcessLookupError: | CODE |
| HIGH | src/oumi/mcp/job_launcher.py | 313 | except ProcessLookupError: | CODE |
| HIGH | src/oumi/mcp/job_logs.py | 189 | except Exception: | CODE |
| HIGH | src/oumi/mcp/job_logs.py | 266 | except Exception: | CODE |
| HIGH | src/oumi/mcp/job_logs.py | 280 | except Exception: | CODE |
| HIGH | src/oumi/mcp/job_logs.py | 194 | except Exception: | CODE |
| HIGH | src/oumi/mcp/preflight_service.py | 662 | except Exception: | CODE |
| HIGH | src/oumi/mcp/preflight_service.py | 638 | except ValueError: | CODE |
| HIGH | src/oumi/mcp/preflight_service.py | 302 | except Exception: | CODE |
| HIGH | src/oumi/mcp/preflight_service.py | 325 | except Exception: | CODE |
| HIGH | src/oumi/mcp/docs_service.py | 464 | except Exception: | CODE |
| HIGH | src/oumi/mcp/config_service.py | 97 | except (OSError, FileNotFoundError): | CODE |
| HIGH | src/oumi/mcp/config_service.py | 90 | except (OSError, FileNotFoundError): | CODE |
| HIGH | src/oumi/mcp/job_runtime.py | 49 | except Exception: | CODE |
| HIGH | src/oumi/cli/completions.py | 283 | except (PermissionError, OSError): | CODE |
| HIGH | src/oumi/inference/adaptive_semaphore.py | 87 | except ValueError: | CODE |
| HIGH | src/oumi/inference/together_inference_engine.py | 199 | except OSError: | CODE |
| HIGH | src/oumi/analyze/discovery.py | 76 | except Exception: | CODE |
| HIGH | src/oumi/telemetry/manager.py | 148 | except OSError: | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | tests/unit/core/test_length_analyzer.py | 240 | # Create a conversation with multiple messages | COMMENT |
| MEDIUM | tests/unit/core/test_length_analyzer.py | 286 | # Create a conversation with multiple messages | COMMENT |
| MEDIUM | tests/unit/core/test_length_analyzer.py | 329 | # Create a conversation with multiple messages | COMMENT |
| MEDIUM | tests/unit/core/datasets/test_base_map_dataset.py | 69 | # Create a test dataset class | COMMENT |
| MEDIUM | tests/unit/core/synthesis/test_synthesis_pipeline.py | 553 | # Create a config with a bad output path by bypassing validation | COMMENT |
| MEDIUM | tests/unit/core/synthesis/test_synthesis_pipeline.py | 702 | # Create a minimal config to avoid model loading | COMMENT |
| MEDIUM | tests/unit/core/analyze/test_dataset_analyzer.py | 123 | # Create a subclass that remembers the analyzer_id | COMMENT |
| MEDIUM | tests/unit/core/analyze/test_dataset_analyzer.py | 260 | # Create a real TextSftJsonLinesDataset from the JSONL file | COMMENT |
| MEDIUM | tests/unit/core/analyze/test_dataset_analyzer.py | 295 | # Create a real dataset from test data | COMMENT |
| MEDIUM | tests/unit/core/analyze/test_dataset_analyzer.py | 812 | # Create a config with sample_count=1 to only analyze 1 conversation | COMMENT |
| MEDIUM | tests/unit/utils/test_analysis_utils.py | 357 | # Create a mock tokenizer with required attributes | COMMENT |
| MEDIUM | tests/unit/utils/test_device_utils.py | 234 | # Create a mock NVMLError_NotSupported exception | COMMENT |
| MEDIUM⚡ | tests/unit/utils/test_io_utils.py | 235 | # Create a dummy file path | COMMENT |
| MEDIUM⚡ | tests/unit/utils/test_io_utils.py | 237 | file_path.touch() # Create an empty file | CODE |
| MEDIUM | tests/unit/cli/test_cli_synth.py | 352 | # Create a file that would collide with the auto-generated name | COMMENT |
| MEDIUM | tests/unit/cli/test_cli_synth.py | 360 | # Create the file that would be auto-generated | COMMENT |
| MEDIUM | tests/unit/cli/test_cli_distributed_run.py | 119 | # Define the redundant OUMI_ variables to activate consistency checks. | COMMENT |
| MEDIUM | tests/unit/cli/test_cli_distributed_run.py | 185 | # Define the redundant OUMI_ variables to activate consistency checks. | COMMENT |
| MEDIUM | tests/unit/cli/test_cli_distributed_run.py | 249 | # Define the redundant OUMI_ variables to activate consistency checks. | COMMENT |
| MEDIUM | tests/unit/cli/test_cli_distributed_run.py | 316 | # Define the redundant OUMI_ variables to activate consistency checks. | COMMENT |
| MEDIUM | tests/unit/cli/test_cli_utils.py | 367 | # Create a mock console | COMMENT |
| MEDIUM⚡ | tests/unit/cli/test_cli_completions.py | 176 | # Create a test yaml file | COMMENT |
| MEDIUM⚡ | tests/unit/cli/test_cli_completions.py | 180 | # Create a subdirectory | COMMENT |
| MEDIUM | tests/unit/inference/test_generation_params.py | 93 | # Create a proper BatchEncoding for the tokenizer to return | COMMENT |
| MEDIUM | tests/unit/inference/test_generation_params.py | 103 | # Create a mock tensor that behaves like model.generate() output | COMMENT |
| MEDIUM⚡ | tests/unit/analyze/test_testing_engine.py | 341 | # Create a non-BaseModel, non-dict object | COMMENT |
| MEDIUM | …/integration/models/test_integration_cnn_classifier.py | 274 | # Create a model. | COMMENT |
| MEDIUM | tests/e2e/test_sambanova_inference.py | 81 | # Create the conversation | COMMENT |
| MEDIUM | tests/e2e/deps/test_circular_deps.py | 70 | # Create a new process with a clean dependency tree. | COMMENT |
| MEDIUM | configs/projects/dcvlr/run_image_synthesis.py | 32 | # Create a white background | COMMENT |
| MEDIUM | configs/projects/dcvlr/run_image_synthesis.py | 121 | # Create a temporary Python file | COMMENT |
| MEDIUM | configs/projects/dcvlr/run_image_synthesis.py | 127 | # Create a temporary directory for execution | COMMENT |
| MEDIUM | …gs/examples/synthesis/instruction_following_synth.yaml | 53 | # Define the domain and complexity attributes | COMMENT |
| MEDIUM | scripts/perlmutter/launcher.sh | 97 | # Create a logs directory for the user if it doesn't exist. | COMMENT |
| MEDIUM | scripts/polaris/launcher.sh | 106 | # Create a logs directory for the user if it doesn't exist. | COMMENT |
| MEDIUM | scripts/polaris/jobs/vllm_worker.sh | 53 | # Create a dir to copy temp files into. | COMMENT |
| MEDIUM | scripts/examples/batch_inference/bulk_infer.py | 6 | ######################## Creating a conversation JSON ######################## | COMMENT |
| MEDIUM | scripts/frontier/launcher.sh | 116 | # Create a logs directory for the user if it doesn't exist. | COMMENT |
| MEDIUM | src/oumi/launcher/clusters/slurm_cluster.py | 339 | # Create the job script by merging envs, setup, and run commands. | COMMENT |
| MEDIUM | src/oumi/launcher/clusters/polaris_cluster.py | 270 | # Create the job script by merging envs, setup, and run commands. | COMMENT |
| MEDIUM | src/oumi/launcher/clusters/frontier_cluster.py | 299 | # Create the job script by merging envs, setup, and run commands. | COMMENT |
| MEDIUM | src/oumi/launcher/clusters/perlmutter_cluster.py | 293 | # Create the job script by merging envs, setup, and run commands. | COMMENT |
| MEDIUM | src/oumi/core/synthesis/attribute_transformation.py | 144 | # Create a TransformationStrategy for the metadata dict transformation | COMMENT |
| MEDIUM | src/oumi/core/evaluation/evaluator.py | 82 | # Create a copy of the evaluation config, without tasks, so that there is no | COMMENT |
| MEDIUM | src/oumi/core/evaluation/utils/save_utils.py | 105 | # Create the output directory: `<base_output_dir> / <backend_name>_<time>`. | COMMENT |
| MEDIUM | src/oumi/core/analyze/dataset_analyzer.py | 594 | # Create a new dataset with only the filtered conversations | COMMENT |
| MEDIUM | src/oumi/datasets/vision_language/huggingface.py | 196 | # Create the image content item | COMMENT |
| MEDIUM | src/oumi/datasets/vision_language/huggingface.py | 199 | # Create the user message with image and text | COMMENT |
| MEDIUM | src/oumi/datasets/evaluation/utils.py | 68 | # Create a dictionary with the instruction, output, metadata. | COMMENT |
| MEDIUM | src/oumi/utils/cache_utils.py | 134 | # Create a wrapper that converts args to hashable JSON strings | COMMENT |
| MEDIUM⚡ | src/oumi/utils/analysis_utils.py | 454 | # Create a DataFrame from raw data for analysis | COMMENT |
| MEDIUM | src/oumi/utils/analysis_utils.py | 132 | # Create a default tokenizer if none provided | COMMENT |
| MEDIUM | src/oumi/utils/analysis_utils.py | 549 | # Create a DataFrame from raw data for analysis | COMMENT |
| MEDIUM | src/oumi/judges/base_judge.py | 751 | # Create a conversation for each judgment prompt | COMMENT |
| MEDIUM | src/oumi/judges/__init__.py | 15 | """This module provides access to various judge configurations for the Oumi project. | STRING |
| MEDIUM | src/oumi/cli/main.py | 401 | # Create the CLI app when building the docs to auto-generate the CLI reference. | COMMENT |
| MEDIUM | src/oumi/inference/native_text_inference_engine.py | 290 | # Create a GenerationConfig object with the new parameters | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tests/markers.py | 8 | CODE | |
| LOW | tests/unit/launcher/clients/test_sky_client.py | 100 | CODE | |
| LOW | tests/unit/launcher/clients/test_sky_client.py | 147 | CODE | |
| LOW | tests/unit/launcher/clients/test_sky_client.py | 209 | CODE | |
| LOW | tests/unit/launcher/clients/test_sky_client.py | 269 | CODE | |
| LOW | tests/unit/launcher/clients/test_sky_client.py | 288 | CODE | |
| LOW | tests/unit/core/test_distributed.py | 112 | CODE | |
| LOW | tests/unit/core/configs/test_parse_configs.py | 29 | CODE | |
| LOW | tests/unit/datasets/test_chat_templates.py | 226 | CODE | |
| LOW | tests/unit/utils/test_conversation_utils.py | 424 | CODE | |
| LOW | tests/unit/cli/test_cli_alias.py | 39 | CODE | |
| LOW | tests/unit/cli/test_cli_alias.py | 70 | CODE | |
| LOW | tests/unit/inference/test_sglang_inference_engine.py | 91 | CODE | |
| LOW | tests/unit/inference/test_inference_engine_init.py | 92 | CODE | |
| LOW | tests/unit/inference/test_generation_params.py | 76 | CODE | |
| LOW | tests/unit/inference/test_generation_params.py | 421 | CODE | |
| LOW | tests/unit/inference/test_remote_inference_engine.py | 1551 | CODE | |
| LOW | tests/unit/inference/test_remote_inference_engine.py | 1693 | CODE | |
| LOW | tests/unit/inference/test_remote_inference_engine.py | 1883 | CODE | |
| LOW | tests/unit/inference/test_remote_inference_engine.py | 3651 | CODE | |
| LOW | tests/unit/inference/test_remote_inference_engine.py | 3658 | CODE | |
| LOW | …/integration/models/test_integration_cnn_classifier.py | 79 | CODE | |
| LOW | tests/e2e/test_eval_e2e.py | 43 | CODE | |
| LOW | tests/e2e/deps/test_circular_deps.py | 27 | CODE | |
| LOW | docs/_summarize_module.py | 197 | CODE | |
| LOW | configs/projects/dcvlr/run_image_synthesis.py | 112 | CODE | |
| LOW | configs/projects/dcvlr/run_image_synthesis.py | 230 | CODE | |
| LOW | configs/projects/dcvlr/synthesize_images_vllm.py | 67 | CODE | |
| LOW | configs/projects/dcvlr/synthesize_images_vllm.py | 182 | CODE | |
| LOW | configs/projects/dcvlr/synthesize_images_vllm.py | 316 | CODE | |
| LOW | configs/projects/dcvlr/synthesize_images_vllm.py | 404 | CODE | |
| LOW | configs/projects/dcvlr/synthesize_images_vllm.py | 734 | CODE | |
| LOW | configs/projects/dcvlr/synthesize_images_vllm.py | 159 | CODE | |
| LOW | scripts/demo.py | 333 | CODE | |
| LOW | scripts/datasets/pretokenize/process_dataset.py | 101 | CODE | |
| LOW | src/oumi/launcher/launcher.py | 81 | CODE | |
| LOW | src/oumi/launcher/clients/slurm_client.py | 70 | CODE | |
| LOW | src/oumi/launcher/clients/sky_client.py | 32 | CODE | |
| LOW | src/oumi/launcher/clients/local_client.py | 66 | CODE | |
| LOW | src/oumi/launcher/clients/local_client.py | 121 | CODE | |
| LOW | src/oumi/launcher/clusters/sky_cluster.py | 40 | CODE | |
| LOW | src/oumi/launcher/clusters/frontier_cluster.py | 58 | CODE | |
| LOW | src/oumi/launcher/clusters/perlmutter_cluster.py | 58 | CODE | |
| LOW | src/oumi/launcher/clouds/sky_cloud.py | 113 | CODE | |
| LOW | src/oumi/core/types/conversation.py | 315 | CODE | |
| LOW | …tors/vision_language_conversation_feature_generator.py | 299 | CODE | |
| LOW | src/oumi/core/callbacks/telemetry_callback.py | 163 | CODE | |
| LOW | src/oumi/core/datasets/packed_sft_dataset.py | 168 | CODE | |
| LOW | src/oumi/core/datasets/base_map_dataset.py | 189 | CODE | |
| LOW | src/oumi/core/datasets/base_map_dataset.py | 274 | CODE | |
| LOW | src/oumi/core/datasets/base_map_dataset.py | 457 | CODE | |
| LOW | src/oumi/core/synthesis/conversation_synthesizer.py | 215 | CODE | |
| LOW | src/oumi/core/synthesis/conversation_synthesizer.py | 525 | CODE | |
| LOW | src/oumi/core/synthesis/dataset_ingestion.py | 193 | CODE | |
| LOW | src/oumi/core/synthesis/document_ingestion.py | 107 | CODE | |
| LOW | src/oumi/core/synthesis/attribute_transformation.py | 69 | CODE | |
| LOW | src/oumi/core/synthesis/synthesis_pipeline.py | 58 | CODE | |
| LOW | src/oumi/core/synthesis/dataset_planner.py | 225 | CODE | |
| LOW | src/oumi/core/synthesis/dataset_planner.py | 303 | CODE | |
| LOW | src/oumi/core/tuners/optuna_tuner.py | 55 | CODE | |
| 114 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH⚡ | tests/unit/launcher/clients/test_modal_client.py | 211 | setup="sudo apt-get update && sudo apt-get install -y zip", | CODE |
| HIGH⚡ | tests/unit/launcher/clients/test_modal_client.py | 219 | assert "apt-get update && apt-get install -y zip" in script | CODE |
| HIGH⚡ | tests/unit/launcher/clients/test_modal_client.py | 299 | src = "sudo apt-get update && sudo apt-get install -y zip\nsudo dpkg -i bar.deb" | CODE |
| HIGH⚡ | tests/unit/launcher/clients/test_modal_client.py | 301 | "apt-get update && apt-get install -y zip\ndpkg -i bar.deb" | CODE |
| HIGH | tests/unit/launcher/clients/test_slurm_client.py | 524 | b" JobState=COMPLETED Reason=None Dependency=(null)\n" | CODE |
| HIGH | tests/unit/datasets/test_huggingface_vision_dataset.py | 193 | assert result.equals(pd.DataFrame({"test": [1, 2, 3]})) | CODE |
| HIGH | tests/unit/datasets/test_huggingface_vision_dataset.py | 226 | assert result.equals(pd.DataFrame({"test": [1, 2, 3]})) | CODE |
| HIGH | tests/unit/utils/test_str_utils.py | 148 | pip install -e uv && uv pip -q install "oumi[gpu,dev]" vllm # comment | STRING |
| HIGH | tests/unit/utils/test_str_utils.py | 153 | pip install -e uv && uv pip -q install -e '.[gpu,dev]' vllm # comment | STRING |
| HIGH | tests/unit/inference/test_remote_inference_engine.py | 2656 | # OpenAI-style error: top-level "error" is null, detail in response.body.error | COMMENT |
| HIGH⚡ | tests/unit/analyze/test_quality_analyzer.py | 402 | Message(role=Role.USER, content="values: NaN, null, undefined"), | CODE |
| HIGH | scripts/demo.py | 688 | "run": "oumi infer -c infer_config.yaml && cat responses.jsonl", | CODE |
| HIGH | scripts/demo.py | 692 | deploy_config["setup"] = "pip install uv && uv pip install --system oumi[gpu]" | CODE |
| HIGH | src/oumi/launcher/clients/slurm_client.py | 125 | if not value or value in ("Unknown", "N/A", "(null)"): | CODE |
| HIGH | src/oumi/environments/lookup_environment.py | 156 | ``output`` may be any JSON value (scalar, list, object, or null). | STRING |
| HIGH⚡ | src/oumi/utils/str_utils.py | 201 | `pip install uv && uv pip -q install oumi[gpu,dev] vllm` | STRING |
| HIGH⚡ | src/oumi/utils/str_utils.py | 203 | `pip install uv && uv pip -q install -e '.[gpu,dev]' vllm` | STRING |
| HIGH | src/oumi/mcp/prompts/cloud_launch.py | 70 | pip install uv && uv pip install --system 'oumi[gpu]' | CODE |
| HIGH | src/oumi/mcp/prompts/cloud_launch.py | 195 | pip install uv && uv pip install --system 'oumi[gpu]' | CODE |
| HIGH⚡ | src/oumi/mcp/prompts/cloud_launch.py | 221 | pip install uv && uv pip install --system 'oumi[gpu]' | CODE |
| HIGH⚡ | src/oumi/mcp/prompts/cloud_launch.py | 228 | pip install uv && uv pip install --system 'oumi[gpu]' | CODE |
| HIGH⚡ | src/oumi/mcp/prompts/cloud_launch.py | 235 | pip install uv && uv pip install --system 'oumi[gpu,evaluation]' | CODE |
| HIGH | src/oumi/analyze/analyzers/quality.py | 98 | - Values serialized as strings (NaN, null, None, undefined) | STRING |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tests/conftest.py | 72 | # Check if test function uses GPU device mapping | COMMENT |
| LOW⚡ | tests/unit/quantize/test_awq_quantizer.py | 94 | # Set _awq to None to simulate missing AWQ | COMMENT |
| LOW⚡ | tests/unit/quantize/test_awq_quantizer.py | 104 | # Set _awq to a mock to simulate AWQ is installed | COMMENT |
| LOW | tests/unit/quantize/test_bnb_quantizer.py | 90 | # Set _bitsandbytes to None to simulate missing BitsAndBytes | COMMENT |
| LOW⚡ | tests/unit/core/test_registry.py | 304 | # Check if the registry contains the key, regardless of case | COMMENT |
| LOW⚡ | tests/unit/core/test_registry.py | 309 | # Check if we can retrieve the builder using different cases | COMMENT |
| LOW⚡ | tests/unit/core/test_registry.py | 314 | # Check if the registry does not contain the key, regardless of case | COMMENT |
| LOW | tests/unit/core/datasets/test_pretraining_dataset.py | 156 | # Check if we've processed all files | COMMENT |
| LOW | tests/unit/utils/test_torch_naming_heuristics.py | 166 | # Check if the guessed class name matches the expected name | COMMENT |
| LOW | tests/unit/inference/test_inference_engine_init.py | 35 | # Check if optional dependencies are available | COMMENT |
| LOW | tests/scripts/e2e_tests_job.yaml | 14 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | tests/scripts/runpod_e2e_tests_job.yaml | 15 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | tests/e2e/test_sambanova_inference.py | 87 | # Print the results | COMMENT |
| LOW | configs/projects/aya/sft/gcp_job.yaml | 20 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | configs/recipes/llama4/sft/scout_base_full/gcp_job.yaml | 25 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | …gs/recipes/llama4/sft/scout_instruct_full/gcp_job.yaml | 25 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | configs/recipes/phi3/dpo/gcp_job.yaml | 18 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | configs/recipes/vision/phi3/sft/full/oumi_gcp_job.yaml | 29 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | configs/recipes/vision/phi3/sft/full/trl_gcp_job.yaml | 29 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | configs/recipes/vision/phi3/sft/lora/gcp_job.yaml | 29 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | configs/recipes/vision/phi4/sft/full/gcp_job.yaml | 24 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | configs/recipes/vision/phi4/sft/lora/gcp_job.yaml | 24 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | …nfigs/recipes/vision/qwen2_vl_2b/sft/full/gcp_job.yaml | 25 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | …nfigs/recipes/vision/qwen2_vl_2b/sft/lora/gcp_job.yaml | 25 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | …igs/recipes/vision/qwen2_5_vl_3b/sft/full/gcp_job.yaml | 25 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | …igs/recipes/vision/qwen2_5_vl_3b/sft/lora/gcp_job.yaml | 25 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | configs/recipes/vision/llava_7b/sft/oumi_gcp_job.yaml | 28 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | configs/recipes/vision/llava_7b/sft/trl_gcp_job.yaml | 18 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | configs/recipes/vision/smolvlm/sft/full/gcp_job.yaml | 25 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | configs/recipes/vision/smolvlm/sft/lora/gcp_job.yaml | 25 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | …cipes/vision/llama3_2_vision/sft/11b_lora/gcp_job.yaml | 30 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | …cipes/vision/llama3_2_vision/sft/90b_full/gcp_job.yaml | 30 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | …cipes/vision/llama3_2_vision/sft/11b_full/gcp_job.yaml | 30 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | configs/recipes/vision/internvl3/sft/full/gcp_job.yaml | 25 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | …es/deepseek_r1/sft/distill_llama_70b/full_gcp_job.yaml | 29 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | configs/recipes/llama3_1/sft/8b_lora/polaris_job.yaml | 23 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | configs/recipes/llama3_1/sft/8b_full/polaris_job.yaml | 23 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | …nfigs/recipes/llama3_1/pretraining/8b/polaris_job.yaml | 23 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | configs/recipes/llama3_3/sft/70b_full/gcp_job.yaml | 31 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | scripts/demo.py | 514 | # Display results | COMMENT |
| LOW | …rimental/configs/projects/zephyr/sft/full_gcp_job.yaml | 20 | num_nodes: 1 # Set it to N for multi-node training. | CODE |
| LOW | src/oumi/launcher/clients/local_client.py | 96 | # Check if the user has specified a logging directory. | COMMENT |
| LOW | src/oumi/launcher/clusters/polaris_cluster.py | 249 | # Check if Oumi is installed in a conda env. If not, install it. | COMMENT |
| LOW | src/oumi/launcher/clusters/frontier_cluster.py | 269 | # Check if Oumi is installed in a conda env. If not, install it. | COMMENT |
| LOW | src/oumi/launcher/clouds/frontier_cloud.py | 51 | # Check if any users have open SSH tunnels to Frontier. | COMMENT |
| LOW | src/oumi/launcher/clouds/perlmutter_cloud.py | 51 | # Check if any users have open SSH tunnels to Perlmutter. | COMMENT |
| LOW | src/oumi/launcher/clouds/polaris_cloud.py | 46 | # Check if any users have open SSH tunnels to Polaris. | COMMENT |
| LOW | …c/oumi/core/datasets/pretraining_async_text_dataset.py | 222 | # Set limit to 1 to account for trailing concat token | COMMENT |
| LOW⚡ | src/oumi/core/synthesis/dataset_planner.py | 401 | # Check if all attributes in the override combination are in the sample | COMMENT |
| LOW⚡ | src/oumi/core/synthesis/dataset_planner.py | 409 | # Check if the attribute values are the same | COMMENT |
| LOW | src/oumi/core/trainers/verl_grpo_trainer.py | 246 | # Check if the conversation_json is valid. | COMMENT |
| LOW | src/oumi/core/configs/params/model_params.py | 349 | # Check if flash-attention-2 is requested and supported | COMMENT |
| LOW | src/oumi/core/configs/params/training_params.py | 990 | dataloader_pin_memory=True, # Set it to True to be explicit. | CODE |
| LOW | src/oumi/core/evaluation/evaluator.py | 423 | # Check if the evaluation function requires an inference engine. | COMMENT |
| LOW | src/oumi/core/analyze/dataset_analyzer.py | 431 | # Check if analysis has been run | COMMENT |
| LOW | src/oumi/core/analyze/dataset_analyzer.py | 540 | # Check if analysis has been run | COMMENT |
| LOW | src/oumi/core/analyze/dataset_analyzer.py | 657 | # Check if we have data to analyze | COMMENT |
| LOW | src/oumi/core/analyze/dataframe_analyzer.py | 255 | # Check if all merge columns exist in both DataFrames | COMMENT |
| LOW | src/oumi/builders/models.py | 227 | # Check if model_kwargs contains quantization_config | COMMENT |
| LOW | src/oumi/builders/data.py | 76 | # Check if the underlying dataset is already packed, or if we need to pack it | COMMENT |
| 17 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tests/unit/deploy/__init__.py | 27 | __all__ = [ | CODE |
| LOW | tests/unit/analyze/test_pipeline.py | 218 | def set_dependencies(self, results: dict[str, Any]) -> None: | CODE |
| LOW | tests/unit/analyze/test_pipeline.py | 273 | def set_dependencies(self, results: dict[str, Any]) -> None: | CODE |
| LOW | configs/projects/dcvlr/synthesize_images_vllm.py | 30 | logger = logging.getLogger(__name__) | CODE |
| LOW | src/oumi/__init__.py | 309 | __all__ = [ | CODE |
| LOW | src/oumi/quantize/__init__.py | 60 | __all__ = [ | CODE |
| LOW | src/oumi/launcher/__init__.py | 52 | __all__ = [ | CODE |
| LOW | src/oumi/launcher/clients/local_client.py | 80 | def _update_job_status(self, job_id: str, status: _JobState) -> None: | CODE |
| LOW | src/oumi/launcher/clouds/__init__.py | 47 | __all__ = [ | CODE |
| LOW | src/oumi/core/distributed.py | 610 | def set_random_seeds(seed: int = 42, set_deterministic: bool = False) -> None: | CODE |
| LOW | src/oumi/core/tokenizers/__init__.py | 25 | __all__ = [ | CODE |
| LOW | src/oumi/core/launcher/__init__.py | 32 | __all__ = [ | CODE |
| LOW | src/oumi/core/types/__init__.py | 56 | __all__ = [ | CODE |
| LOW | src/oumi/core/types/exceptions.py | 26 | __all__ = ["HardwareException"] | CODE |
| LOW | src/oumi/core/feature_generators/__init__.py | 29 | __all__ = [ | CODE |
| LOW | src/oumi/core/callbacks/__init__.py | 32 | __all__ = [ | CODE |
| LOW | src/oumi/core/datasets/__init__.py | 46 | __all__ = [ | CODE |
| LOW | src/oumi/core/models/__init__.py | 40 | __all__ = [ | CODE |
| LOW | src/oumi/core/tuners/__init__.py | 33 | __all__ = [ | CODE |
| LOW | src/oumi/core/trainers/oumi_trainer.py | 851 | def _set_sampler_epoch(self, epoch: int) -> None: | CODE |
| LOW | src/oumi/core/trainers/__init__.py | 37 | __all__ = [ | CODE |
| LOW | src/oumi/core/configs/__init__.py | 172 | __all__ = [ | CODE |
| LOW | src/oumi/core/configs/params/rubric_judge_params.py | 24 | logger = logging.getLogger(__name__) | CODE |
| LOW | src/oumi/core/registry/__init__.py | 30 | __all__ = [ | CODE |
| LOW | src/oumi/core/inference/__init__.py | 29 | __all__ = [ | CODE |
| LOW | src/oumi/core/evaluation/__init__.py | 33 | __all__ = [ | CODE |
| LOW | src/oumi/core/evaluation/backends/lm_harness.py | 216 | def _set_random_seeds(random_seed, numpy_random_seed, torch_random_seed) -> None: | CODE |
| LOW | src/oumi/core/analyze/__init__.py | 26 | __all__ = [ | CODE |
| LOW | src/oumi/builders/__init__.py | 52 | __all__ = [ | CODE |
| LOW⚡ | src/oumi/deploy/parasail_api.py | 34 | logger = logging.getLogger(__name__) | CODE |
| LOW | src/oumi/deploy/deploy_config.py | 34 | logger = logging.getLogger(__name__) | CODE |
| LOW | src/oumi/deploy/__init__.py | 33 | __all__ = [ | CODE |
| LOW | src/oumi/deploy/utils.py | 32 | logger = logging.getLogger(__name__) | CODE |
| LOW | src/oumi/deploy/fireworks_client.py | 69 | logger = logging.getLogger(__name__) | CODE |
| LOW | src/oumi/deploy/parasail_client.py | 69 | logger = logging.getLogger(__name__) | CODE |
| LOW | src/oumi/environments/__init__.py | 67 | __all__ = [ | CODE |
| LOW | src/oumi/datasets/__init__.py | 109 | __all__ = [ | CODE |
| LOW | src/oumi/datasets/vision_language/__init__.py | 42 | __all__ = [ | CODE |
| LOW | src/oumi/datasets/preference_tuning/__init__.py | 21 | __all__ = [ | CODE |
| LOW | src/oumi/datasets/grpo/__init__.py | 29 | __all__ = [ | CODE |
| LOW | src/oumi/datasets/grpo/rewards/rar_medicine_reward.py | 29 | logger = logging.getLogger(__name__) | CODE |
| LOW | src/oumi/datasets/grpo/rewards/__init__.py | 26 | __all__ = [ | CODE |
| LOW | src/oumi/datasets/grpo/rollouts/__init__.py | 17 | __all__ = [] | CODE |
| LOW | src/oumi/datasets/evaluation/__init__.py | 19 | __all__ = [ | CODE |
| LOW | src/oumi/datasets/sft/__init__.py | 32 | __all__ = [ | CODE |
| LOW | src/oumi/datasets/pretraining/__init__.py | 33 | __all__ = [ | CODE |
| LOW | src/oumi/utils/logging.py | 177 | def update_logger_level(name: str, level: str = "info") -> None: | CODE |
| LOW | src/oumi/utils/analysis_utils.py | 28 | logger = logging.getLogger(__name__) | CODE |
| LOW | src/oumi/utils/vllm_utils/gemma4_compat.py | 47 | logger = logging.getLogger(__name__) | CODE |
| LOW | src/oumi/mcp/job_launcher.py | 33 | logger = logging.getLogger(__name__) | CODE |
| LOW | src/oumi/mcp/server.py | 134 | logger = logging.getLogger(__name__) | CODE |
| LOW | src/oumi/mcp/job_logs.py | 28 | logger = logging.getLogger(__name__) | CODE |
| LOW | src/oumi/mcp/preflight_service.py | 79 | logger = logging.getLogger(__name__) | CODE |
| LOW | src/oumi/mcp/job_service.py | 59 | logger = logging.getLogger(__name__) | CODE |
| LOW | src/oumi/mcp/job_service.py | 61 | __all__ = [ | CODE |
| LOW | src/oumi/mcp/sync_service.py | 44 | logger = logging.getLogger(__name__) | CODE |
| LOW | src/oumi/mcp/docs_service.py | 55 | logger = logging.getLogger(__name__) | CODE |
| LOW | src/oumi/mcp/environment_service.py | 24 | logger = logging.getLogger(__name__) | CODE |
| LOW | src/oumi/mcp/config_service.py | 56 | logger = logging.getLogger(__name__) | CODE |
| LOW | src/oumi/mcp/job_registry.py | 31 | logger = logging.getLogger(__name__) | CODE |
| 23 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | docs/user_guides/infer/configuration.md | 157 | api_key: "your-api-key" # API key for authentication | CODE |
| HIGH | configs/examples/deploy/README.md | 195 | export FIREWORKS_API_KEY="your-api-key" | CODE |
| HIGH | …gs/examples/synthesis/instruction_following_synth.yaml | 7 | # - example: `export ANTHROPIC_API_KEY=your_api_key` | COMMENT |
| HIGH | configs/examples/synthesis/conversation_synth.yaml | 7 | # - example: `export ANTHROPIC_API_KEY=your_api_key` | COMMENT |
| HIGH | configs/examples/synthesis/question_answer_synth.yaml | 7 | # - example: `export ANTHROPIC_API_KEY=your_api_key` | COMMENT |
| HIGH⚡ | configs/examples/synthesis/dynamic_few_shot_synth.yaml | 7 | # - example: `export ANTHROPIC_API_KEY=your_api_key` | COMMENT |
| HIGH | configs/examples/synthesis/data_augmentation_synth.yaml | 7 | # - example: `export ANTHROPIC_API_KEY=your_api_key` | COMMENT |
| HIGH | …s/examples/synthesis/multiturn_conversation_synth.yaml | 6 | # - example: `export ANTHROPIC_API_KEY=your_api_key` | COMMENT |
| HIGH | configs/examples/synthesis/domain_qa_synth.yaml | 7 | # - example: `export ANTHROPIC_API_KEY=your_api_key` | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | tests/unit/core/evaluation/test_backend_lm_harness.py | 342 | # A dual-mode model with text_only=True must route to the text harness model | COMMENT |
| LOW | tests/unit/inference/test_generation_params.py | 477 | # To avoid running inference, we just call the method that converts | COMMENT |
| LOW | tests/e2e/test_eval_e2e.py | 88 | # or impossible. Let's just create a modified config copy instead. | COMMENT |
| MEDIUM | configs/projects/judges/code/code_quality.yaml | 2 | # Evaluates code for readability, maintainability, and best practices | COMMENT |
| MEDIUM | configs/projects/judges/code/performance.yaml | 2 | # Evaluates code for efficiency and performance best practices | COMMENT |
| MEDIUM | configs/recipes/qwen3/inference/14b_infer.yaml | 24 | # Set based on recommended Qwen3 best practices for thinking model: | COMMENT |
| MEDIUM | configs/recipes/qwen3/inference/0.6b_infer.yaml | 24 | # Set based on recommended Qwen3 best practices for thinking model: | COMMENT |
| MEDIUM | configs/recipes/qwen3/inference/30b_a3b_infer.yaml | 24 | # Set based on recommended Qwen3 best practices for thinking model: | COMMENT |
| MEDIUM | configs/recipes/qwen3/inference/4b_infer.yaml | 24 | # Set based on recommended Qwen3 best practices for thinking model: | COMMENT |
| MEDIUM | configs/recipes/qwen3/inference/8b_infer.yaml | 24 | # Set based on recommended Qwen3 best practices for thinking model: | COMMENT |
| MEDIUM | configs/recipes/qwen3/inference/32b_infer.yaml | 24 | # Set based on recommended Qwen3 best practices for thinking model: | COMMENT |
| MEDIUM | configs/recipes/qwen3/inference/1.7b_infer.yaml | 24 | # Set based on recommended Qwen3 best practices for thinking model: | COMMENT |
| MEDIUM | src/oumi/core/callbacks/bitnet_callback.py | 54 | """You need `onebitllms` to be installed in order to save | STRING |
| MEDIUM | src/oumi/core/configs/params/evaluation_params.py | 148 | """Number of few-shot examples (with responses) to add in the prompt, in order to | STRING |
| MEDIUM | src/oumi/core/evaluation/backends/lm_harness.py | 94 | # LM Harness `model` | Class | File in lm-evaluation-harness repo # | COMMENT |
| MEDIUM⚡ | src/oumi/core/evaluation/backends/lm_harness.py | 115 | # https://github.com/EleutherAI/lm-evaluation-harness/blob/365fcda9b85bbb6e0572d91976b8daf409164500/lm_eval/models/h | COMMENT |
| MEDIUM⚡ | src/oumi/core/evaluation/backends/lm_harness.py | 119 | # https://github.com/EleutherAI/lm-evaluation-harness/blob/main/docs/interface.md | COMMENT |
| MEDIUM | src/oumi/core/analyze/dataset_analyzer.py | 645 | """Generate a comprehensive summary of dataset analysis results. | STRING |
| MEDIUM | src/oumi/core/analyze/dataset_analyzer.py | 678 | """Get the comprehensive analysis summary. | STRING |
| MEDIUM | src/oumi/builders/models.py | 94 | """Please install `onebitllms` in order to fine-tune | STRING |
| MEDIUM | src/oumi/builders/models.py | 428 | # Identify the tokenizer we need to leverage for this model. | COMMENT |
| MEDIUM | src/oumi/utils/system_info.py | 165 | """Get comprehensive system information. | STRING |
| LOW | src/oumi/utils/cache_utils.py | 138 | # (In practice, we just pass through to the original function) | COMMENT |
| MEDIUM | src/oumi/utils/torch_naming_heuristics.py | 17 | # TODO(OPE-303): These should be replaced with something more robust. | STRING |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | platform/banking77-classifier/SPEC.md | 7 | ### Step 1: Download and read dataset | COMMENT |
| LOW | platform/banking77-classifier/SPEC.md | 10 | ### Step 2: Transform into Oumi-compatible dataset | COMMENT |
| LOW | platform/banking77-classifier/SPEC.md | 59 | ### Step 3: Create transformed dataset | COMMENT |
| LOW⚡ | tests/unit/inference/test_remote_inference_engine.py | 3682 | # Step 11: Concurrency should be 12, window size 1 | COMMENT |
| LOW⚡ | tests/unit/inference/test_remote_inference_engine.py | 3684 | # Step 21: Backoff to 10, window size 1 | COMMENT |
| LOW⚡ | tests/unit/inference/test_remote_inference_engine.py | 3686 | # Step 31: Concurrency should stay 10, first good window | COMMENT |
| LOW⚡ | tests/unit/inference/test_remote_inference_engine.py | 3688 | # Step 41: 2nd good window, exit backoff, warmup to 12, window size 1 | COMMENT |
| LOW | docs/user_guides/evaluate/custom_evals.md | 9 | ### Step 1: Defining Evaluation Configuration | COMMENT |
| LOW | docs/user_guides/evaluate/custom_evals.md | 31 | ### Step 2: Defining Custom Evaluation Function | COMMENT |
| LOW | docs/user_guides/evaluate/custom_evals.md | 56 | ### Step 3: Executing the Evaluation | COMMENT |
| LOW | docs/user_guides/evaluate/custom_evals.md | 95 | ### Step 1: Defining the Evaluation Configuration | COMMENT |
| LOW | docs/user_guides/evaluate/custom_evals.md | 112 | ### Step 2: Defining Custom Evaluation Function | STRING |
| LOW | docs/user_guides/evaluate/custom_evals.md | 137 | ### Step 3: Executing the Evaluation | STRING |
| LOW⚡ | src/oumi/quantize/awq_quantizer.py | 85 | # Step 1: AWQ quantization | COMMENT |
| LOW⚡ | src/oumi/quantize/awq_quantizer.py | 88 | # Step 2: Save as PyTorch format | COMMENT |
| LOW | …/collators/trl_data_collator_for_completion_only_lm.py | 302 | # Step 1: mask everything. | COMMENT |
| LOW | …/collators/trl_data_collator_for_completion_only_lm.py | 317 | # Step 2: find every assistant response start position. | COMMENT |
| LOW⚡ | …/collators/trl_data_collator_for_completion_only_lm.py | 332 | # Step 3: find the next end_of_turn after content_start. | COMMENT |
| LOW⚡ | …/collators/trl_data_collator_for_completion_only_lm.py | 342 | # Step 4: unmask this assistant response span, including the | COMMENT |
| LOW⚡ | src/oumi/deploy/fireworks_client.py | 471 | # Step 1: Resolve model source to a local directory so we can | COMMENT |
| LOW⚡ | src/oumi/deploy/fireworks_client.py | 479 | # Step 2: Collect the file manifest. For HF_BASE_MODEL uploads the | COMMENT |
| LOW | src/oumi/mcp/prompts/post_training.py | 38 | ## Step 1: Download Model Weights | COMMENT |
| LOW | src/oumi/mcp/prompts/post_training.py | 58 | ## Step 2: Run Evaluation (Optional, On-Cluster) | COMMENT |
| LOW | src/oumi/mcp/prompts/post_training.py | 69 | ## Step 3: Tear Down the Cluster | COMMENT |
| LOW | src/oumi/mcp/prompts/post_training.py | 82 | ## Step 4: Merge LoRA Adapter (Local) | COMMENT |
| LOW | src/oumi/mcp/prompts/post_training.py | 104 | ## Step 5: Push to HuggingFace Hub (Optional) | COMMENT |
| LOW | src/oumi/inference/together_inference_engine.py | 133 | # Step 1: Request signed upload URL | COMMENT |
| LOW | src/oumi/inference/together_inference_engine.py | 167 | # Step 2: Upload file content to signed URL | COMMENT |
| LOW | src/oumi/inference/together_inference_engine.py | 181 | # Step 3: Finalize upload by calling preprocess endpoint | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | pyproject.toml | 135 | # When updating verl version, make sure to also update the default config: | COMMENT |
| MEDIUM | …s/unit/core/synthesis/test_conversation_synthesizer.py | 949 | "Let me know if you need any changes." | CODE |
| LOW | configs/projects/aya/sft/gcp_job.yaml | 26 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| LOW | configs/recipes/vision/phi3/sft/full/oumi_gcp_job.yaml | 23 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| LOW | configs/recipes/vision/phi3/sft/full/trl_gcp_job.yaml | 23 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| LOW | configs/recipes/vision/phi3/sft/lora/gcp_job.yaml | 23 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| LOW | configs/recipes/vision/llava_7b/sft/oumi_gcp_job.yaml | 22 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| LOW | configs/recipes/vision/llava_7b/sft/trl_gcp_job.yaml | 24 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| LOW | …cipes/vision/llama3_2_vision/sft/11b_lora/gcp_job.yaml | 24 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| LOW | …cipes/vision/llama3_2_vision/sft/90b_full/gcp_job.yaml | 24 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| LOW | …cipes/vision/llama3_2_vision/sft/11b_full/gcp_job.yaml | 24 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| LOW | …es/deepseek_r1/sft/distill_llama_70b/full_gcp_job.yaml | 23 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| LOW | …s/deepseek_r1/sft/distill_llama_70b/qlora_gcp_job.yaml | 22 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| LOW | …es/deepseek_r1/sft/distill_llama_70b/lora_gcp_job.yaml | 22 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| LOW | configs/recipes/llama3_1/sft/405b_qlora/gcp_job.yaml | 24 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| LOW | configs/recipes/llama3_1/sft/405b_lora/gcp_job.yaml | 25 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| LOW | configs/recipes/llama3_1/sft/70b_qlora/gcp_job.yaml | 24 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| LOW | configs/recipes/llama3_1/sft/70b_full/gcp_job.yaml | 25 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| LOW | configs/recipes/llama3_3/sft/70b_lora/gcp_job.yaml | 24 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| LOW | configs/recipes/llama3_3/sft/70b_qlora/gcp_job.yaml | 24 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| LOW | configs/recipes/llama3_3/sft/70b_full/gcp_job.yaml | 25 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| MEDIUM | .github/workflows/codeql.yml | 56 | # If you are analyzing a compiled language, you can modify the 'build-mode' for that language to customize how | COMMENT |
| LOW | …rimental/configs/projects/zephyr/sft/full_gcp_job.yaml | 26 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| LOW | …imental/configs/projects/zephyr/sft/qlora_gcp_job.yaml | 24 | # However, make sure you are saving your output to a mounted cloud storage in case of | COMMENT |
| MEDIUM | src/oumi/builders/models.py | 396 | When ``text_only`` is True the model is treated as a language model regardless | STRING |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW⚡ | tests/unit/inference/test_remote_inference_engine.py | 3682 | # Step 11: Concurrency should be 12, window size 1 | COMMENT |
| LOW⚡ | tests/unit/inference/test_remote_inference_engine.py | 3684 | # Step 21: Backoff to 10, window size 1 | COMMENT |
| LOW⚡ | tests/unit/inference/test_remote_inference_engine.py | 3686 | # Step 31: Concurrency should stay 10, first good window | COMMENT |
| LOW⚡ | tests/unit/inference/test_remote_inference_engine.py | 3688 | # Step 41: 2nd good window, exit backoff, warmup to 12, window size 1 | COMMENT |
| LOW⚡ | src/oumi/quantize/awq_quantizer.py | 85 | # Step 1: AWQ quantization | COMMENT |
| LOW⚡ | src/oumi/quantize/awq_quantizer.py | 88 | # Step 2: Save as PyTorch format | COMMENT |
| LOW | …/collators/trl_data_collator_for_completion_only_lm.py | 302 | # Step 1: mask everything. | COMMENT |
| LOW | …/collators/trl_data_collator_for_completion_only_lm.py | 317 | # Step 2: find every assistant response start position. | COMMENT |
| LOW⚡ | …/collators/trl_data_collator_for_completion_only_lm.py | 332 | # Step 3: find the next end_of_turn after content_start. | COMMENT |
| LOW⚡ | …/collators/trl_data_collator_for_completion_only_lm.py | 342 | # Step 4: unmask this assistant response span, including the | COMMENT |
| LOW⚡ | src/oumi/deploy/fireworks_client.py | 471 | # Step 1: Resolve model source to a local directory so we can | COMMENT |
| LOW⚡ | src/oumi/deploy/fireworks_client.py | 479 | # Step 2: Collect the file manifest. For HF_BASE_MODEL uploads the | COMMENT |
| LOW | src/oumi/mcp/prompts/post_training.py | 38 | ## Step 1: Download Model Weights | COMMENT |
| LOW | src/oumi/mcp/prompts/post_training.py | 58 | ## Step 2: Run Evaluation (Optional, On-Cluster) | COMMENT |
| LOW | src/oumi/mcp/prompts/post_training.py | 69 | ## Step 3: Tear Down the Cluster | COMMENT |
| LOW | src/oumi/mcp/prompts/post_training.py | 82 | ## Step 4: Merge LoRA Adapter (Local) | COMMENT |
| LOW | src/oumi/mcp/prompts/post_training.py | 104 | ## Step 5: Push to HuggingFace Hub (Optional) | COMMENT |
| LOW | src/oumi/inference/together_inference_engine.py | 133 | # Step 1: Request signed upload URL | COMMENT |
| LOW | src/oumi/inference/together_inference_engine.py | 167 | # Step 2: Upload file content to signed URL | COMMENT |
| LOW | src/oumi/inference/together_inference_engine.py | 181 | # Step 3: Finalize upload by calling preprocess endpoint | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | README.md | 26 | - [2026/09] [Oumi v0.9 released](https://github.com/oumi-ai/oumi/releases/tag/v0.9): an end-to-end stack for agentic, to | CODE |
| MEDIUM | README.md | 46 | - [2025/11] [Example notebook to perform RLVF fine-tuning with OpenEnv](https://github.com/oumi-ai/oumi/blob/main/notebo | CODE |
| MEDIUM | configs/examples/synthesis/README.md | 12 | 6. [Dynamic Few-Shot Sampling](#6-dynamic-few-shot-sampling-dynamic_few_shot_synthyaml) - Randomly sample examples for d | CODE |
| MEDIUM | configs/examples/synthesis/README.md | 495 | # Enable dynamic few-shot sampling | COMMENT |
| MEDIUM⚡ | configs/examples/synthesis/dynamic_few_shot_synth.yaml | 2 | # Demonstrates the new dynamic few-shot sampling feature for synthesis. | COMMENT |
| MEDIUM | configs/examples/synthesis/dynamic_few_shot_synth.yaml | 56 | # Generate diverse task instructions using the randomly sampled few-shot examples | COMMENT |
| MEDIUM | scripts/datasets/save_conversations.py | 9 | # --subset zero-shot-exp --split train --max-conversations 100 \ | COMMENT |
| MEDIUM | src/oumi/core/datasets/packed_sft_dataset.py | 97 | # the context window. There is no way to handle this sample | COMMENT |
| MEDIUM | src/oumi/core/evaluation/backends/lm_harness.py | 45 | # Used to set the few-shot seed for lm_eval.api.task.Task. The value is consistent with | COMMENT |
| MEDIUM | src/oumi/datasets/mmlu.py | 81 | "dev", # For few-shot development; 5 questions per subject (285 questions) | CODE |
| MEDIUM | src/oumi/judges/base_judge.py | 736 | # Build few-shot examples: convert field values to prompts and responses | COMMENT |
| MEDIUM | src/oumi/judges/base_judge.py | 898 | # Add few-shot examples as alternating user/assistant pairs | COMMENT |
| MEDIUM | src/oumi/judges/rubric_judge.py | 43 | # Prompt scaffolding: introducing the rubric to the judge. | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| CRITICAL⚡ | tests/unit/launcher/clients/test_modal_client.py | 142 | fake_modal.Sandbox.create.return_value.set_tags.assert_called_once_with( | CODE |
| CRITICAL | src/oumi/core/trainers/hf_trainer.py | 169 | self._hf_trainer.accelerator.state.fsdp_plugin.set_state_dict_type( | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | scripts/benchmarks/minimal_multimodal_training.py | 206 | CODE | |
| LOW | src/oumi/launcher/clients/slurm_client.py | 620 | CODE | |
| LOW | src/oumi/core/distributed.py | 607 | CODE | |
| LOW | …tors/vision_language_conversation_feature_generator.py | 69 | CODE | |
| LOW | src/oumi/core/datasets/base_sft_dataset.py | 38 | CODE | |
| LOW | src/oumi/core/datasets/base_map_dataset.py | 126 | CODE | |
| LOW | src/oumi/core/datasets/vision_language_dataset.py | 64 | CODE | |
| LOW | src/oumi/core/datasets/vision_language_dpo_dataset.py | 67 | CODE | |
| LOW | …c/oumi/core/datasets/pretraining_async_text_dataset.py | 38 | CODE | |
| LOW | src/oumi/core/trainers/verl_grpo_trainer.py | 422 | CODE | |
| LOW | …re/collators/text_completions_collator_with_padding.py | 29 | CODE | |
| LOW | src/oumi/core/collators/vision_language_sft_collator.py | 73 | CODE | |
| LOW | …/collators/trl_data_collator_for_completion_only_lm.py | 93 | CODE | |
| LOW | src/oumi/utils/device_utils.py | 121 | CODE | |
| LOW | src/oumi/cli/train.py | 29 | CODE | |
| LOW | src/oumi/cli/infer.py | 33 | CODE | |
| LOW | src/oumi/cli/analyze.py | 393 | CODE | |
| LOW | src/oumi/inference/vllm_inference_engine.py | 286 | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW⚡ | tests/unit/inference/test_gcp_inference_engine.py | 119 | mock_credentials.from_service_account_file.return_value.token = "fake_token" | CODE |
| LOW⚡ | tests/unit/inference/test_gcp_inference_engine.py | 121 | assert token == "fake_token" | CODE |
| LOW⚡ | tests/unit/inference/test_gcp_inference_engine.py | 129 | with patch.object(gcp_engine, "_get_api_key", return_value="fake_token"): | CODE |
| LOW⚡ | tests/unit/inference/test_gcp_inference_engine.py | 133 | "Authorization": "Bearer fake_token", | CODE |
| LOW | tests/integration/judges/test_rule_based_judge_e2e.py | 146 | {"text": "user@example.com"}, # Valid email | CODE |
| LOW | tests/integration/judges/test_rule_based_judge_e2e.py | 204 | test_data = [{"response": "555-1234"}] | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tests/unit/core/test_distributed.py | 101 | def test_function(): | CODE |
| LOW | tests/unit/core/test_distributed.py | 254 | def test_function(): | CODE |
| LOW | tests/unit/core/test_distributed.py | 269 | def test_function(): | CODE |