Making large AI models cheaper, faster and more accessible
3803 matches across 17 categories. Click a row to expand file-level details.
| Severity | File | Line | Snippet |
|---|---|---|---|
| MEDIUM | …sts/test_shardformer/test_model/test_shard_deepseek.py | 130 | # =================================================================================== |
| MEDIUM | tests/test_shardformer/test_model/test_shard_mixtral.py | 123 | # =================================================================================== |
| MEDIUM | tests/kit/model_zoo/timm/timm.py | 6 | ## ============== |
| MEDIUM | tests/kit/model_zoo/timm/timm.py | 8 | ## ============== |
| MEDIUM | tests/kit/model_zoo/timm/timm.py | 150 | # ============== |
| MEDIUM | tests/kit/model_zoo/timm/timm.py | 152 | # ============== |
| MEDIUM | tests/kit/model_zoo/transformers/command.py | 14 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/command.py | 16 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/mistral.py | 7 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/mistral.py | 9 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/t5.py | 6 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/t5.py | 8 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/gptj.py | 8 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/gptj.py | 10 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/albert.py | 6 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/albert.py | 8 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/albert.py | 69 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/albert.py | 71 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/opt.py | 6 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/opt.py | 8 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/chatglm2.py | 7 | # ================================ |
| MEDIUM | tests/kit/model_zoo/transformers/chatglm2.py | 9 | # ================================ |
| MEDIUM | tests/kit/model_zoo/transformers/whisper.py | 6 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/whisper.py | 8 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/llama.py | 14 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/llama.py | 16 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/llama.py | 23 | # ----------------------------------- |
| MEDIUM | tests/kit/model_zoo/transformers/llama.py | 28 | # ----------------------------------- |
| MEDIUM | tests/kit/model_zoo/transformers/vit.py | 6 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/vit.py | 8 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/deepseek.py | 8 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/deepseek.py | 10 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/bert.py | 6 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/bert.py | 8 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/qwen3.py | 14 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/qwen3.py | 16 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/qwen3.py | 23 | # ----------------------------------- |
| MEDIUM | tests/kit/model_zoo/transformers/qwen3.py | 28 | # ----------------------------------- |
| MEDIUM | tests/kit/model_zoo/transformers/deepseek_v3.py | 8 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/deepseek_v3.py | 10 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/falcon.py | 6 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/falcon.py | 8 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/sam.py | 6 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/sam.py | 8 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/qwen2.py | 14 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/qwen2.py | 16 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/qwen2.py | 23 | # ----------------------------------- |
| MEDIUM | tests/kit/model_zoo/transformers/qwen2.py | 28 | # ----------------------------------- |
| MEDIUM | tests/kit/model_zoo/transformers/mixtral.py | 8 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/mixtral.py | 10 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/gpt.py | 8 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/gpt.py | 10 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/bloom.py | 6 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/bloom.py | 8 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/blip2.py | 6 | # =============================== |
| MEDIUM | tests/kit/model_zoo/transformers/blip2.py | 8 | # =============================== |
| MEDIUM | tests/test_pipeline/test_schedule/test_zerobubble_pp.py | 890 | # =================================================================================== |
| MEDIUM | tests/test_pipeline/test_schedule/test_zerobubble_pp.py | 1045 | # =================================================================================== |
| MEDIUM | …/test_fx/test_profiler/test_profiler_meta_info_prop.py | 67 | # ===================================================== |
| MEDIUM | …/test_fx/test_profiler/test_profiler_meta_info_prop.py | 83 | # ===================================================== |
| 471 more matches not shown… | |||
| Severity | File | Line | Snippet |
|---|---|---|---|
| HIGH | tests/test_shardformer/test_layer/test_dist_log_prob.py | 0 | compute the log probabilities from logits for the given labels. args: logits (torch.tensor): the input logits. labels (t |
| HIGH | applications/ColossalChat/coati/distributed/utils.py | 0 | compute the log probabilities from logits for the given labels. args: logits (torch.tensor): the input logits. labels (t |
| HIGH | applications/ColossalChat/coati/models/utils.py | 0 | compute the log probabilities from logits for the given labels. args: logits (torch.tensor): the input logits. labels (t |
| HIGH | tests/test_infer/test_kernels/triton/kernel_utils.py | 0 | allocate block tables based on provided context lengths; and copy kv to blocked kv cache. |
| HIGH | tests/test_infer/test_kernels/triton/kernel_utils.py | 0 | allocate block tables based on provided context lengths; and copy kv to blocked kv cache. |
| HIGH | tests/test_infer/test_kernels/triton/kernel_utils.py | 0 | allocate block tables based on provided context lengths; and copy kv to blocked kv cache. |
| HIGH | tests/test_infer/test_kernels/triton/kernel_utils.py | 0 | allocate block tables based on provided context lengths; and copy kv to blocked kv cache. |
| HIGH | colossalai/pipeline/schedule/interleaved_pp.py | 0 | load a batch from data iterator. args: data_iter (iterable): data iterator. device (optional[torch.device], optional): t |
| HIGH | colossalai/pipeline/schedule/one_f_one_b.py | 0 | load a batch from data iterator. args: data_iter (iterable): data iterator. device (optional[torch.device], optional): t |
| HIGH | colossalai/pipeline/schedule/generate.py | 0 | load a batch from data iterator. args: data_iter (iterable): data iterator. device (optional[torch.device], optional): t |
| HIGH | colossalai/pipeline/schedule/zero_bubble_pp.py | 0 | load a batch from data iterator. args: data_iter (iterable): data iterator. device (optional[torch.device], optional): t |
| HIGH | colossalai/nn/optimizer/lamb.py | 0 | performs a single optimization step. arguments: closure (callable, optional): a closure that reevaluates the model and r |
| HIGH | colossalai/nn/optimizer/lars.py | 0 | performs a single optimization step. arguments: closure (callable, optional): a closure that reevaluates the model and r |
| HIGH | colossalai/nn/optimizer/galore.py | 0 | performs a single optimization step. arguments: closure (callable, optional): a closure that reevaluates the model and r |
| HIGH | colossalai/nn/optimizer/distributed_lamb.py | 0 | performs a single optimization step. arguments: closure (callable, optional): a closure that reevaluates the model and r |
| HIGH | colossalai/nn/optimizer/distributed_galore.py | 0 | performs a single optimization step. arguments: closure (callable, optional): a closure that reevaluates the model and r |
| HIGH | colossalai/nn/optimizer/fused_lamb.py | 0 | performs a single optimization step. arguments: closure (callable, optional): a closure that reevaluates the model and r |
| HIGH | colossalai/nn/optimizer/fused_sgd.py | 0 | performs a single optimization step. arguments: closure (callable, optional): a closure that reevaluates the model and r |
| HIGH | colossalai/nn/optimizer/came.py | 0 | performs a single optimization step. args: closure (callable, optional): a closure that reevaluates the model and return |
| HIGH | colossalai/nn/optimizer/distributed_came.py | 0 | performs a single optimization step. args: closure (callable, optional): a closure that reevaluates the model and return |
| HIGH | examples/images/diffusion/ldm/util.py | 0 | performs a single optimization step. args: closure (callable, optional): a closure that reevaluates the model and return |
| HIGH | colossalai/nn/layer/utils.py | 0 | only allow exact division. args: numerator (int): numerator of the division. denominator (int): denominator of the divis |
| HIGH | colossalai/legacy/nn/layer/utils/common.py | 0 | only allow exact division. args: numerator (int): numerator of the division. denominator (int): denominator of the divis |
| HIGH | colossalai/legacy/tensor/dist_spec_mgr.py | 0 | only allow exact division. args: numerator (int): numerator of the division. denominator (int): denominator of the divis |
| HIGH | …tched_bias_addition_function/bias_addition_function.py | 0 | this method is used to create the non_bias_func proxy, the node created by this proxy will compute the main computation, |
| HIGH | …ddition_patch/patched_bias_addition_function/addbmm.py | 0 | this method is used to create the non_bias_func proxy, the node created by this proxy will compute the main computation, |
| HIGH | …h/patched_bias_addition_module/bias_addition_module.py | 0 | this method is used to create the non_bias_func proxy, the node created by this proxy will compute the main computation, |
| HIGH | colossalai/fx/codegen/activation_checkpoint_codegen.py | 0 | add an obj to be tracked as a global. we call this for names that reference objects external to the graph, like function |
| HIGH | colossalai/fx/codegen/activation_checkpoint_codegen.py | 0 | add an obj to be tracked as a global. we call this for names that reference objects external to the graph, like function |
| HIGH | colossalai/autochunk/autochunk_codegen.py | 0 | add an obj to be tracked as a global. we call this for names that reference objects external to the graph, like function |
| HIGH | colossalai/_analyzer/fx/codegen.py | 0 | add an obj to be tracked as a global. we call this for names that reference objects external to the graph, like function |
| HIGH | colossalai/fx/codegen/activation_checkpoint_codegen.py | 0 | delete values after their last use. this ensures that values that are not used in the remainder of the code are freed an |
| HIGH | colossalai/fx/codegen/activation_checkpoint_codegen.py | 0 | delete values after their last use. this ensures that values that are not used in the remainder of the code are freed an |
| HIGH | colossalai/autochunk/autochunk_codegen.py | 0 | delete values after their last use. this ensures that values that are not used in the remainder of the code are freed an |
| HIGH | colossalai/_analyzer/fx/codegen.py | 0 | delete values after their last use. this ensures that values that are not used in the remainder of the code are freed an |
| HIGH | colossalai/booster/plugin/gemini_plugin.py | 0 | save model to checkpoint but only on master process. |
| HIGH | colossalai/booster/plugin/torch_fsdp_plugin.py | 0 | save model to checkpoint but only on master process. |
| HIGH | colossalai/booster/plugin/torch_fsdp_plugin.py | 0 | save model to checkpoint but only on master process. |
| HIGH | colossalai/booster/plugin/torch_fsdp_plugin.py | 0 | save model to checkpoint but only on master process. |
| HIGH | colossalai/booster/plugin/torch_ddp_plugin.py | 0 | save model to checkpoint but only on master process. |
| HIGH | colossalai/booster/plugin/torch_ddp_plugin.py | 0 | save model to checkpoint but only on master process. |
| HIGH | colossalai/booster/plugin/torch_ddp_plugin.py | 0 | save model to checkpoint but only on master process. |
| HIGH | colossalai/booster/plugin/torch_fsdp_plugin.py | 0 | save optimizer to checkpoint but only on master process. |
| HIGH | colossalai/booster/plugin/torch_fsdp_plugin.py | 0 | save optimizer to checkpoint but only on master process. |
| HIGH | colossalai/booster/plugin/torch_ddp_plugin.py | 0 | save optimizer to checkpoint but only on master process. |
| HIGH | colossalai/booster/plugin/hybrid_parallel_plugin.py | 0 | compute and return the gradient norm for gradient clipping. args: param_gradient_pairs (list[tuple[tensor]]): list of (p |
| HIGH | colossalai/booster/plugin/hybrid_parallel_plugin.py | 0 | compute and return the gradient norm for gradient clipping. args: param_gradient_pairs (list[tuple[tensor]]): list of (p |
| HIGH | colossalai/amp/naive_amp/mixed_precision_optimizer.py | 0 | compute and return the gradient norm for gradient clipping. args: param_gradient_pairs (list[tuple[tensor]]): list of (p |
| HIGH | colossalai/shardformer/policies/command.py | 0 | if under pipeline parallel setting, replacing the original forward method of huggingface to customized forward method, a |
| HIGH | colossalai/shardformer/policies/mistral.py | 0 | if under pipeline parallel setting, replacing the original forward method of huggingface to customized forward method, a |
| HIGH | colossalai/shardformer/policies/gpt2.py | 0 | if under pipeline parallel setting, replacing the original forward method of huggingface to customized forward method, a |
| HIGH | colossalai/shardformer/policies/t5.py | 0 | if under pipeline parallel setting, replacing the original forward method of huggingface to customized forward method, a |
| HIGH | colossalai/shardformer/policies/gptj.py | 0 | if under pipeline parallel setting, replacing the original forward method of huggingface to customized forward method, a |
| HIGH | colossalai/shardformer/policies/opt.py | 0 | if under pipeline parallel setting, replacing the original forward method of huggingface to customized forward method, a |
| HIGH | colossalai/shardformer/policies/chatglm2.py | 0 | if under pipeline parallel setting, replacing the original forward method of huggingface to customized forward method, a |
| HIGH | colossalai/shardformer/policies/whisper.py | 0 | if under pipeline parallel setting, replacing the original forward method of huggingface to customized forward method, a |
| HIGH | colossalai/shardformer/policies/llama.py | 0 | if under pipeline parallel setting, replacing the original forward method of huggingface to customized forward method, a |
| HIGH | colossalai/shardformer/policies/deepseek.py | 0 | if under pipeline parallel setting, replacing the original forward method of huggingface to customized forward method, a |
| HIGH | colossalai/shardformer/policies/bert.py | 0 | if under pipeline parallel setting, replacing the original forward method of huggingface to customized forward method, a |
| HIGH | colossalai/shardformer/policies/qwen3.py | 0 | if under pipeline parallel setting, replacing the original forward method of huggingface to customized forward method, a |
| 265 more matches not shown… | |||
| Severity | File | Line | Snippet |
|---|---|---|---|
| LOW | …heckpoint_io/test_plugins_huggingface_compatibility.py | 78 | def test_huggingface_compatibility(world_size): |
| LOW | tests/test_checkpoint_io/test_gemini_checkpoint_io.py | 39 | def exam_state_dict_with_origin( |
| LOW | tests/test_checkpoint_io/test_gemini_checkpoint_io.py | 193 | def exam_lazy_from_pretrained(): |
| LOW | tests/test_checkpoint_io/test_general_checkpoint_io.py | 23 | def test_unsharded_checkpoint(use_safetensors: bool, use_async: bool): |
| LOW | tests/test_checkpoint_io/test_general_checkpoint_io.py | 77 | def test_sharded_model_checkpoint(use_safetensors: bool, use_async: bool): |
| LOW | tests/test_checkpoint_io/test_general_checkpoint_io.py | 117 | def test_sharded_optimizer_checkpoint(use_async: bool): |
| LOW | tests/test_checkpoint_io/test_general_checkpoint_io.py | 186 | def test_sharded_optimizer_multiple_param_groups(use_async: bool): |
| LOW | …ts/test_checkpoint_io/test_gemini_torch_compability.py | 24 | def exam_torch_load_from_gemini(shard: bool, model_name: str): |
| LOW | …ts/test_checkpoint_io/test_gemini_torch_compability.py | 90 | def exam_gemini_load_from_torch(shard: bool, model_name: str): |
| LOW | tests/test_zero/test_gemini/test_zeroddp_state_dict.py | 18 | def ignore_the_first_parameter(model: torch.nn.Module): |
| LOW | …sts/test_zero/test_gemini/test_zerooptim_state_dict.py | 23 | def exam_zero_optim_state_dict(placement_config, keep_gathered): |
| LOW | tests/test_shardformer/test_flash_attention.py | 67 | def post_process_kwargs_for_raw_attn(attn_kwargs: dict): |
| LOW | tests/test_shardformer/test_with_torch_ddp.py | 18 | def check_shardformer_with_ddp(lazy_init: bool): |
| LOW | tests/test_shardformer/test_model/_utils.py | 120 | def build_model_from_hybrid_plugin( |
| LOW | tests/test_shardformer/test_model/_utils.py | 174 | def build_model_from_low_level_zero_plugin( |
| LOW | tests/test_shardformer/test_model/_utils.py | 200 | def run_forward_backward_with_hybrid_plugin( |
| LOW | tests/test_shardformer/test_model/_utils.py | 263 | def run_forward_backward_with_low_level_zero_plugin( |
| LOW | tests/test_shardformer/test_model/_utils.py | 301 | def check_output_hidden_state( |
| LOW | tests/test_shardformer/test_model/_utils.py | 368 | def get_grad_tensors_for_check( |
| LOW | …/test_shardformer/test_layer/test_sequence_parallel.py | 173 | def test_all_to_all_attention(): |
| LOW | tests/test_shardformer/test_layer/test_linear_1d.py | 121 | def check_linear_without_weight_grad_store(lazy_init: bool, seq_parallel_mode: bool): |
| LOW | tests/test_shardformer/test_layer/test_linear_1d.py | 161 | def check_linear_with_weight_grad_store(lazy_init: bool, seq_parallel_mode: bool): |
| LOW | tests/test_shardformer/test_layer/test_linear_1d.py | 208 | def check_linear_col_plus_row(lazy_init: bool, seq_parallel_mode: bool, overlap: bool): |
| LOW | tests/test_shardformer/test_layer/test_dropout.py | 10 | def check_dropout_parallel_input(): |
| LOW | tests/test_shardformer/test_layer/test_dropout.py | 42 | def check_dropout_replicated_input(): |
| LOW | tests/test_legacy/test_context/test_hybrid_parallel.py | 32 | def check_pipeline_parallel_rank(rank): |
| LOW | tests/test_legacy/test_context/test_hybrid_parallel.py | 44 | def check_model_parallel_rank(rank): |
| LOW | tests/test_legacy/test_context/test_hybrid_parallel.py | 51 | def check_tensor_parallel_rank(rank): |
| LOW | tests/test_legacy/test_utils/test_memory.py | 9 | def _run_colo_set_process_memory_fraction_and_colo_device_memory_capacity(): |
| LOW | …est_legacy/test_utils/test_activation_checkpointing.py | 45 | def test_activation_checkpointing(cpu_offload, use_reentrant): |
| LOW | tests/test_legacy/test_layers/test_cache_embedding.py | 224 | def run_parallel_freq_aware_embed_tablewise(rank, world_size): |
| LOW | tests/test_legacy/test_layers/test_cache_embedding.py | 310 | def run_parallel_freq_aware_embed_columnwise(rank, world_size): |
| LOW | tests/test_legacy/test_layers/test_cache_embedding.py | 389 | def test_parallel_freq_aware_embed(world_size): |
| LOW | …legacy/test_layers/test_1d/checks_1d/check_layer_1d.py | 203 | def check_vocab_parallel_embed(): |
| LOW | …legacy/test_layers/test_1d/checks_1d/check_layer_1d.py | 245 | def check_classifier_no_given_weight(): |
| LOW | …legacy/test_layers/test_1d/checks_1d/check_layer_1d.py | 311 | def check_vocab_parallel_classifier_no_given_weight(): |
| LOW | …legacy/test_layers/test_1d/checks_1d/check_layer_1d.py | 371 | def check_classifier_given_embed_weight(): |
| LOW | …legacy/test_layers/test_1d/checks_1d/check_layer_1d.py | 422 | def check_vocab_parallel_classifier_given_embed_weight(): |
| LOW | …legacy/test_layers/test_1d/checks_1d/check_layer_1d.py | 474 | def check_vocab_parallel_loss(): |
| LOW | …legacy/test_layers/test_1d/checks_1d/check_layer_1d.py | 510 | def check_linear_row_stream_inference(): |
| LOW | tests/test_legacy/test_layers/test_2d/test_2d.py | 51 | def check_layer_and_operation(rank, world_size, port): |
| LOW | …legacy/test_layers/test_2d/checks_2d/check_layer_2d.py | 276 | def check_vocab_parallel_embed(): |
| LOW | …legacy/test_layers/test_2d/checks_2d/check_layer_2d.py | 323 | def check_classifier_no_given_weight(): |
| LOW | …legacy/test_layers/test_2d/checks_2d/check_layer_2d.py | 401 | def check_vocab_parallel_classifier_no_given_weight(): |
| LOW | …legacy/test_layers/test_2d/checks_2d/check_layer_2d.py | 469 | def check_classifier_given_embed_weight(): |
| LOW | …legacy/test_layers/test_2d/checks_2d/check_layer_2d.py | 521 | def check_vocab_parallel_classifier_given_embed_weight(): |
| LOW | …legacy/test_layers/test_2d/checks_2d/check_layer_2d.py | 610 | def check_vocab_parallel_loss(): |
| LOW | tests/test_legacy/test_layers/test_2p5d/test_2p5d.py | 39 | def check_layer_and_operation(rank, world_size, port): |
| LOW | …/test_layers/test_2p5d/checks_2p5d/check_layer_2p5d.py | 280 | def check_vocab_parallel_embed(): |
| LOW | …/test_layers/test_2p5d/checks_2p5d/check_layer_2p5d.py | 328 | def check_classifier_no_given_weight(): |
| LOW | …/test_layers/test_2p5d/checks_2p5d/check_layer_2p5d.py | 407 | def check_vocab_parallel_classifier_no_given_weight(): |
| LOW | …/test_layers/test_2p5d/checks_2p5d/check_layer_2p5d.py | 474 | def check_classifier_given_embed_weight(): |
| LOW | …/test_layers/test_2p5d/checks_2p5d/check_layer_2p5d.py | 526 | def check_vocab_parallel_classifier_given_embed_weight(): |
| LOW | …/test_layers/test_2p5d/checks_2p5d/check_layer_2p5d.py | 615 | def check_vocab_parallel_loss(): |
| LOW | tests/test_legacy/test_layers/test_3d/test_3d.py | 45 | def check_layer_and_operation(rank, world_size, port): |
| LOW | …legacy/test_layers/test_3d/checks_3d/check_layer_3d.py | 206 | def check_classifier_no_given_weight(): |
| LOW | …legacy/test_layers/test_3d/checks_3d/check_layer_3d.py | 306 | def check_vocab_parallel_classifier_no_given_weight(): |
| LOW | …legacy/test_layers/test_3d/checks_3d/check_layer_3d.py | 413 | def check_classifier_given_embed_weight(): |
| LOW | …legacy/test_layers/test_3d/checks_3d/check_layer_3d.py | 497 | def check_vocab_parallel_classifier_given_embed_weight(): |
| LOW | …legacy/test_layers/test_3d/checks_3d/check_layer_3d.py | 747 | def check_vocab_parallel_embed(): |
| 1008 more matches not shown… | |||
| Severity | File | Line | Snippet |
|---|---|---|---|
| LOW | setup.py | 8 | |
| LOW | tests/test_legacy/test_layers/test_1d/test_1d.py | 6 | |
| LOW | tests/test_legacy/test_layers/test_2p5d/test_2p5d.py | 3 | |
| LOW | …/test_layers/test_2p5d/checks_2p5d/check_layer_2p5d.py | 22 | |
| LOW | …t_layers/test_2p5d/checks_2p5d/check_operation_2p5d.py | 9 | |
| LOW | tests/test_legacy/test_tensor/common_utils/__init__.py | 1 | |
| LOW | tests/test_auto_parallel/test_offload/test_perf.py | 18 | |
| LOW | tests/test_auto_parallel/test_offload/test_solver.py | 11 | |
| LOW | tests/kit/model_zoo/__init__.py | 3 | |
| LOW | tests/kit/model_zoo/__init__.py | 3 | |
| LOW | tests/kit/model_zoo/__init__.py | 3 | |
| LOW | tests/kit/model_zoo/__init__.py | 3 | |
| LOW | tests/kit/model_zoo/__init__.py | 3 | |
| LOW | tests/kit/model_zoo/__init__.py | 3 | |
| LOW | tests/kit/model_zoo/__init__.py | 4 | |
| LOW | tests/kit/model_zoo/__init__.py | 4 | |
| LOW | tests/kit/model_zoo/__init__.py | 5 | |
| LOW | tests/kit/model_zoo/torchrec/__init__.py | 1 | |
| LOW | tests/kit/model_zoo/diffusers/__init__.py | 1 | |
| LOW | tests/kit/model_zoo/torchvision/__init__.py | 1 | |
| LOW | tests/kit/model_zoo/timm/__init__.py | 1 | |
| LOW | tests/kit/model_zoo/transformers/__init__.py | 1 | |
| LOW | tests/kit/model_zoo/transformers/__init__.py | 2 | |
| LOW | tests/kit/model_zoo/transformers/__init__.py | 3 | |
| LOW | tests/kit/model_zoo/transformers/__init__.py | 4 | |
| LOW | tests/kit/model_zoo/transformers/__init__.py | 5 | |
| LOW | tests/kit/model_zoo/transformers/__init__.py | 6 | |
| LOW | tests/kit/model_zoo/transformers/__init__.py | 7 | |
| LOW | tests/kit/model_zoo/transformers/__init__.py | 8 | |
| LOW | tests/kit/model_zoo/transformers/__init__.py | 9 | |
| LOW | tests/kit/model_zoo/transformers/__init__.py | 10 | |
| LOW | tests/kit/model_zoo/transformers/__init__.py | 11 | |
| LOW | tests/kit/model_zoo/transformers/__init__.py | 12 | |
| LOW | tests/kit/model_zoo/transformers/__init__.py | 13 | |
| LOW | tests/kit/model_zoo/transformers/__init__.py | 14 | |
| LOW | tests/kit/model_zoo/transformers/__init__.py | 15 | |
| LOW | tests/kit/model_zoo/transformers/__init__.py | 16 | |
| LOW | tests/kit/model_zoo/transformers/__init__.py | 17 | |
| LOW | tests/kit/model_zoo/transformers/__init__.py | 18 | |
| LOW | tests/kit/model_zoo/transformers/__init__.py | 19 | |
| LOW | tests/kit/model_zoo/transformers/__init__.py | 20 | |
| LOW | tests/kit/model_zoo/torchaudio/__init__.py | 1 | |
| LOW | tests/kit/model_zoo/custom/__init__.py | 1 | |
| LOW | tests/kit/model_zoo/custom/__init__.py | 2 | |
| LOW | tests/kit/model_zoo/custom/__init__.py | 3 | |
| LOW | tests/kit/model_zoo/custom/__init__.py | 4 | |
| LOW | tests/kit/model_zoo/custom/__init__.py | 5 | |
| LOW | …/test_infer/test_kernels/triton/test_rmsnorm_triton.py | 10 | |
| LOW | tests/test_infer/test_kernels/triton/test_xine_copy.py | 8 | |
| LOW | …nfer/test_kernels/triton/test_rotary_embdding_unpad.py | 13 | |
| LOW | …s/test_infer/test_kernels/triton/test_decoding_attn.py | 19 | |
| LOW | …ts/test_infer/test_kernels/triton/test_kvcache_copy.py | 14 | |
| LOW | …t_infer/test_kernels/triton/test_context_attn_unpad.py | 15 | |
| LOW | …fer/test_kernels/triton/test_fused_rotary_embedding.py | 12 | |
| LOW | extensions/pybind/optimizer/__init__.py | 1 | |
| LOW | extensions/pybind/layernorm/__init__.py | 1 | |
| LOW | extensions/pybind/cpu_adam/__init__.py | 1 | |
| LOW | extensions/pybind/cpu_adam/__init__.py | 2 | |
| LOW | extensions/pybind/softmax/__init__.py | 1 | |
| LOW | extensions/pybind/softmax/__init__.py | 2 | |
| 900 more matches not shown… | |||
| Severity | File | Line | Snippet |
|---|---|---|---|
| LOW | tests/test_checkpoint_io/test_safetensors_async_io.py | 118 | |
| LOW | …ts/test_checkpoint_io/test_torch_fsdp_checkpoint_io.py | 18 | |
| LOW | …/test_shardformer/test_model/test_shard_deepseek_v3.py | 26 | |
| LOW | tests/test_legacy/test_comm/test_object_list_p2p_v2.py | 85 | |
| LOW | tests/test_lora/test_lora.py | 20 | |
| LOW | …arallel/test_ckpt_solvers/test_C_solver_consistency.py | 29 | |
| LOW | …test_auto_parallel/test_ckpt_solvers/test_linearize.py | 32 | |
| LOW | …test_auto_parallel/test_ckpt_solvers/test_linearize.py | 96 | |
| LOW | …_tensor_shard/test_node_handler/test_matmul_handler.py | 50 | |
| LOW | …_parallel/test_tensor_shard/test_node_handler/utils.py | 66 | |
| LOW | tests/kit/model_zoo/registry.py | 64 | |
| LOW | tests/test_pipeline/test_schedule/test_zerobubble_pp.py | 114 | |
| LOW | tests/test_tensor/test_dtensor/test_dtensor.py | 22 | |
| LOW | tests/test_optimizer/test_dist_came.py | 63 | |
| LOW | tests/test_optimizer/test_dist_adafactor.py | 196 | |
| LOW | tests/test_optimizer/_utils.py | 230 | |
| LOW | tests/test_moe/test_moe_checkpoint.py | 42 | |
| LOW | tests/test_moe/test_moe_checkpoint.py | 90 | |
| LOW | …ochunk_transformer/test_autochunk_transformer_utils.py | 81 | |
| LOW | tests/test_fp8/test_fp8_fsdp_comm_hook.py | 33 | |
| LOW | extensions/utils.py | 154 | |
| LOW | colossalai/pipeline/weight_grad_store.py | 19 | |
| LOW | colossalai/pipeline/schedule/v_schedule.py | 81 | |
| LOW | colossalai/pipeline/schedule/v_schedule.py | 290 | |
| LOW | colossalai/pipeline/schedule/v_schedule.py | 318 | |
| LOW | colossalai/pipeline/schedule/generate.py | 203 | |
| LOW | colossalai/pipeline/schedule/generate.py | 348 | |
| LOW | colossalai/pipeline/schedule/zero_bubble_pp.py | 239 | |
| LOW | colossalai/pipeline/schedule/zero_bubble_pp.py | 297 | |
| LOW | colossalai/pipeline/schedule/zero_bubble_pp.py | 871 | |
| LOW | colossalai/interface/model.py | 9 | |
| LOW | colossalai/nn/optimizer/adafactor.py | 86 | |
| LOW | colossalai/nn/optimizer/lars.py | 40 | |
| LOW | colossalai/nn/optimizer/galore.py | 118 | |
| LOW | colossalai/nn/optimizer/galore.py | 235 | |
| LOW | colossalai/nn/optimizer/distributed_adafactor.py | 286 | |
| LOW | colossalai/nn/optimizer/came.py | 64 | |
| LOW | colossalai/nn/optimizer/cpu_adam.py | 127 | |
| LOW | colossalai/nn/optimizer/distributed_lamb.py | 104 | |
| LOW | colossalai/nn/optimizer/distributed_galore.py | 143 | |
| LOW | colossalai/nn/optimizer/fused_lamb.py | 105 | |
| LOW | colossalai/nn/optimizer/distributed_came.py | 68 | |
| LOW | colossalai/nn/optimizer/distributed_came.py | 333 | |
| LOW | colossalai/nn/optimizer/hybrid_adam.py | 93 | |
| LOW | colossalai/nn/layer/scaled_softmax.py | 132 | |
| LOW | colossalai/fx/_meta_regist_12.py | 435 | |
| LOW | colossalai/fx/tracer/experimental.py | 497 | |
| LOW | colossalai/fx/tracer/experimental.py | 525 | |
| LOW | colossalai/fx/tracer/experimental.py | 595 | |
| LOW | colossalai/fx/tracer/experimental.py | 201 | |
| LOW | colossalai/fx/tracer/experimental.py | 334 | |
| LOW | colossalai/fx/tracer/experimental.py | 366 | |
| LOW | colossalai/fx/tracer/experimental.py | 370 | |
| LOW | colossalai/fx/tracer/tracer.py | 83 | |
| LOW | colossalai/fx/tracer/tracer.py | 152 | |
| LOW | colossalai/fx/tracer/tracer.py | 218 | |
| LOW | colossalai/fx/tracer/tracer.py | 317 | |
| LOW | colossalai/fx/tracer/tracer.py | 157 | |
| LOW | …ai/fx/tracer/meta_patch/patched_function/arithmetic.py | 8 | |
| LOW | …lai/fx/tracer/meta_patch/patched_function/torch_ops.py | 146 | |
| 485 more matches not shown… | |||
| Severity | File | Line | Snippet |
|---|---|---|---|
| MEDIUM | …t_hybrid_parallel_grad_clip_norm/test_amp_optimizer.py | 74 | # Create a list of parameter-gradient pairs containing working parameters and their gradients |
| MEDIUM | …est_tensor_shard/test_metainfo/test_linear_metainfo.py | 23 | """This function is for linear memory test |
| MEDIUM | …est_tensor_shard/test_metainfo/test_linear_metainfo.py | 61 | """This function is for linear memory test |
| MEDIUM | …st_tensor_shard/test_metainfo/test_pooling_metainfo.py | 14 | """This function is for AdaptiveAvgPool memory test |
| MEDIUM | …st_tensor_shard/test_metainfo/test_pooling_metainfo.py | 55 | """This function is for MaxPool memory test |
| MEDIUM | …/test_tensor_shard/test_metainfo/test_conv_metainfo.py | 23 | """This function is for conv memory test |
| MEDIUM | …/test_tensor_shard/test_metainfo/test_conv_metainfo.py | 65 | """This function is for conv function memory test |
| MEDIUM | …hard/test_metainfo/test_binary_elementwise_metainfo.py | 24 | """This function is for binary elementwise ops memory test |
| MEDIUM | …/test_tensor_shard/test_metainfo/test_norm_metainfo.py | 18 | """This function is for batchnorm memory test |
| MEDIUM | tests/test_infer/test_streamingllm.py | 115 | result_list = manager.list([-1] * 1) # Create a shared list |
| MEDIUM | tests/test_infer/test_inference_engine.py | 90 | result_list = manager.list([-1] * world_size) # Create a shared list |
| MEDIUM | tests/test_infer/test_models/test_baichuan.py | 82 | result_list = manager.list([-1] * world_size) # Create a shared list |
| MEDIUM | tests/test_infer/test_models/test_custom_model.py | 84 | result_list = manager.list([-1] * world_size) # Create a shared list |
| MEDIUM | tests/test_infer/test_kernels/triton/kernel_utils.py | 7 | # This function is adapted from src/transformers/models/llama/modeling_llama.py |
| MEDIUM | colossalai/fx/codegen/activation_checkpoint_codegen.py | 168 | """This function is to find the offload regions |
| MEDIUM | colossalai/booster/plugin/hybrid_parallel_plugin.py | 1403 | # Create a context for gradient synchronization based on the optimizer type. |
| MEDIUM | colossalai/shardformer/modeling/gpt2.py | 137 | # This function is modified on the basis of transformers.models.gpt2.modeling_gpt2.GPT2Model.forward. |
| MEDIUM | colossalai/shardformer/modeling/gpt2.py | 555 | # This function is modified on the basis of transformers.models.gpt2.modeling_gpt2.GPT2ForQuestionAnswering.forw |
| MEDIUM | colossalai/shardformer/modeling/gpt2.py | 643 | # This function is modified on the basis of transformers.models.gpt2.modeling_gpt2.GPT2ForTokenClassification.fo |
| MEDIUM | colossalai/shardformer/modeling/gpt2.py | 718 | # This function is modified on the basis of transformers.models.gpt2.modeling_gpt2.GPT2ForSequenceClassification |
| MEDIUM | colossalai/shardformer/modeling/t5.py | 54 | # This function is modified on the basis of transformers.models.t5.modeling_t5.T5Stack.forward. |
| MEDIUM | colossalai/shardformer/modeling/t5.py | 290 | # This function is modified on the basis of transformers.models.t5.modeling_t5.T5Model.forward. |
| MEDIUM | colossalai/shardformer/modeling/t5.py | 430 | # This function is modified on the basis of transformers.models.t5.modeling_t5.T5ForConditionalGeneration.forwar |
| MEDIUM | colossalai/shardformer/modeling/gptj.py | 99 | # This function is modified on the v4.51.3 transformers.models.gptj.modeling_gptj.GPTJModel.forward. |
| MEDIUM | colossalai/shardformer/modeling/gptj.py | 288 | # This function is modified on the basis of transformers.models.gptj.modeling_gptj.GPTJForCausalLM.forward. |
| MEDIUM | colossalai/shardformer/modeling/gptj.py | 381 | # This function is modified on the v4.51.3 transformers.models.gptj.modeling_gptj.GPTJForSequenceClassification. |
| MEDIUM | colossalai/shardformer/modeling/gptj.py | 503 | # This function is modified on the basis of transformers.models.gptj.modeling_gptj.GPTJForQuestionAnswering.forw |
| MEDIUM | colossalai/shardformer/modeling/gptj.py | 583 | # This function is modified on the v4.51.3 transformers.models.gptj.modeling_gptj.GPTJAttention.forward. |
| MEDIUM | colossalai/legacy/pipeline/middleware/topo.py | 127 | # This class is a middleware between partition splitter |
| MEDIUM | colossalai/legacy/context/parallel_context.py | 25 | """This class provides interface functions for users to get the parallel context, |
| MEDIUM | colossalai/legacy/context/random/seed_manager.py | 11 | """This class is a manager of all random seeds involved in the system. |
| MEDIUM | colossalai/legacy/nn/loss/loss_1d.py | 30 | # Create a mask of valid vocab ids (1 means it needs to be masked). |
| MEDIUM | …lossalai/legacy/zero/sharded_optim/sharded_optim_v2.py | 189 | # This function is called except the last stage of pipeline parallel |
| MEDIUM | colossalai/legacy/amp/naive_amp/_fp16_optimizer.py | 128 | # Create a fp32 copy |
| MEDIUM | colossalai/utils/tensor_detector/tensor_detector.py | 17 | """This class is a detector to detect tensor on different devices. |
| MEDIUM | colossalai/quantization/bnb.py | 219 | # Create a copy of the model |
| MEDIUM | colossalai/zero/gemini/gemini_ddp.py | 261 | """This function is only triggered for inference.""" |
| MEDIUM | colossalai/zero/gemini/gemini_ddp.py | 296 | """This function is only triggered for inference.""" |
| MEDIUM | colossalai/zero/gemini/gemini_optimizer.py | 306 | # This function is called except the last stage of pipeline parallel |
| MEDIUM | …ssalai/zero/gemini/memory_tracer/memstats_collector.py | 76 | # The following code work for ZeroInitContext, which is deprecated in v0.1.12 |
| MEDIUM | colossalai/lazy/lazy_init.py | 80 | """This class is only for correctness verification.""" |
| MEDIUM | colossalai/inference/batch_bucket.py | 483 | ########## The following methods are expected to be used in modeling ########### |
| MEDIUM | colossalai/tensor/comm_spec.py | 423 | # Create a new member `logical_process_axes` to distinguish from original flatten |
| MEDIUM | …/tutorial/sequence_parallel/loss_func/cross_entropy.py | 15 | # Create a mask of valid vocab ids (1 means it needs to be masked). |
| MEDIUM | …ce_parallel/data/datasets/test/test_indexed_dataset.py | 1 | # This file isn't really a formal automated test, it's just a place to |
| MEDIUM | examples/images/diffusion/main.py | 56 | # Create an ArgumentParser object with specifies kwargs |
| MEDIUM | examples/images/diffusion/main.py | 404 | # Create a list of exponentially increasing log steps, starting from 1 and ending at batch_frequency |
| MEDIUM | examples/images/diffusion/main.py | 661 | # Create the checkpoint and configuration directories within the log directory. |
| MEDIUM | examples/images/diffusion/main.py | 755 | # Create an empty OmegaConf configuration object |
| MEDIUM | examples/images/diffusion/main.py | 796 | # Create a Trainer object with the specified command-line arguments and keyword arguments, and set the log direc |
| MEDIUM | examples/images/diffusion/main.py | 800 | # Create a data module based on the configuration file |
| MEDIUM | examples/images/diffusion/ldm/data/lsun.py | 10 | # This class is used to create a dataset of images from LSUN dataset for training |
| MEDIUM | examples/images/dreambooth/train_dreambooth.py | 689 | # Create the pipeline using using the trained modules and save it. |
| MEDIUM | examples/images/dreambooth/train_dreambooth_inpaint.py | 726 | # Create the pipeline using using the trained modules and save it. |
| MEDIUM | examples/language/gpt/titans/model/embed.py | 262 | # Create a mask of valid vocab ids (1 means it needs to be masked). |
| MEDIUM | examples/language/gpt/gemini/commons/model_zoo.py | 5 | ## Define the Model and Loss Based on Huggingface transformers GPT2LMHeadModel |
| MEDIUM | …anguage/gpt/experiments/pipeline_parallel/model_zoo.py | 5 | ## Define the Model and Loss Based on Huggingface transformers GPT2LMHeadModel |
| MEDIUM | …ples/community/roberta/pretraining/model/deberta_v2.py | 102 | >>> # Create a mask |
| MEDIUM | …ons/ColossalQA/colossalqa/retrieval_conversation_en.py | 31 | # Define the retriever |
| MEDIUM | …ons/ColossalQA/colossalqa/retrieval_conversation_en.py | 34 | # Define the chain to preprocess the input |
| 15 more matches not shown… | |||
| Severity | File | Line | Snippet |
|---|---|---|---|
| HIGH | colossalai/pipeline/schedule/_utils.py | 72 | Get the batch size (size of dimension-0) of the first tensor in the batch. Args: batch (Any): Batch to be i |
| HIGH | colossalai/context/config.py | 52 | Reads a python file and constructs a corresponding :class:`Config` object. Args: filename (str): Na |
| HIGH | colossalai/shardformer/layer/normalization.py | 140 | Convert a native PyTorch layer normalization module to a specific layer normalization module, and optio |
| HIGH | colossalai/legacy/context/parallel_context.py | 130 | Returns the local rank of the current device. Args: parallel_mode (:class:`colossalai.legacy.contex |
| HIGH | colossalai/legacy/context/parallel_context.py | 160 | Returns the global rank of the next device. Args: parallel_mode (:class:`colossalai.legacy.context. |
| HIGH | colossalai/legacy/context/parallel_context.py | 182 | Returns the global rank of the previous device. Args: parallel_mode (:class:`colossalai.legacy.cont |
| HIGH | colossalai/legacy/context/parallel_context.py | 204 | Returns a boolean value indicating whether the current device is the first one among its group for `parallel_mod |
| HIGH | colossalai/legacy/context/parallel_context.py | 222 | Returns a boolean value indicating whether the current device is the last one among its group for `parallel_mode |
| HIGH | colossalai/legacy/context/parallel_context.py | 256 | Returns the world size for `parallel_mode`. Args: parallel_mode (:class:`colossalai.legacy.context. |
| HIGH | colossalai/legacy/context/parallel_context.py | 286 | Returns the group of the current device for `parallel_mode`. Args: parallel_mode (:class:`colossala |
| HIGH | colossalai/legacy/context/parallel_context.py | 342 | Returns the rank of the current device for `parallel_mode` in the group. Args: parallel_mode (:clas |
| HIGH | colossalai/legacy/utils/checkpointing.py | 221 | Loads training states from a checkpoint file. Args: file: a file-like object (has to implement read(), read |
| HIGH | …lossalai/legacy/zero/gemini/tensor_placement_policy.py | 64 | Evict tensors from CUDA device. Args: hold_cuda_tensor_list (List[StatefulTensor]): the li |
| HIGH | colossalai/legacy/registry/registry.py | 28 | Registers a module represented in `module_class`. Args: module_class (class): The module to be regi |
| HIGH | colossalai/legacy/registry/registry.py | 45 | Retrieves a module with name `module_name` and returns the module if it has already been registered before. |
| HIGH | colossalai/legacy/builder/builder.py | 10 | Returns an object of :class:`module` constructed from `config`. Args: module: A python or user-defined clas |
| HIGH | colossalai/legacy/builder/builder.py | 27 | Returns an object constructed from `config`, the type of the object is specified by `registry`. Note: t |
| HIGH | colossalai/cli/launcher/run.py | 59 | Parse an inclusion or exclusion string and filter a hostfile dictionary. Examples: include_str="worker-0,wo |
| HIGH | colossalai/_analyzer/fx/tracer/symbolic_trace.py | 37 | Traces a ``torch.nn.Module`` or a function and returns a ``GraphModule`` with ``Node``s and ``MetaInfo`` attach |
| HIGH | colossalai/_analyzer/fx/passes/graph_profile.py | 233 | Run a specific node ``n`` and profile its execution time and memory usage. Calls into call_function, ca |
| HIGH | …lossalai/auto_parallel/checkpoint/ckpt_solver_rotor.py | 311 | Backtrack the cost table and retrieve the optimal checkpointing strategy. Args: chain (Chain): A ba |
| HIGH | colossalai/zero/gemini/placement_policy.py | 158 | Evict tensors from CUDA device. Args: can_evict_chunks (List[StatefulTensor]): the list of |
| HIGH | colossalai/tensor/d_tensor/layout_converter.py | 453 | Check if the process groups required for sharding have been deleted by torch.distributed.destroy_proces |
| HIGH | …/ColossalChat/coati/distributed/reward/reward_utils.py | 80 | Modified from: https://gist.github.com/lewtun/9c2ce1937b741404090a3dc4c7c022b3 Retrieves the content from the l |
| HIGH | applications/ColossalChat/coati/models/generation.py | 279 | Generates new tokens using a streaming approach. Args: model (Any): The model used for token generatio |
| Severity | File | Line | Snippet |
|---|---|---|---|
| LOW | …est_checkpoint_io/test_low_level_zero_checkpoint_io.py | 149 | except Exception as e: |
| MEDIUM | …est_checkpoint_io/test_low_level_zero_checkpoint_io.py | 95 | def run_fn(stage, shard, offload, model_fn, data_gen_fn, output_transform_fn, lo |
| LOW | tests/test_shardformer/test_model/test_shard_gpt2.py | 226 | except Exception as e: |
| LOW | tests/test_shardformer/test_model/test_shard_gpt2.py | 271 | except Exception as e: |
| LOW | tests/test_shardformer/test_model/test_shard_command.py | 286 | except Exception as e: |
| LOW | tests/test_shardformer/test_model/test_shard_llama.py | 81 | except Exception as e: |
| LOW | tests/test_shardformer/test_model/test_shard_llama.py | 305 | except Exception as e: |
| LOW | tests/test_shardformer/test_model/test_shard_llama.py | 359 | except Exception as e: |
| LOW | …sts/test_shardformer/test_model/test_shard_chatglm2.py | 241 | except Exception as e: |
| LOW | tests/test_shardformer/test_model/test_shard_qwen3.py | 214 | except Exception as e: |
| LOW | tests/test_shardformer/test_model/test_shard_qwen3.py | 263 | except Exception as e: |
| LOW | tests/test_shardformer/test_model/test_shard_qwen2.py | 218 | except Exception as e: |
| LOW | tests/test_shardformer/test_model/test_shard_qwen2.py | 267 | except Exception as e: |
| LOW | tests/test_booster/test_plugin/test_gemini_plugin.py | 62 | except Exception as e: |
| MEDIUM | tests/test_booster/test_plugin/test_gemini_plugin.py | 19 | def run_fn(init_method, model_fn, data_gen_fn, output_transform_fn, zero_size, t |
| LOW | tests/test_booster/test_plugin/test_3d_plugin.py | 82 | except Exception as e: |
| MEDIUM | tests/test_booster/test_plugin/test_3d_plugin.py | 48 | def run_fn(init_method, model_fn, data_gen_fn, output_transform_fn) -> Optional[ |
| LOW | …test_booster/test_plugin/test_low_level_zero_plugin.py | 56 | except Exception as e: |
| LOW | tests/test_fx/test_tracer/test_patched_op.py | 13 | except Exception as e: |
| MEDIUM | tests/test_fx/test_tracer/test_patched_op.py | 9 | def _run(data, patch_fn): |
| LOW | tests/test_fx/test_tracer/test_patched_module.py | 16 | except Exception as e: |
| MEDIUM | tests/test_fx/test_tracer/test_patched_module.py | 7 | def _run(data, module, patch_fn): |
| LOW | …racer/test_torchvision_model/test_torchvision_model.py | 41 | except Exception as e: |
| LOW | …s/test_fx/test_tracer/test_hf_model/hf_tracer_utils.py | 23 | except Exception as e: |
| LOW | …ts/test_fx/test_pipeline/test_timm_model/timm_utils.py | 29 | except Exception as e: |
| LOW | tests/test_fx/test_pipeline/test_topo/topo_utils.py | 45 | except Exception as e: |
| LOW | tests/test_fx/test_pipeline/test_hf_model/hf_utils.py | 32 | except Exception as e: |
| LOW | tests/test_optimizer/test_dist_lamb.py | 148 | except Exception as e: |
| LOW | tests/test_optimizer/test_dist_lamb.py | 230 | except Exception as e: |
| LOW | tests/test_optimizer/test_dist_lamb.py | 250 | except Exception as e: |
| LOW | tests/test_optimizer/test_dist_galore.py | 80 | except Exception as e: |
| LOW | tests/test_optimizer/test_dist_galore.py | 171 | except Exception as e: |
| LOW | tests/test_optimizer/test_dist_galore.py | 245 | except Exception as e: |
| LOW | tests/test_optimizer/test_dist_galore.py | 264 | except Exception as e: |
| LOW | tests/test_optimizer/test_dist_galore.py | 289 | except Exception as e: |
| LOW | …est_autochunk_diffuser/benchmark_autochunk_diffuser.py | 134 | except Exception as e: |
| LOW | …t_autochunk_alphafold/benchmark_autochunk_alphafold.py | 123 | except Exception as e: |
| LOW | …tochunk_transformer/benchmark_autochunk_transformer.py | 135 | except Exception as e: |
| LOW | colossalai/fx/tracer/experimental.py | 589 | except Exception as e: |
| LOW | colossalai/fx/tracer/tracer.py | 312 | except Exception as e: |
| LOW | colossalai/shardformer/layer/parallel_module.py | 152 | except Exception as ex: |
| LOW | colossalai/shardformer/layer/parallel_module.py | 330 | except Exception as ex: |
| LOW | colossalai/shardformer/layer/normalization.py | 22 | except Exception: |
| LOW | colossalai/shardformer/shard/sharder.py | 200 | except Exception as e: |
| LOW | …lossalai/legacy/zero/sharded_model/sharded_model_v2.py | 208 | except Exception as e: |
| LOW | …lossalai/legacy/zero/sharded_model/sharded_model_v2.py | 556 | except Exception as ex: |
| LOW | colossalai/legacy/inference/async_manager.py | 148 | except Exception: |
| MEDIUM | colossalai/legacy/inference/async_manager.py | 135 | def start_dynamic_batching(args, tp_engine, waiting_req_list): |
| LOW | colossalai/legacy/inference/manager.py | 293 | except Exception: |
| MEDIUM | colossalai/legacy/inference/manager.py | 280 | def start_dynamic_batching(args, tp_engine, waiting_req_list): |
| LOW | …egacy/inference/quant/smoothquant/models/base_model.py | 228 | except Exception as e: |
| LOW | colossalai/legacy/builder/builder.py | 55 | except Exception as e: |
| MEDIUM | colossalai/legacy/builder/builder.py | 56 | print(f"An error occurred when building {mod_type} from registry {registry.name}", flush=True) |
| LOW | colossalai/utils/safetensors.py | 12 | except Exception: |
| LOW | colossalai/cli/launcher/multinode_runner.py | 55 | except Exception as e: |
| LOW | colossalai/_analyzer/fx/graph_module.py | 71 | except Exception as e: |
| MEDIUM | colossalai/_analyzer/fx/graph_module.py | 65 | def __call__(self, obj, *args, **kwargs): |
| LOW | colossalai/_analyzer/fx/passes/graph_profile.py | 257 | except Exception as e: |
| LOW | colossalai/zero/gemini/gemini_ddp.py | 727 | except Exception as ex: |
| LOW | colossalai/lazy/pretrained.py | 259 | except Exception: |
| 51 more matches not shown… | |||
| Severity | File | Line | Snippet |
|---|---|---|---|
| LOW | …legacy/test_layers/test_2d/checks_2d/check_layer_2d.py | 641 | out_grad = torch.chunk(out_grad, DEPTH, dim=0)[i] |
| LOW | …legacy/test_layers/test_2d/checks_2d/check_layer_2d.py | 661 | # ) |
| LOW | …legacy/test_layers/test_2d/checks_2d/check_layer_2d.py | 681 | # out.backward(grad) |
| LOW | …legacy/test_layers/test_2d/checks_2d/check_layer_2d.py | 701 | # torch.distributed.broadcast(A_master, src=0) |
| LOW | …legacy/test_layers/test_2d/checks_2d/check_layer_2d.py | 721 | # INPUT_SIZE = HIDDEN_SIZE |
| LOW | …legacy/test_layers/test_2d/checks_2d/check_layer_2d.py | 741 | # mask_shape = (BATCH_SIZE // DEPTH, NUM_ATTENTION_HEADS // DEPTH, SEQ_LENGTH, SEQ_LENGTH) |
| LOW | …/test_layers/test_2p5d/checks_2p5d/check_layer_2p5d.py | 661 | |
| LOW | …/test_layers/test_2p5d/checks_2p5d/check_layer_2p5d.py | 681 | # assert out.shape == (BATCH_SIZE // TESSERACT_DIM, SEQ_LENGTH, INPUT_SIZE // TESSERACT_DIM) |
| LOW | …/test_layers/test_2p5d/checks_2p5d/check_layer_2p5d.py | 701 | # HIDDEN_SIZE, |
| LOW | …/test_layers/test_2p5d/checks_2p5d/check_layer_2p5d.py | 721 | # grad = torch.randn(grad_shape, dtype=dtype, device=device) |
| LOW | …/test_layers/test_2p5d/checks_2p5d/check_layer_2p5d.py | 741 | # attention_dropout_prob=0.5, |
| LOW | …/test_layers/test_2p5d/checks_2p5d/check_layer_2p5d.py | 761 | # grad_shape = out.shape |
| LOW | …rallel/test_tensor_shard/test_solver_with_resnet_v2.py | 21 | mesh_shape = (2, 4) |
| LOW | …rallel/test_tensor_shard/test_solver_with_resnet_v2.py | 41 | # %layer1_0_conv2 : [#users=1] = call_module[target=layer1.0.conv2](args = (%layer1_0_relu,), kwargs = {}) |
| LOW | tests/kit/model_zoo/transformers/mistral.py | 1 | import torch |
| LOW | tests/kit/model_zoo/transformers/whisper.py | 1 | import torch |
| LOW | tests/kit/model_zoo/transformers/deepseek.py | 1 | # modified from tests/kit/model_zoo/transformers/mistral.py |
| LOW | tests/kit/model_zoo/transformers/bert.py | 1 | import torch |
| LOW | tests/kit/model_zoo/transformers/deepseek_v3.py | 1 | # modified from tests/kit/model_zoo/transformers/mistral.py |
| LOW | tests/kit/model_zoo/transformers/sam.py | 1 | import torch |
| LOW | tests/kit/model_zoo/transformers/mixtral.py | 1 | # modified from tests/kit/model_zoo/transformers/mistral.py |
| LOW | tests/kit/model_zoo/transformers/mixtral.py | 81 | # name="transformers_mixtral_for_sequence_classification", |
| LOW | tests/kit/model_zoo/transformers/blip2.py | 1 | import torch |
| LOW | tests/test_tensor/test_comm_spec_apply.py | 81 | if rank in (0, 1): |
| LOW | tests/test_tensor/test_dtensor/test_comm_spec.py | 61 | sharded_tensor_1 = torch.ones(2, 1) * 3 |
| LOW | tests/test_fx/test_tracer/test_bias_addition_module.py | 41 | # %linear : [#users=1] = call_function[target=torch._C._nn.linear](args = (%x, %linear_weight), kwargs = {}) |
| LOW | …/test_fx/test_profiler/test_profiler_meta_info_prop.py | 61 | torch.cuda.reset_peak_memory_stats() |
| LOW | …/test_fx/test_profiler/test_profiler_meta_info_prop.py | 101 | # If we need to dive deep into the memory usage by |
| LOW | …e/en/features/mixed_precision_training_with_booster.md | 221 | ```python |
| LOW | extensions/csrc/common/micros.h | 101 | // https://github.com/pytorch/pytorch/commit/3aeb78079bcd68282fe9117088e138b77318e288 |
| LOW | extensions/csrc/common/micros.h | 221 | } else { \ |
| LOW | extensions/csrc/funcs/cast_functor.h | 1 | #pragma once |
| LOW | extensions/csrc/funcs/binary_functor.h | 1 | #pragma once |
| LOW | extensions/csrc/funcs/ternary_functor.h | 1 | #pragma once |
| LOW | extensions/csrc/funcs/unary_functor.h | 1 | #pragma once |
| LOW | extensions/csrc/kernel/cuda/attention/attention_utils.h | 21 | |
| LOW | extensions/csrc/kernel/x86/cpu_adam.h | 21 | */ |
| LOW | extensions/csrc/kernel/x86/cpu_adam.h | 41 | #define SIMD_WIDTH 16 |
| LOW | extensions/csrc/kernel/x86/cpu_adam.h | 61 | #define SIMD_LOAD(x) _mm256_loadu_ps(x) |
| LOW | colossalai/pipeline/schedule/v_schedule.py | 1 | # Refer from Zero Bubble Pipeline Parallelism. |
| LOW | colossalai/nn/optimizer/adafactor.py | 1 | # coding=utf-8 |
| LOW | colossalai/shardformer/modeling/bert.py | 1081 | # if uni-directional self-attention (decoder) save Tuple(torch.Tensor, torch.Tensor) of |
| LOW | colossalai/legacy/pipeline/middleware/topo.py | 121 | return res |
| LOW | colossalai/legacy/moe/openmoe/train.sh | 21 | # ep |
| LOW | colossalai/legacy/moe/openmoe/test_ci.sh | 1 | # pip install -r requirements.txt |
| LOW | colossalai/legacy/moe/openmoe/test_ci.sh | 21 | # torchrun --standalone --nproc_per_node 4 train.py \ |
| LOW | …salai/legacy/moe/openmoe/model/convert_openmoe_ckpt.py | 1 | # coding=utf-8 |
| LOW | colossalai/legacy/moe/openmoe/model/modeling_openmoe.py | 1 | # coding=utf-8 |
| LOW | colossalai/_analyzer/fx/graph_module.py | 21 | from torch.fx.graph_module import _exec_with_source, _forward_from_src |
| LOW | colossalai/auto_parallel/checkpoint/ckpt_solver_base.py | 101 | """ |
| LOW | …salai/auto_parallel/passes/runtime_preparation_pass.py | 241 | getitem_index = node.args[1] |
| LOW | colossalai/device/device_mesh.py | 421 | # key is axis name |
| LOW | colossalai/tensor/colo_parameter.py | 81 | |
| LOW | …utorial/sequence_parallel/lr_scheduler/annealing_lr.py | 1 | # coding=utf-8 |
| LOW | examples/tutorial/sequence_parallel/data/__init__.py | 21 | (train_dataloader, valid_dataloader, test_dataloader) = (None, None, None) |
| LOW | …/sequence_parallel/data/tokenizer/bert_tokenization.py | 1 | # coding=utf-8 |
| LOW | …/tutorial/sequence_parallel/data/tokenizer/__init__.py | 1 | # coding=utf-8 |
| LOW | …tutorial/sequence_parallel/data/tokenizer/tokenizer.py | 1 | # coding=utf-8 |
| LOW | …l/sequence_parallel/data/datasets/blendable_dataset.py | 1 | # coding=utf-8 |
| LOW | …torial/sequence_parallel/data/datasets/bert_dataset.py | 1 | # coding=utf-8 |
| 26 more matches not shown… | |||
| Severity | File | Line | Snippet |
|---|---|---|---|
| LOW | applications/ColossalQA/data/tests/test.txt | 28 | * Lorem ipsum dolor sit amet. |
| LOW | applications/ColossalQA/data/tests/test.txt | 28 | * Lorem ipsum dolor sit amet. |
| LOW | applications/ColossalQA/data/tests/test.txt | 2 | Lorem ipsum dolor sit amet, consectetuer adipiscing elit |
| LOW | applications/ColossalQA/data/tests/test.txt | 2 | Lorem ipsum dolor sit amet, consectetuer adipiscing elit |
| LOW | applications/ColossalQA/data/tests/test.txt | 10 | Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh. |
| LOW | applications/ColossalQA/data/tests/test.txt | 10 | Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh. |
| LOW | applications/ColossalQA/data/tests/test.txt | 13 | Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh. |
| LOW | applications/ColossalQA/data/tests/test.txt | 13 | Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh. |
| LOW | applications/ColossalQA/data/tests/test.txt | 16 | Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh. |
| LOW | applications/ColossalQA/data/tests/test.txt | 16 | Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh. |
| LOW | applications/ColossalQA/data/tests/test.txt | 20 | Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh euismod tincidunt ut laoreet dolore. |
| LOW | applications/ColossalQA/data/tests/test.txt | 20 | Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh euismod tincidunt ut laoreet dolore. |
| LOW | applications/ColossalQA/data/tests/test.txt | 23 | Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam. |
| LOW | applications/ColossalQA/data/tests/test.txt | 23 | Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam. |
| LOW | applications/ColossalQA/data/tests/test.txt | 26 | Lorem ipsum dolor sit amet, consectetuer adipiscing elit. |
| LOW | applications/ColossalQA/data/tests/test.txt | 26 | Lorem ipsum dolor sit amet, consectetuer adipiscing elit. |
| LOW | applications/ColossalQA/data/tests/test.txt | 33 | Lorem ipsum dolor sit amet Consectetuer adipiscing elit, Sed diam nonummy |
| LOW | applications/ColossalQA/data/tests/test.txt | 33 | Lorem ipsum dolor sit amet Consectetuer adipiscing elit, Sed diam nonummy |
| LOW | applications/ColossalQA/data/tests/test.txt | 35 | Lorem ipsum dolor sit amet Consectetuer adipiscing elit, Sed diam nonummy |
| LOW | applications/ColossalQA/data/tests/test.txt | 35 | Lorem ipsum dolor sit amet Consectetuer adipiscing elit, Sed diam nonummy |
| LOW | applications/ColossalQA/data/tests/test.txt | 38 | Lorem ipsum, Dolor sit amet, Consectetuer |
| LOW | applications/ColossalQA/data/tests/test.txt | 38 | Lorem ipsum, Dolor sit amet, Consectetuer |
| Severity | File | Line | Snippet |
|---|---|---|---|
| LOW | tests/test_shardformer/test_model/_utils.py | 324 | # Check if the output sequence is gathered before cross entropy |
| LOW | colossalai/booster/plugin/hybrid_parallel_plugin.py | 173 | # Check if the DP group size is 1, meaning no synchronization is needed. |
| LOW | colossalai/booster/plugin/hybrid_parallel_plugin.py | 854 | # Check if the list of gradients is empty |
| LOW | colossalai/booster/plugin/hybrid_parallel_plugin.py | 1427 | # Check if the optimizer is a HybridParallelZeroOptimizer and synchronize data parallelism gradients if so. |
| LOW | colossalai/quantization/bnb.py | 160 | # Check if the current key is not in the `modules_to_not_convert` |
| LOW | colossalai/quantization/bnb.py | 232 | # Check if it is a base model |
| LOW | colossalai/tensor/d_tensor/layout_converter.py | 472 | # Check if each process group is still alive |
| LOW | examples/images/diffusion/main.py | 257 | # Check if the train dataset is iterable |
| LOW | examples/images/diffusion/main.py | 274 | # Check if the validation dataset is iterable |
| LOW | examples/images/diffusion/main.py | 289 | # Check if the test dataset is iterable |
| LOW | examples/images/diffusion/main.py | 620 | # Check if the "resume" option is specified, resume training from the checkpoint if it is true |
| LOW | examples/images/diffusion/main.py | 854 | # Assign melk to SIGUSR1 signal and divein to SIGUSR2 signal |
| LOW | …flows/scripts/example_checks/detect_changed_example.py | 19 | # Output the result using print. Then the shell can get the values. |
| LOW | …ons/ColossalChat/coati/distributed/reward/reward_fn.py | 140 | # Check if the sequence is over length |
| LOW | …ons/ColossalChat/coati/distributed/reward/reward_fn.py | 205 | # Check if the sequence is over length |
| LOW | …ons/ColossalChat/coati/distributed/reward/reward_fn.py | 286 | # Check if the sequence is over length |
| LOW | …tions/ColossalChat/coati/dataset/tokenization_utils.py | 112 | # Check if all labels are ignored, this may happen when the tokenized length is too long |
| LOW | …tions/ColossalChat/coati/dataset/tokenization_utils.py | 321 | # Check if loss mask is all 0s (no loss), this may happen when the tokenized length is too long |
| LOW | applications/ColossalChat/tests/test_train.sh | 406 | # Check if the plugin is either "gemini_auto" or "gemini" and set grad_accu to '1' |
| LOW | applications/ColossalChat/tests/test_lora.py | 59 | # Check if model has overfitted |
| LOW | …ns/ColossalChat/examples/training_scripts/train_kto.py | 182 | # Check if the user specified weights fit into the theoratical lower and upper bounds from Eq. (8) of https://arxiv. |
| LOW | …t/examples/data_preparation_scripts/prepare_dataset.py | 173 | # Set stop_ids to eos_token_id for other dataset types if not exist |
| Severity | File | Line | Snippet |
|---|---|---|---|
| LOW | colossalai/fx/tracer/tracer.py | 97 | # if no extra manipulation is applied, we just pass the origin arguments to create_proxy function |
| MEDIUM | colossalai/legacy/context/parallel_mode.py | 17 | # this is added to facilitate amp and grad clipping in hybrid parallel |
| LOW | …lossalai/legacy/zero/sharded_optim/sharded_optim_v2.py | 339 | # We just set p.data = p.colo_attr.saved_grad.payload here |
| MEDIUM | …egacy/zero/shard_utils/bucket_tensor_shard_strategy.py | 40 | # Ensure we utilize maximum PCIE bandwidth |
| LOW | colossalai/autochunk/estimate_memory.py | 197 | # if node is placeholder, just add the size of the node |
| LOW | colossalai/cli/launcher/hostinfo.py | 36 | port = 22 # no port specified, lets just use the ssh port |
| LOW | colossalai/zero/gemini/chunk/chunk.py | 332 | # just use another way for the movement |
| LOW | colossalai/lazy/lazy_init.py | 340 | # inplace op, just return origin lazy tensor |
| MEDIUM | colossalai/kernel/triton/context_attn_unpad.py | 602 | # NOTE use `triton.next_power_of_2` here to utilize the cache mechanism of triton |
| MEDIUM | colossalai/kernel/triton/flash_decoding.py | 409 | # NOTE use `triton.next_power_of_2` here to utilize the cache mechanism of triton |
| LOW | …utorial/sequence_parallel/lr_scheduler/annealing_lr.py | 67 | # If the learning rate is constant, just return the initial value. |
| LOW | examples/language/gpt/gemini/train_gpt_demo.py | 210 | # we just use randomly generated data here |
| LOW | …uage/gpt/experiments/pipeline_parallel/train_gpt_pp.py | 143 | # we just use randomly generated data here |
| LOW | …pt/experiments/auto_parallel/auto_parallel_with_gpt.py | 81 | # we just use randomly generated data here |
| LOW | …les/community/roberta/pretraining/utils/global_vars.py | 113 | # pollutes the runs list, so we just add each as a scalar |
| MEDIUM | …/ColossalChat/coati/distributed/reward/reward_utils.py | 21 | """Performs comprehensive validation of response structure. |
| MEDIUM | …cations/ColossalChat/coati/utils/reward_score/utils.py | 21 | """Performs comprehensive validation of response structure. |
| Severity | File | Line | Snippet |
|---|---|---|---|
| MEDIUM | colossalai/shardformer/shard/shard_config.py | 116 | # you can add all the optimization flag here |
| MEDIUM | …s/tutorial/opt/inference/script/processing_ckpt_66b.py | 7 | # you can use whether wget or git lfs |
| MEDIUM | …ples/community/roberta/preprocessing/sentence_split.py | 143 | # if you have multiple server, you can use code below or modify code to openmpi |
| MEDIUM | …mples/community/roberta/preprocessing/tokenize_mask.py | 239 | # if you have multiple server, you can use code below or modify code to openmpi |
| MEDIUM | applications/ColossalQA/data/tests/test.html | 1113 | <p>If you are using {}-formatting (<a class="reference internal" href="https://docs.python.org/3/library/stdtypes.html#s |
| MEDIUM | applications/ColossalQA/data/tests/test.html | 1296 | counterparts in <a class="reference internal" href="https://docs.python.org/3/library/logging.html#logging.Logger" title |
| Severity | File | Line | Snippet |
|---|---|---|---|
| HIGH | colossalai/legacy/nn/layer/parallel_3d/_utils.py | 90 | return _async_grad_bucket.push(op, grad, param_id) |
| HIGH | colossalai/legacy/moe/layer/layers.py | 312 | | C1 || C2 || C3 || C4 | |
| HIGH | colossalai/legacy/moe/layer/layers.py | 313 | | A1 || A2 | | R1 | A3 || R2 | A4 || R3 | | R4 | |
| Severity | File | Line | Snippet |
|---|---|---|---|
| CRITICAL | …ns/ColossalChat/coati/distributed/inference_backend.py | 284 | self.llm.llm_engine.model_executor.driver_worker.model_runner.model.load_weights(state_dict.items()) |
| Severity | File | Line | Snippet |
|---|---|---|---|
| LOW | colossalai/fx/codegen/activation_checkpoint_codegen.py | 586 | # we need to check if the checkpoint need to offload the input |
| LOW | colossalai/fx/codegen/activation_checkpoint_codegen.py | 593 | # we need to check if the checkpoint need use_reentrant=False |
| LOW | .github/workflows/scripts/check_doc_i18n.py | 6 | # First, we need to check if the two directories exist |
| Severity | File | Line | Snippet |
|---|---|---|---|
| LOW | colossalai/auto_parallel/tensor_shard/utils/misc.py | 20 | def do_something(): |
| LOW | colossalai/testing/utils.py | 235 | def do_something(rank, world_size, port): |
| LOW | colossalai/testing/utils.py | 241 | def do_something(rank, world_size, port, arg1, arg2): |