MCPcopy Create free account

hub / github.com/tile-ai/tilelang / functions

Functions7,254 in github.com/tile-ai/tilelang

Methodforward
( ctx, q_unpad, k_unpad, v_unpad, sinks, cu_seqlens_q, cu_seqlens_k, N_CTX, max_seqlen_q, max_seqlen_k
examples/attention_sink/example_gqa_sink_bwd_varlen.py:503
Methodforward
(ctx, q, k, v, sinks, window_size)
examples/attention_sink/example_mha_sink_bwd_bhsd.py:342
Methodforward
(ctx, q, k, v, sinks, window_size, groups)
examples/attention_sink/example_gqa_sink_bwd_bhsd.py:334
Methodforward
(ctx, x, gamma, beta, eps)
examples/norm/layernorm.py:138
Methodforward
(self, input)
examples/bitnet-1.58b/utils_quant.py:199
Methodforward
(self, input)
examples/bitnet-1.58b/utils_quant.py:222
Methodforward
(self, hidden_states)
examples/bitnet-1.58b/modeling_bitnet.py:97
Methodforward
(self, x, position_ids)
examples/bitnet-1.58b/modeling_bitnet.py:144
Methodforward
(self, x)
examples/bitnet-1.58b/modeling_bitnet.py:224
Methodforward
(self, x)
examples/bitnet-1.58b/modeling_bitnet.py:263
Methodforward
( self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor] = None,
examples/bitnet-1.58b/modeling_bitnet.py:354
Methodforward
( self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor] = None,
examples/bitnet-1.58b/modeling_bitnet.py:487
Methodforward
( self, hidden_states: torch.Tensor, attention_mask: Optional[torch.LongTensor] = None
examples/bitnet-1.58b/modeling_bitnet.py:563
Methodforward
Args: hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)` attention_mask (
examples/bitnet-1.58b/modeling_bitnet.py:751
Methodforward
( self, input_ids: torch.LongTensor = None, attention_mask: Optional[torch.Tensor] = N
examples/bitnet-1.58b/modeling_bitnet.py:984
Methodforward
r""" Args: labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*): Labels for computing
examples/bitnet-1.58b/modeling_bitnet.py:1162
Methodforward
r""" labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*): Labels for computing the sequence classification/regression
examples/bitnet-1.58b/modeling_bitnet.py:1499
Methodforward
r""" start_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*): Labels for position (index) of the start of the la
examples/bitnet-1.58b/modeling_bitnet.py:1616
Methodforward
(self, query, key, value, block_indices, cache_seqlens)
examples/blocksparse_attention/example_tilelang_sparse_gqa_decode_varlen_indice.py:172
Methodforward
(self, query, key, value, block_mask, cache_seqlens)
examples/blocksparse_attention/example_tilelang_sparse_gqa_decode_varlen_mask.py:166
Methodforward
(ctx, q, k, v, block_sparse_dense, sm_scale)
examples/blocksparse_attention/block_sparse_attn_triton.py:240
Methodforward
(self, x: torch.Tensor)
examples/fusedmoe/example_fusedmoe_torch.py:20
Methodforward
(self, x: torch.Tensor)
examples/fusedmoe/example_fusedmoe_torch.py:35
Methodforward
(self, x: torch.Tensor)
examples/fusedmoe/example_fusedmoe_torch.py:52
Methodforward
(self, x: torch.Tensor)
examples/fusedmoe/example_fusedmoe_tilelang.py:241
Methodforward
(self, x: torch.Tensor)
examples/fusedmoe/example_fusedmoe_tilelang.py:256
Methodforward
(self, x: torch.Tensor)
examples/fusedmoe/example_fusedmoe_tilelang.py:330
Methodforward
(ctx, q, k, v, block_sparse_dense, sm_scale)
examples/seer_attention/block_sparse_attn_triton.py:235
Methodforward
Given the row index `i` and column index `j` in the fragment,
tilelang/cuda/intrinsics/macro/mma_sm70_macro_generator.py:396
Methodforward_fp32_simulated
(self, input)
examples/bitnet-1.58b/utils_quant.py:182
Functionforward_index
Given the row index `i` and column index `j` in the fragment,
examples/plot_layout/fragment_mma_load_a.py:86
Functionforward_index
Given the row index `i` and column index `j` in the fragment,
examples/plot_layout/fragment_mfma_load_a.py:91
Methodforward_index
(i, j)
tilelang/rocm/intrinsics/wmma_macro_generator.py:432
Methodforward_index
Given the row index `i` and column index `j` in the fragment,
tilelang/rocm/intrinsics/mfma_macro_generator.py:659
Methodforward_index
Given the row index `i` and column index `j` in the fragment, map them to a local index in a single thread according
tilelang/cuda/intrinsics/macro/mma_sm70_macro_generator.py:485
Methodforward_index
Given the row index `i` and column index `j` in the fragment,
tilelang/cuda/intrinsics/macro/mma_macro_generator.py:757
Methodforward_index
Given the row index `i` and column index `j` in the fragment,
tilelang/cuda/intrinsics/macro/mma_sp_macro_generator.py:708
Methodforward_index
Given the row index `i` and column index `j` in the fragment,
tilelang/cuda/intrinsics/macro/wgmma_macro_generator.py:790
Methodforward_index
(i: int, j: int)
tilelang/cuda/intrinsics/macro/wgmma_sp_macro_generator.py:530
Functionforward_thread
Given the row index `i` and column index `j` in the fragment,
examples/plot_layout/fragment_mma_load_a.py:79
Functionforward_thread
Given the row index `i` and column index `j` in the fragment,
examples/plot_layout/fragment_mfma_load_a.py:84
Methodforward_thread
(i, j)
tilelang/rocm/intrinsics/wmma_macro_generator.py:428
Methodforward_thread
Given the row index `i` and column index `j` in the fragment,
tilelang/rocm/intrinsics/mfma_macro_generator.py:652
Methodforward_thread
Given the row index `i` and column index `j` in the fragment, map them to a thread index according to `inverse_mma_store_layo
tilelang/cuda/intrinsics/macro/mma_sm70_macro_generator.py:468
Methodforward_thread
Given the row index `i` and column index `j` in the fragment,
tilelang/cuda/intrinsics/macro/mma_macro_generator.py:750
Methodforward_thread
Given the row index `i` and column index `j` in the fragment,
tilelang/cuda/intrinsics/macro/mma_sp_macro_generator.py:701
Methodforward_thread
Given the row index `i` and column index `j` in the fragment,
tilelang/cuda/intrinsics/macro/wgmma_macro_generator.py:783
Methodforward_thread
(i: int, j: int)
tilelang/cuda/intrinsics/macro/wgmma_sp_macro_generator.py:526
Functionfp8_gemm_kernel_
( A: T.Tensor[(M, K), FP8], B: T.Tensor[(N, K), FP8], C: T.Tensor[(M, N), out_dtype],
examples/deepseek_v32/inference/kernel.py:118
Functionfp8_index_kernel_
( q: T.Tensor[(b, m, h, d), FP8], q_s: T.Tensor[(b, m, h), FP32], k: T.Tensor[(b, n, d
examples/deepseek_v32/inference/kernel.py:201
Functionfp8_remove_negative_zeros_
(tensor: torch.Tensor)
tilelang/utils/tensor.py:25
Functionfragment_forward_A_colmajor_gfx11
Replicated fragment forward map for gfx11 transposed A.
tilelang/rocm/intrinsics/wmma_layout.py:296
Functionfragment_forward_A_gfx11
Replicated fragment forward map for gfx11 A. The canonical owner lives in the lower half-wave and `rep` selects whether the logical element i
tilelang/rocm/intrinsics/wmma_layout.py:286
Functionfragment_forward_B_colmajor_gfx11
Replicated fragment forward map for gfx11 transposed B.
tilelang/rocm/intrinsics/wmma_layout.py:308
Functionfragment_forward_B_gfx11
Replicated fragment forward map for gfx11 B.
tilelang/rocm/intrinsics/wmma_layout.py:302
Methodfrom_code
( cls, name: str, kind: ASTPatKind, match: str, rewrite: str,
tilelang/autodd.py:334
Methodfrom_database
( cls, params: list[TensorType], result_idx: list[int], target: str, f
tilelang/jit/adapter/tvm_ffi.py:263
Methodfrom_database
( cls, params: list[TensorType], result_idx: list[int], target: str, f
tilelang/jit/adapter/cython/adapter.py:153
Methodfrom_database
( cls, params: list[KernelParam], result_idx: list[int], target: str,
tilelang/jit/adapter/nvrtc/adapter.py:100
Methodfrom_database
(cls, *args, **kwargs)
tilelang/jit/adapter/nvrtc/__init__.py:55
Methodfrom_database
Rebuild a CuTeDSL adapter from persisted cache artifacts.
tilelang/jit/adapter/cutedsl/adapter.py:109
Methodfrom_dict
(cls, dic: dict)
tilelang/carver/roller/hint.py:225
Methodfrom_ptr
Create a buffer from a pointer, shape, and data type. Args: pointer_var: The pointer variable shape: The shape of the
tilelang/language/proxy.py:58
Methodfrom_ptr
Create a buffer from a pointer, shape, and data type. Args: pointer_var: The pointer variable shape: The shape of the
tilelang/language/proxy.py:133
Methodfrom_source
(cls, source: str, *args, **kwargs)
tilelang/autodd.py:616
Methodfrom_source
(cls, source, *args, **kwargs)
tilelang/autodd.py:627
Methodfrom_source
(cls, source, *args, **kwargs)
tilelang/autodd.py:692
Methodfrom_tilelang_function
Alternative constructor to create a TorchFunction directly from a TileLang PrimFunc. Parameters ---------- tilelang_
tilelang/jit/kernel.py:413
Functionfunc
( A: T.Tensor((128,), T.float32), B: T.Tensor((128,), T.float32), )
testing/python/transform/test_tilelang_transform_lexical_alloc_scope.py:73
Functionfunc
(x: T.Tensor((256,), "int64"))
testing/python/transform/test_tilelang_transform_plan_update_buffer_allocation_location.py:45
Functionfunc
()
testing/python/transform/test_tilelang_transform_lower_shared_tmem.py:33
Functionfunc
()
testing/python/transform/test_tilelang_transform_lower_shared_barrier.py:65
Functionfunc
(X: T.Tensor((256, 256), T.float16), Y: T.Tensor((256, 256), T.float16))
testing/python/transform/test_tilelang_transform_inject_tcgen05_fence.py:98
Functionfunc
( A: T.Buffer([16, 16], "int32"), B: T.Buffer([16, 16], "int32"), C: T.Buffer([16, 16]
testing/python/transform/test_tilelang_transform_make_packed_api.py:168
Functionfunc
(A: T.Buffer((128,), "float32"), B: T.Buffer((128,), "float32"))
testing/python/transform/test_tilelang_transform_lower_ldgstg.py:48
Functionfunc1
(A: T.Tensor((M, N), T.float32), B: T.Tensor((M, N), T.float32))
testing/python/cache/test_tilelang_kernel_cache.py:227
Functionfunc2
(A: T.Tensor((M, N), T.float32), B: T.Tensor((M, N), T.float32))
testing/python/cache/test_tilelang_kernel_cache.py:245
Functionfunc_attr
The PrimFunc annotation statement. Parameters ---------- attrs : Dict[str, Any] The annotations of the PrimFunc.
tilelang/language/ast/ir.py:219
Methodfunc_call_args
(s, function_args)
tilelang/jit/adapter/wrapper.py:814
Functionfunc_name
The PrimFunc naming statement. Parameters ---------- name : str The name of the PrimFunc.
tilelang/language/ast/ir.py:208
Functionfunc_ret
The PrimFunc return type statement. Parameters ---------- ret_type : Type The return type of the PrimFunc. Returns -----
tilelang/language/ast/ir.py:230
Functionfused_chunk_linear_attn_bwd
( Q: T.Tensor([B, S, H, DK], dtype), # type: ignore K: T.Tensor([B, S, H, DK], dtype), # typ
examples/linear_attention/example_linear_attn_bwd.py:34
Functionfused_chunk_linear_attn_fwd
( Q: T.Tensor([B, S, H, DK], dtype), # type: ignore K: T.Tensor([B, S, H, DK], dtype), # typ
examples/linear_attention/example_linear_attn_fwd.py:37
Functionfvisit
(x)
tilelang/carver/roller/shape_inference/tir.py:338
Functionfwd_main
(KV: T.Tensor((M, N), dtype), ids: T.Tensor((4,), T.int32))
testing/python/issue/test_tilelang_issue_1210.py:10
Functionfwd_main
( KV: T.Tensor((M, N), dtype), ids: T.Tensor((4,), "int32"), ids2: T.Tensor((4,), "int
testing/python/issue/test_tilelang_issue_1263.py:9
Functiongemm
(A, B, block_M, block_N, block_K, in_dtype, out_dtype, accum_dtype, num_stages, use_tma_store=True)
examples/gemm_sm100/gemm_tcgen5mma_ws.py:10
Functiongemm
( A: T.Tensor((M, K), dtype), B: T.Tensor((K, N), dtype), C: T.Tensor((M, N), dtype),
testing/python/profiler/test_tilelang_profiler_dynamic_symbolic.py:16
Functiongemm
( A: T.Tensor((M, K), dtype), B: T.Tensor((K, N), dtype), C: T.Tensor((M, N), dtype),
testing/python/profiler/test_tilelang_profiler.py:9
Functiongemm
( A: T.Tensor((M, K), dtype), B: T.Tensor((K, N), dtype), C: T.Tensor((M, N), dtype),
testing/python/language/test_tilelang_language_cluster.py:9
Functiongemm
( a: T.Tensor((M, K), "bfloat16"), b: T.Tensor((K, N), "bfloat16"), c: T.Tensor((M, N)
testing/python/issue/test_tilelang_issue_sm120_tma_smem_alignment.py:62
Functiongemm
( A: T.Tensor((M, K), dtype), B: T.Tensor((K, N), dtype), C: T.Tensor((M, N), dtype),
testing/python/metal/test_metal_codegen.py:11
Functiongemm
( A: T.Tensor((M, K), T.float16), B: T.Tensor((K, N), T.float16), C: T.Tensor((M, N),
testing/python/transform/test_tilelang_transform_pipeline_barrier_ownership.py:34
Functiongemm_2cta
(A, B, block_M, block_N, block_K, in_dtype, out_dtype, accum_dtype, num_stages, use_tma_store=True)
examples/gemm_sm100/gemm_tcgen5mma_ws.py:73
Functiongemm_autotune
( A: T.Tensor((M, K), dtype), B: T.Tensor((N, K), dtype), C: T.Tensor((M, N), dtype),
examples/gemm/example_gemm_autotune.py:190
Functiongemm_autotune
( A: T.Tensor((M, K), dtype), B: T.Tensor((N, K), dtype), C: T.Tensor((M, N), dtype),
examples/gemm/example_gemm_advanced_autotune.py:248
Functiongemm_fp8_rs
( A: T.Tensor((M, K), dtype), B: T.Tensor((N, K), dtype), C: T.Tensor((M, N), accum_dt
examples/gemm_fp8/example_tilelang_gemm_amd.py:62
Functiongemm_fp8_ss
( A: T.Tensor((M, K), dtype), B: T.Tensor((N, K), dtype), C: T.Tensor((M, N), accum_dt
examples/gemm_fp8/example_tilelang_gemm_amd.py:81
Functiongemm_infer_layout
(gemm, target: Target, thread_bounds: Range)
tilelang/tileop/gemm/__init__.py:13
Functiongemm_kernel
( A: T.Tensor((M, K), dtype), B: T.Tensor((K, N), dtype), C: T.Tensor((M, N), accum_dt
benchmark/matmul_metal/benchmark_matmul_metal.py:24
← previousnext →4,501–4,600 of 7,254, ranked by callers