127 Pull requests merged by 60 people

• [CINN] Improve subgraph topo index update algorithm

  #70380 merged Dec 23, 2024

• [Prim] Add index_add_double_grad in dygraph composite

  #70367 merged Dec 23, 2024

• [Inference]Support plugin in python converter

  #70251 merged Dec 23, 2024

• [CINN][Add Backend Pass Comment No.12] Add comment for TransformGpuForloop

  #70289 merged Dec 23, 2024

• [CINN][Backend Pass Update No.9] Update RemoveScheduleBlock pass

  #70334 merged Dec 23, 2024

• [CodeStyle][Typos][M-[1-22]] Fix typo(macor, mantain, mak, manualy, mannualy, maping, maxmium, Maxinum, menas, momory, Meatadata,miliseconds, Minium, minumum, minimun, modication, modifed, modifing, modifty, moduels, Modul, Morevoer, mutiple, Mutiply, multipy, Multiplie, mutiply, Muti, muti, mutexs)

  #70392 merged Dec 23, 2024

• [SCU][Docathon][Add API Legend No.7] Add legend for expand API

  #70143 merged Dec 23, 2024

• [SCU][Docathon][Add API Legend No.44] Add legend for repeat_interleave API

  #70128 merged Dec 23, 2024

• [Inference] FIX OpenVINO COMPILE BUG

  #70372 merged Dec 23, 2024

• [Kernel] Fix IndexAddGradKernel

  #70353 merged Dec 23, 2024

• [CINN]fix DimExpr sort

  #70336 merged Dec 23, 2024

• 【api】support “dcu” for set_device and device_guard api

  #70308 merged Dec 23, 2024

• [SOT] Visualize subgraph relation info by info collector

  #70387 merged Dec 21, 2024

• [CodeStyle][Typos][K-[2-4]] Fix typo (keyward, Keyward)

  #70356 merged Dec 21, 2024

• Disable WITH_OPENVINO

  #70365 merged Dec 20, 2024

• [Fix] Reduce unnecessary GpuMemGetInfo in Tensor.to

  #70359 merged Dec 20, 2024

• [AutoParallel] fix shard stage2 memory error

  #70363 merged Dec 20, 2024

• [CINN]fix fused gemm grad infer meta bug

  #70256 merged Dec 20, 2024

• [CINN] Disable transpose for TileBroadcastTactic

  #70350 merged Dec 20, 2024

• [CINN]optimize hardswish grad decomp

  #70327 merged Dec 20, 2024

• Fixed pir fitness when exporting model and comm init when model inferring

  #70237 merged Dec 20, 2024

• [inference]update pir argmax and bilinear trt op convert

  #70325 merged Dec 20, 2024

• Correct test cases that were never executed

  #70206 merged Dec 20, 2024

• [XPU] fix logits_max allreduce_type of XPU c_softmax_with_cross_entropy

  #70351 merged Dec 20, 2024

• 【SCU】【Paddle TensorRT No.23】Add pd_op.elementwise_pow converter

  #69580 merged Dec 20, 2024

• [Inference] release tensor in predictor destructor

  #70314 merged Dec 20, 2024

• [fluid_ops] Modify custom_allocator.cc include

  #70346 merged Dec 20, 2024

• 【CINN】Add longlong2int method for if and select.

  #70257 merged Dec 20, 2024

• [TJU][CINN][Add Backend Pass Comment No.2] Add comment for eliminate_common_global_memory_read

  #70304 merged Dec 20, 2024

• [CINN] Add fuesd_attention_grad symbol infer

  #70337 merged Dec 20, 2024

• [cherry-pick] fix doc of save_state_dict (#70285)

  #70311 merged Dec 20, 2024

• [CodeStyle][Typos][K-1] Fix typo (kenerl, kerenl, kernal)

  #70354 merged Dec 20, 2024

• [CodeStyle][Typos][[N-[1-5],N-[9-10],N-[12-14]] Fix typo(namess, nams, neccessary, Neet, neet, neeed, nedd,neighor,netwrok,normlize,noraml,numer)

  #70283 merged Dec 20, 2024

• [PIR] pir pylayer update input

  #70335 merged Dec 20, 2024

• 【Paddle Tensor 第二期 常用API复数类型支持 NO.6】 添加 full 函数复数类型支持

  #70277 merged Dec 19, 2024

• [Eager] Allow set dynamic attribute to eager tensor instance

  #70291 merged Dec 19, 2024

• [Paddle TensorRT] add pd_op.logical_not converter

  #70267 merged Dec 19, 2024

• [CINN] Implement the new RearrangeLoadInstruction pass

  #70258 merged Dec 19, 2024

• [CINN] Adding unit tests for transpose subgraphs from tipc models

  #70284 merged Dec 19, 2024

• 【Comm】Add out_shape attribute for to p_recv in phi

  #69019 merged Dec 19, 2024

• [fluid_ops] clean c_reduce_prod

  #70300 merged Dec 19, 2024

• [Inference]OpenVINO submodule clear

  #70313 merged Dec 19, 2024

• 【TJU】【Paddle Tensor 第二期 API支持 0-size Tensor】paddle.broadcast_to 支持 0-size tensor

  #70125 merged Dec 19, 2024

• [CINN] Restore "Enhance BuildCinnPass efficiency"

  #70249 merged Dec 19, 2024

• [SOT] Mark some APIs can be directly run in simulation mode

  #70293 merged Dec 19, 2024

• [XPU] replace xpu::where with xpu::nonzero_compute for nonzero kernel

  #70317 merged Dec 19, 2024

• [CINN]fix ArangeOpInferSymbolicShape

  #70305 merged Dec 19, 2024

• [CINN] add no_need_buffer for recompute algorithm

  #70255 merged Dec 19, 2024

• [Distributed] fix fusion optimizer helper

  #70323 merged Dec 19, 2024

• 【 Paddle Tensor 规范化二期 】删去redece_all_kernel.cc文件中多余的头文件

  #70252 merged Dec 19, 2024

• [SCU][CINN][Add Backend Pass Comment No.13] Add comment for UpdateBufferAxis

  #70271 merged Dec 19, 2024

• [CINN] Fix maximum recursion depth exceeded bug

  #70320 merged Dec 19, 2024

• Add dcu paddlex

  #68448 merged Dec 19, 2024

• [CINN]fix backend infer shape func new symbol error

  #70303 merged Dec 19, 2024

• 【SCU】【Paddle Tensor 第二期 API 支持 0-size TensorNo.25、47】paddle.tensordot 支持 0-size Tensor

  #70238 merged Dec 19, 2024

• [XPU] fuse swiglu in fc

  #70141 merged Dec 19, 2024

• Update hybrid_parallel_util.py

  #70297 merged Dec 19, 2024

• [Dy2st]Fix while to static bug

  #70287 merged Dec 19, 2024

• 【BUPT】[Paddle Tensor 第二期 API 鲁棒性增强] paddle.atan2 API 鲁棒性增强

  #70253 merged Dec 19, 2024

• [SOT][3.13] Re-enable SOT in Python 3.13 and run all unittest in CI

  #70319 merged Dec 19, 2024

• 【Inference Bug fixes】Fixed the problem of flash_attn modifying seed internally when dropout=0.0

  #70298 merged Dec 19, 2024

• Revert "[PIR] support pylayer pir updata"

  #70331 merged Dec 19, 2024

• [Comp] Add p_norm_grad in details.h

  #69675 merged Dec 19, 2024

• [Kernel] Optimize p_norm gpu (#69660)

  #70312 merged Dec 19, 2024

• [CINN]remove pow forward decomp

  #69546 merged Dec 19, 2024

• [CINN]add cinn jit instruction shape check

  #70266 merged Dec 19, 2024

• [Inference]Update the conditions for bilinear_interp_v2_op to be included in TRT.

  #69995 merged Dec 18, 2024

• [TJU][CINN][Add Backend Pass Comment No.11] Add comment for schedule_block_dce

  #70279 merged Dec 18, 2024

• 【custom】Add Custom engine dialect and give a engineOp test case

  #69815 merged Dec 18, 2024

• [fluid_ops] clean c_allreduce_avg

  #69745 merged Dec 18, 2024

• fix doc of save_state_dict

  #70285 merged Dec 18, 2024

• [PIR] support pylayer pir updata

  #70240 merged Dec 18, 2024

• 【Hackathon 7th No.20】为 Paddle 新增 Tensor.resize_ -part

  #69841 merged Dec 18, 2024

• 【HEU】【Paddle Tensor 第二期 API支持 0-size Tensor】paddle.eigh paddle.eigvalsh支持 0-size tensor

  #70254 merged Dec 18, 2024

• 【Paddle Tensor 第二期】增强 paddle.linalg.svd 鲁棒性

  #70235 merged Dec 18, 2024

• [Inference]fix openvino engine bug

  #70288 merged Dec 18, 2024

• 【SCU】【Paddle TensorRT No.24】Add pd_op.minimum converter

  #69736 merged Dec 18, 2024

• [CINN][New Hardware Update] SYCL first PR

  #69554 merged Dec 18, 2024

• [fluid_ops] clean c_reduce_max

  #70272 merged Dec 18, 2024

• [fluid_ops] clean c_reduce_min

  #69544 merged Dec 18, 2024

• [fluid_ops] clean alltoall

  #69446 merged Dec 18, 2024

• [CodeStyle][Typos][C-[60-66]]Fix typos(convience,coodinate,Copyed,copyed,copys,corener,corrrect,Correspoding)

  #70275 merged Dec 18, 2024

• [fluid_ops] clean c_reduce_sum

  #70268 merged Dec 18, 2024

• [fluid_ops]clean c_reduce_avg

  #69921 merged Dec 18, 2024

• [fluid_ops] clean c_allgather

  #69445 merged Dec 18, 2024

• [Inference] Some effective methods to reduce the loading time of pir models

  #70219 merged Dec 18, 2024

• 【Hackathon 7th No.32】为 paddle.nn.functional.scaled_dot_product_attention 进行功能增强

  #70166 merged Dec 18, 2024

• [Prim][Pir] Only set primitive backward for dynamic shape tests.

  #70270 merged Dec 18, 2024

• [CINN]fix FullWithTensorOpInferSymbolicShape

  #70269 merged Dec 18, 2024

• [Recompute] Fix recompute full_int_array bug

  #70243 merged Dec 18, 2024

• [Comp] Fix take_along_axis_grad when duplicated entries in indices

  #70250 merged Dec 18, 2024

• [CodeStyle][Typos][N-[6-8]] Fix typo (negetive, nagative, negtive, Neighboor, neigbhors, Neigbors)

  #70281 merged Dec 18, 2024

• [Paddle Tensorrt] change int32 to int64 in test file

  #69191 merged Dec 17, 2024

• 【Paddle Tensor 第二期 复数类型支持问题No.9、10】为paddle.not_equal、paddle.equal添加复数类型支持。

  #69968 merged Dec 17, 2024

• [XPU] support flash_attn_unpadded on xpu

  #70176 merged Dec 17, 2024

• 【Hackathon 7th No.21】为 Paddle 新增 reset_max_memory_reserved/reset_max_memory_allocated API -part

  #70032 merged Dec 17, 2024

• [fluid_ops] clean c_allreduce_prod

  #69754 merged Dec 17, 2024

• [Paddle TensorRT] Modify the issue when validating model inference precision in FP16.

  #70168 merged Dec 17, 2024

• support load merged checkpoint

  #70105 merged Dec 17, 2024

• 【prim】uniform Decomp list

  #70204 merged Dec 17, 2024

• 【SCU】[Paddle TensorRT No.28] Add pd_op.logical_xor converter

  #70170 merged Dec 17, 2024

• [CINN]fix IfFusionPass condition

  #70239 merged Dec 17, 2024

• [CINN] Add fusion result loop alignment check

  #69793 merged Dec 17, 2024

• [CINN] Remove group_schedule_tiling_first flag

  #69986 merged Dec 17, 2024

• [AutoParallel] opt memory usage for tensor-fusion save&load

  #70195 merged Dec 17, 2024

• [Distributed] fused param of sharding use ipc storage format

  #70246 merged Dec 17, 2024

• Integrate Flux

  #70097 merged Dec 17, 2024

• [CodeStyle][Typos][D-[20-23]] Fix typo(Destory,DEIVCE,dictionnary,diffrence)

  #70248 merged Dec 17, 2024

• [CodeStyle][Typos][C-59] Fix typos (Conver)

  #70259 merged Dec 17, 2024

• [CodeStyle][Typos][N-11] Fix typos (Normlized, normlized)

  #70263 merged Dec 17, 2024

• [CINN] Add the TileBroadcastTactic for NCHW broadcast

  #70092 merged Dec 16, 2024

• fused_bias_act op support fp8_e4m3

  #69865 merged Dec 16, 2024

• 【BUPT】【Paddle Tensor 第二期 API支持 0-size Tensor】paddle.matrix_rank 支持 0-size tensor

  #70130 merged Dec 16, 2024

• [Auto Parallel] add dropout spmd rules

  #70216 merged Dec 16, 2024

• [Dy2St][PIR] Add PIR mode into AST dy2st cache key

  #70218 merged Dec 16, 2024

• 【Paddle Tensor 第二期】使paddle.all支持0-size输入

  #70228 merged Dec 16, 2024

• nn.Embedding/F.embedding support scale_grad_by_freq 易用性提升

  #70090 merged Dec 16, 2024

• [CINN]fix symbol arg binding in bc optimize

  #70193 merged Dec 16, 2024

• Revert "[Pir] Fast inverse_sort_op"

  #70158 merged Dec 16, 2024

• optmize dropout decomp

  #70234 merged Dec 16, 2024

• [CINN][Add Backend Pass Comment No.9] Add comment for RemoveScheduleBlock

  #70225 merged Dec 16, 2024

• fix returns in to_distributed docs; test=document_fix

  #70209 merged Dec 16, 2024

• [CINN] Add new infer pass before cinn in inference

  #70220 merged Dec 16, 2024

• [SCU][CINN][Add Backend Pass Comment No.7] Add comment for CanMergeBlocks

  #70213 merged Dec 16, 2024

• 【Inference】Fix Bug for OpenVINO Config.h

  #70212 merged Dec 16, 2024

• [AutoParallel]:fix parallel tp error

  #70138 merged Dec 16, 2024

• [CINN]fix DimExpr Simplify check

  #70205 merged Dec 16, 2024

67 Pull requests opened by 39 people

• CI测试不review[fluid_ops] CustomDevice3

  #70241 opened Dec 16, 2024

• [Docathon][Add API Legend No.16] Add legend and update doc for atleast_2d -part

  #70242 opened Dec 16, 2024

• 【TJU】【Paddle Tensor 第二期 API支持 0-size Tensor】paddle.copysign 支持 0-size tensor

  #70244 opened Dec 16, 2024

• Revert "[CINN] Add the TileBroadcastTactic for NCHW broadcast (#70092)"

  #70264 opened Dec 17, 2024

• 【Inference】Revert "Support OpenVINO CPU High-Performance Inference"

  #70273 opened Dec 17, 2024

• CI测试不review[fluid_ops] c_allreduce_sum

  #70274 opened Dec 17, 2024

• add test

  #70280 opened Dec 17, 2024

• [CINN] fix slice op sym infer with input(with data) and symbol starts or ends

  #70282 opened Dec 17, 2024

• [Paddle TensorRT] support TensorRT Refittable

  #70286 opened Dec 17, 2024

• [CINN] skip CacheGradOpSymbolicShape check for fused_attention

  #70294 opened Dec 18, 2024

• fix

  #70295 opened Dec 18, 2024

• CI测试不review[fluid_ops] c_allreduce_max

  #70299 opened Dec 18, 2024

• 【GPUPS】add env for gpups and fix cache table

  #70301 opened Dec 18, 2024

• CI测试不review[fluid_ops] c_broadcast

  #70302 opened Dec 18, 2024

• 【PIR】Allow bn_&depthwiseConv change layout

  #70306 opened Dec 18, 2024

• [CINN][new hardware] SYCL second PR: add SYCL backend

  #70307 opened Dec 18, 2024

• [AutoParallel] add FLAGS_enable_sharding_stage1_tensor_fusion flag

  #70309 opened Dec 18, 2024

• Remove `return` seems to be unreachable

  #70310 opened Dec 18, 2024

• [CINN]Add support for TensorList input in the Pad3d infer symbolic interface

  #70318 opened Dec 18, 2024

• fix predictor mem release

  #70321 opened Dec 18, 2024

• [CINN][optimize warning message] optimize `SetSymbolForValueByStaticShape` && `CheckInferSymWithInferMeta`

  #70326 opened Dec 18, 2024

• try to fix long long bug

  #70328 opened Dec 18, 2024

• CI测试不review[fluid_ops] send_v2

  #70329 opened Dec 19, 2024

• CI测试不review[fluid_ops] recv_v2

  #70330 opened Dec 19, 2024

• fix typo

  #70332 opened Dec 19, 2024

• test

  #70338 opened Dec 19, 2024

• [Cherr-pick] Fix `take_along_axis_grad` and `p_norm_decomp`

  #70340 opened Dec 19, 2024

• The trace level has a higher logical priority

  #70341 opened Dec 19, 2024

• 【BUPT】[Paddle Tensor 第二期 API 支持 0-size Tensor] paddle.unstack 支持 0-size tensor

  #70342 opened Dec 19, 2024

• [CINN] Process 0D tensor in ReduceInferDim

  #70343 opened Dec 19, 2024

• CI测试不review[fluid_ops] allocator_facade.cc

  #70344 opened Dec 19, 2024

• [HEU][Paddle TensorRT No.21] Add (pd_op.dequantize_linear,pd_op.dequantize_linear_) converter

  #70345 opened Dec 19, 2024

• [XPU] feat: also support 5d in addition to 4d in interpolate op

  #70347 opened Dec 19, 2024

• CI测试不review[fluid_ops] inlined_vector

  #70348 opened Dec 19, 2024

• [CINN] Fix no_need_buffer bug

  #70349 opened Dec 19, 2024

• 【Comm】Comm init in dynamic mode

  #70352 opened Dec 19, 2024

• 【CINN】Add new visit method for IndexExpr

  #70355 opened Dec 19, 2024

• CI测试不review[fluid_ops] allreduce_type

  #70358 opened Dec 20, 2024

• fix

  #70360 opened Dec 20, 2024

• [CINN]support backend PassManager run on block

  #70361 opened Dec 20, 2024

• [SOT][Faster Guard] add `StringCompareGuard` and support `RangeVariable`

  #70362 opened Dec 20, 2024

• [PIR] translate old ir to pir_program and trt program

  #70364 opened Dec 20, 2024

• 【custom】add loadCustomEnginelib method

  #70366 opened Dec 20, 2024

• Fix DCU doc

  #70368 opened Dec 20, 2024

• 【 Paddle Tensor 规范化二期 】解决隐式广播不能够处理0-size的问题

  #70369 opened Dec 20, 2024

• index put bug

  #70370 opened Dec 20, 2024

• Directly use `paddle.full` to create complex tensor in `paddle.equal`.

  #70373 opened Dec 20, 2024

• 【cherry-pick】Fixed pir fitness when exporting model and comm init when model inferring

  #70374 opened Dec 20, 2024

• support pylayer for moe

  #70375 opened Dec 20, 2024

• Expr div

  #70376 opened Dec 20, 2024

• [PIR] Add state_dict to program_patch to enable default paramater

  #70377 opened Dec 20, 2024

• add xpu local compilation

  #70378 opened Dec 20, 2024

• 【SCU】【Paddle Tensor 第二期 API 支持 0-size TensorNo.26、50】paddle.linalg.vecdot 支持 0-size Tensor

  #70379 opened Dec 20, 2024

• [CINN] Fallback FuionOp with single reshape to PHI kernel

  #70381 opened Dec 20, 2024

• optimize cinn compile tools

  #70382 opened Dec 20, 2024

• Add general function for PIR Custom Pass

  #70383 opened Dec 20, 2024

• fix compile [[maybe_unused]]

  #70384 opened Dec 20, 2024

• [CINN] Accelerate recompute ips

  #70386 opened Dec 20, 2024

• CI测试不review[fluid_ops] c_allreduce_max 2

  #70388 opened Dec 20, 2024

• Update `F.multi_label_soft_margin_loss/nn.MultiLabelSoftMarginLoss` docs 易用性提升

  #70389 opened Dec 21, 2024

• [Windows CI] Record the exact size of dll and whl files and save them to disk D

  #70390 opened Dec 21, 2024

• Transform to trt program

  #70394 opened Dec 22, 2024

• [CI] Add clone pipeline in action

  #70395 opened Dec 23, 2024

• clear sym shape str print

  #70396 opened Dec 23, 2024

• [Warning] Refine `to_tensor` warning

  #70398 opened Dec 23, 2024

• cuda_error_check

  #70399 opened Dec 23, 2024

• [Paddle TensorRT] add pd_op.logical_not converter

  #70400 opened Dec 23, 2024

7 Issues closed by 3 people

• 电脑更新GPU后出现ImportError: DLL load failed while importing libpaddle: %1 不是有效的 Win32 应用程序。

  #70157 closed Dec 23, 2024

• 日志打印问题

  #69433 closed Dec 19, 2024

• 无法定位程序输入点

  #59559 closed Dec 17, 2024

• grid_sample 输出结果与torch的不一致

  #58907 closed Dec 17, 2024

• bfloat16 tensor can not be assigned with a number value

  #57950 closed Dec 17, 2024

• Docker镜像使用gpu问题。

  #53821 closed Dec 17, 2024

• paddle报错：KeyError: 'create_py_reader'

  #53474 closed Dec 17, 2024

11 Issues opened by 10 people

• 安装报错

  #70393 opened Dec 22, 2024

• 调试代码错误

  #70391 opened Dec 21, 2024

• TransformerEncoderLayer中的activation激活参数需求。内容很简短烦请请大佬赐教。

  #70385 opened Dec 20, 2024

• PaddleOCR Multi-Machine Distributed Training

  #70357 opened Dec 20, 2024

• 模型剪枝后的速度问题

  #70339 opened Dec 19, 2024

• paddlepaddle 需要支持 python3.13

  #70316 opened Dec 18, 2024

• 怎么在hadoop上跑paddle预估产出embedding

  #70315 opened Dec 18, 2024

• 编译最新 develop ，在链接 `inline_translator_test` 大量报错 `undefined reference to xxxx`

  #70290 opened Dec 18, 2024

• paddle 1.x版本推理模型如何转换成paddle 2.x版本推理模型

  #70276 opened Dec 17, 2024

• 请问下profiler如何能完整打印kernel的名字和类型？

  #70265 opened Dec 17, 2024

• 3.0bata版本的lstm参数输出有误

  #70245 opened Dec 16, 2024

100 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• [HEU][Paddle TensorRT No.62、64] Add (pd_op.index_put、pd_op.pow) converter

  #69889 commented on Dec 19, 2024 • 17 new comments

• 支持非均衡VPP编排的灵活模型层分配策略

  #70230 commented on Dec 18, 2024 • 9 new comments

• 【BUPT】【Paddle TensorRT No.18】Add pd_op.conv3d and pd_op.conv3d_transpose converter

  #69757 commented on Dec 16, 2024 • 5 new comments

• support c_softmax_with_multi_label_cross_entropy

  #69337 commented on Dec 20, 2024 • 4 new comments

• 【BUPT】【Paddle Tensor 第二期 API支持 0-size Tensor】paddle.matrix_norm 支持 0-size tensor

  #70186 commented on Dec 19, 2024 • 4 new comments

• [CINN][Add Backend Pass Comment No.10] Add comment for replace_cross_thread_reduction

  #70227 commented on Dec 16, 2024 • 3 new comments

• [Docathon][Add API Legend No.55、22] add paddle.index_fill and paddle.shard_index API Paddle

  #70215 commented on Dec 20, 2024 • 3 new comments

• [SCU][Docathon][Add API Legend No.48]add illustration for as_strided API

  #69874 commented on Dec 22, 2024 • 3 new comments

• [Paddle TensorRT No.50] Add pd_op.prelu converter

  #69394 commented on Dec 18, 2024 • 2 new comments

• [SCU][Paddle TensorRT No.58] Add pd_op.unbind converter

  #70066 commented on Dec 20, 2024 • 2 new comments

• [Paddle TensorRT No.43、44] Add pd_op.{`leaky_relu`, `logsigmoid`} converter

  #69379 commented on Dec 18, 2024 • 2 new comments

• [CINN][Backend Pass Update No.3] Update extern_call_process pass

  #70191 commented on Dec 16, 2024 • 2 new comments

• 【SCU】【Paddle Tensor 第二期 常用API复数类型支持No.7】添加 nonzero 函数复数类型支持

  #70200 commented on Dec 16, 2024 • 2 new comments

• 【BUPT】【Paddle TensorRT No.38】Add pd_op.gather converter

  #69714 commented on Dec 16, 2024 • 2 new comments

• [Docathon][Add API Legend No.40、51] add figures for unfold and unsqueeze API

  #70048 commented on Dec 20, 2024 • 2 new comments

• [CINN] Add Div shape op

  #70224 commented on Dec 20, 2024 • 2 new comments

• 【Hackathon 7th No.28】为 paddle.clip 进行功能增强

  #69269 commented on Dec 18, 2024 • 1 new comment

• 【SCU】【Paddle TensorRT No.11、12、65】Add `pd_op.bitwise_and`、`pd_op.bitwise_or`、`pd_op.bitwise_not` converter

  #69599 commented on Dec 20, 2024 • 1 new comment

• [SCU][Docathon][Add API Legend No.53] Add legend for masked_scatter API

  #70145 commented on Dec 23, 2024 • 1 new comment

• [SCU][Paddle TensorRT No.40] Add pd_op.index_select converter

  #69762 commented on Dec 20, 2024 • 1 new comment

• decomp histogram(temp)

  #70096 commented on Dec 20, 2024 • 0 new comments

• [Hackathon 7th No.55] BUG FIX `paddle.nn.functional.fold`

  #70102 commented on Dec 18, 2024 • 0 new comments

• [XPU] disable use_stride_kernel for XPU

  #70108 commented on Dec 18, 2024 • 0 new comments

• 【SCU】【Paddle Tensor 第二期 API其他类型支持No.5】为 `round` 函数添加 `int` 、复数支持

  #70094 commented on Dec 20, 2024 • 0 new comments

• [AutoParallel]:fix FLAGS_log_memory_stats wrong log

  #70081 commented on Dec 17, 2024 • 0 new comments

• [Prim][Pir] Add backward decomp and support dynamic for angle_grad(not complex)

  #70080 commented on Dec 19, 2024 • 0 new comments

• 【Hackathon 7th No.26】add `lu_solve` api for paddle

  #70078 commented on Dec 17, 2024 • 0 new comments

• support pir_s_to_p_reshard and test

  #70075 commented on Dec 16, 2024 • 0 new comments

• [PIR-Auto-Parallel]refactor refined recompute pass in PIR mode

  #70064 commented on Dec 20, 2024 • 0 new comments

• Pass init fix

  #70063 commented on Dec 17, 2024 • 0 new comments

• Check

  #70062 commented on Dec 20, 2024 • 0 new comments

• [Inference] Fused Moe Optimization

  #70059 commented on Dec 17, 2024 • 0 new comments

• Test transpose pass from ChenControl

  #70057 commented on Dec 17, 2024 • 0 new comments

• fix cinn.concat infer symbolic bug

  #70056 commented on Dec 17, 2024 • 0 new comments

• CI测试不review[fluid_ops] pybind.cc 2

  #70055 commented on Dec 17, 2024 • 0 new comments

• [CINN] Update codegen_device_util.h without visitor

  #70236 commented on Dec 16, 2024 • 0 new comments

• [CINN][Add Backend Pass Comment No.3] Add comment for extern_call_process

  #70233 commented on Dec 16, 2024 • 0 new comments

• 【CINN】Set index for IrNode

  #70208 commented on Dec 20, 2024 • 0 new comments

• [AutoParallel] add while op test demo and fix has_elements bug

  #70207 commented on Dec 23, 2024 • 0 new comments

• Revert "fix_cinnapi.so_symbol_bug"

  #70185 commented on Dec 20, 2024 • 0 new comments

• [Prim][Pir] Decomp diag op

  #70183 commented on Dec 20, 2024 • 0 new comments

• [XPU] rotary interface update && xhpc update

  #70175 commented on Dec 16, 2024 • 0 new comments

• [CINN] draft:Update codegen_device_util without visitor

  #70172 commented on Dec 20, 2024 • 0 new comments

• Cinn support no need buffer

  #70171 commented on Dec 18, 2024 • 0 new comments

• [SCU][Docathon][Add API Legend No.39]add the picture of unique_consecutive

  #70163 commented on Dec 20, 2024 • 0 new comments

• Fixed a bug where required header was missing

  #70159 commented on Dec 20, 2024 • 0 new comments

• fix pir_p_to_s_reshard unbalanced split_axis and dst_type

  #70155 commented on Dec 16, 2024 • 0 new comments

• [SOT][Faster Guard] implement basic guard tree mechanism

  #70154 commented on Dec 20, 2024 • 0 new comments

• Upgrade normalize for number() > int32

  #70152 commented on Dec 20, 2024 • 0 new comments

• Update paddlex modules

  #70148 commented on Dec 20, 2024 • 0 new comments

• 【SCU】【Paddle Tensor 第二期 API支持 0-size Tensor】（paddle.sum、paddle.mean、paddle.prod、 paddle.var、 paddle.std）支持 0-size tensor

  #70146 commented on Dec 20, 2024 • 0 new comments

• 【CINN】Add pass to fold consecutive IfThenElse.

  #70142 commented on Dec 19, 2024 • 0 new comments

• mem check cuda 700 big tensor

  #70131 commented on Dec 20, 2024 • 0 new comments

• 【SCU】【Paddle Tensor 第二期】为flip(isnan, isfinite, isinf)添加复数类型支持

  #70126 commented on Dec 20, 2024 • 0 new comments

• [XPU] Update xhpc to 1209

  #70114 commented on Dec 20, 2024 • 0 new comments

• strided_copy bug

  #70110 commented on Dec 20, 2024 • 0 new comments

• CI测试不review [fluid_ops] test

  #69275 commented on Dec 19, 2024 • 0 new comments

• [Dist Dialect] Add placements and dist_reshape api in pir

  #69262 commented on Dec 17, 2024 • 0 new comments

• CI测试不review [fluid_ops] broadcast2

  #69081 commented on Dec 19, 2024 • 0 new comments

• 【Hackathon 7th Fundable Projects 2 No.89】 [fluid_ops] custom operator

  #68540 commented on Dec 16, 2024 • 0 new comments

• 【开源任务】CINN编译器后端Pass注释添加

  #70113 commented on Dec 23, 2024 • 0 new comments

• 【开源任务】CINN编译器后端Pass改造

  #69639 commented on Dec 23, 2024 • 0 new comments

• 安装报错。

  #70069 commented on Dec 23, 2024 • 0 new comments

• [CodeStyle][Typos][A-N] ✏️ Typos 工具引入计划

  #69441 commented on Dec 22, 2024 • 0 new comments

• Strided_slice 官网文档中未描述stride 不能整除的情况，给出的示例也未说明出现这中情况时算子的行为

  #70203 commented on Dec 21, 2024 • 0 new comments

• 【Hackathon 7th】开源贡献个人挑战赛

  #68244 commented on Dec 20, 2024 • 0 new comments

• paddle的最新gpu版本3.0.0b2，报错：Illegal instruction (core dumped)

  #70223 commented on Dec 20, 2024 • 0 new comments

• 【开源任务】PIR-TensorRT converter推全升级

  #69178 commented on Dec 20, 2024 • 0 new comments

• 【Hackathon 7th】Fundable Projects

  #68245 commented on Dec 20, 2024 • 0 new comments

• 【Hackathon 7th】开源贡献个人挑战赛（偷跑版）

  #67603 commented on Dec 19, 2024 • 0 new comments

• 【HACKATHON 7th Code Camp】黑客松护航计划集训营（正式批）

  #67601 commented on Dec 19, 2024 • 0 new comments

• 海光服务器+paddle inference专用推理库+麒麟OS

  #69792 commented on Dec 17, 2024 • 0 new comments

• 请问单gpu如何提高并发量

  #68885 commented on Dec 16, 2024 • 0 new comments

• 【快乐开源】Paddle Tensor 规范化二期

  #69908 commented on Dec 16, 2024 • 0 new comments

• 请问ARM平台上没有提供paddle2.5.1版本，只有2.5.2版本的paddle吗

  #69945 commented on Dec 16, 2024 • 0 new comments

• LSTM禁用偏置项，就会导致程序无法运行

  #70231 commented on Dec 16, 2024 • 0 new comments

• 百度昆仑芯R200下，使用Paddle Inference如何指定模型运行精度（如INT8）？

  #70117 commented on Dec 16, 2024 • 0 new comments

• SOT Guard 机制性能优化

  #69264 commented on Dec 16, 2024 • 0 new comments

• CI测试不review[fluid_ops] test LoD

  #70051 commented on Dec 20, 2024 • 0 new comments

• [CI]update scripts of Auto-Parallel CI

  #70041 commented on Dec 20, 2024 • 0 new comments

• Revert "[CINN] Update loop fusion relax control flow"

  #70028 commented on Dec 17, 2024 • 0 new comments

• [Dist dialect] Add pass for dist_reshape and nd_mesh_all2all in PIR

  #70027 commented on Dec 20, 2024 • 0 new comments

• CI测试不review[fluid_ops] load_combine

  #70001 commented on Dec 18, 2024 • 0 new comments

• 【Auto-Parallel | Comm】fix communication hang issue on the GPU-H

  #69997 commented on Dec 17, 2024 • 0 new comments

• CI测试不review[fluid_ops] SetLoDLevel

  #69973 commented on Dec 20, 2024 • 0 new comments

• 【SCU】[Paddle TensorRT No.28] Add pd_op.logical_xor converter

  #69958 commented on Dec 17, 2024 • 0 new comments

• CI测试不review[fluid_ops] all_to_all

  #69951 commented on Dec 16, 2024 • 0 new comments

• [Auto Parallel]add align mode code for dp

  #69941 commented on Dec 20, 2024 • 0 new comments

• [hackathon 7th] PaddleSpeech line: fix when shape=() can not load

  #69907 commented on Dec 20, 2024 • 0 new comments

• [CINN] open data check for sym infer

  #69891 commented on Dec 23, 2024 • 0 new comments

• 【SCU】【Paddle TensorRT No.55】Add pd_op.take_along_axis converter

  #69871 commented on Dec 20, 2024 • 0 new comments

• 【SCU】【Paddle TensorRT No.57】Add `pd_op.temporal_shift` converter

  #69848 commented on Dec 19, 2024 • 0 new comments

• [XPU] update XPU matmul

  #69755 commented on Dec 20, 2024 • 0 new comments

• [CINN] apply vectorize Primitive in IRSchedule

  #69732 commented on Dec 20, 2024 • 0 new comments

• 【SCU】【Paddle TensorRT No.54】Add pd_op.shuffle_channell converter

  #69718 commented on Dec 17, 2024 • 0 new comments

• 【SCU】【Paddle TensorRT No.56】Add `pd_op.tanh_shrink` converter

  #69693 commented on Dec 18, 2024 • 0 new comments

• 【SCU】【Paddle TensorRT No.48】Add `pd_op.pad` converter

  #69673 commented on Dec 20, 2024 • 0 new comments

• [HEU][Paddle TensorRT No.3、51] Add (pd_op.selu、pd_op.pool3d) converter

  #69663 commented on Dec 16, 2024 • 0 new comments

• CI测试不review[fluid_ops] run_program

  #69626 commented on Dec 20, 2024 • 0 new comments

• [Paddle TensorRT No.19] Add pd_op.cumsum converter

  #69330 commented on Dec 16, 2024 • 0 new comments