Paddle icon indicating copy to clipboard operation
Paddle copied to clipboard

PArallel Distributed Deep LEarning: Machine Learning Framework from Industrial Practice (『飞桨』核心框架,深度学习&机器学习高性能单机、分布式训练和跨平台部署)

Results 1049 Paddle issues
Sort by recently updated
recently updated
newest added
trafficstars

### PR types Others ### PR changes Others ### Describe Others

contributor
status: proposed

### PR types Others ### PR changes Others ### Describe - conv2d_transpose至少从7134开始就支持不对称的padding,可认为其一直支持 - conv3d_transpose自8406开始支持不对称的padding了

### PR types New features ### PR changes APIs ### Describe Add paddle.geometric.send_ue_recv API, mainly used in graph learning domain.

### bug描述 Describe the Bug `paddle.masked_select` 返回错误的值 我本要实现一个这样的需求: ```python >>> import numpy as np >>> A = np.arange(6).reshape(3, 2) >>> A [[0 1] [2 3] [4 5]] >>> mask =...

status/new-issue
type/bug-report

## 为啥要自己编译 - 因为官方包不支持ubuntu22.04(系统自带gcc版本与glibc版本太高) ```bash ImportError: /home/ubuntu/anaconda3/envs/paddle/lib/python3.9/site-packages/paddle/fluid/core_avx.so: undefined symbol: _dl_sym, version GLIBC_PRIVATE ``` - 参考链接:https://www.paddlepaddle.org.cn/documentation/docs/zh/install/compile/linux-compile.html - 原理,将利用conda安装gcc/glibc,不改系统环境的gcc/glibc,防止系统出现莫名其妙的bug。 ## 环境 - 自带python环境(其实没啥影响,只是展示一下) ```bash $ /usr/bin/python3 --version Python 3.10.4 ``` - cmake环境(建议版本装高一下,貌似要3.19以上)...

status/following-up
type/build

### PR types New features ### PR changes OPs ### Describe 将multihead_matmul convert到oss:主要是针对下面的情况,attention的输入里每个batch的length都一样,这样可以不用设定mask, pos, mask等全局变量,提升性能也顺便解决用户使用不方便的特性。

### PR types Function optimization ### PR changes Others ### Describe This PR make more api run in fast through path in eager mode

### PR types New features ### PR changes APIs ### Describe solve this https://github.com/PaddlePaddle/Paddle/issues/44073#task23 docs - https://github.com/PaddlePaddle/docs/pull/5107

PaddlePaddle Hackathon
contributor

### bug描述 Describe the Bug 环境配置:单机四卡 import lightgbm as lgb notebook 上可以正常使用CPU训练 lgb 参数配置: params = { ............ "device":"gpu", "gpu_platform_id":-1 } 运行命令: python -m paddle.distributed.launch --gpus="0,1,2,3" run001.py 报错信息: ''' aunch...

status/new-issue
type/bug-report

### bug描述 Describe the Bug imag_vin是一张(155, 1355, 3)维度的图片 使用gpu 3090 rec_client = pdx.deploy.Predictor('./model_rec', use_gpu=True) s1 = time.perf_counter() # results = rec_client.predict(image_vin) results = rec_client.predict([image_vin, image_vin, image_vin, image_vin, image_vin, image_vin, image_vin,...

status/new-issue
type/bug-report