djl icon indicating copy to clipboard operation
djl copied to clipboard

An Engine-Agnostic Deep Learning Framework in Java

Results 322 djl issues
Sort by recently updated
recently updated
newest added
trafficstars

Here is my pom.xml: ``` ai.djl api 0.18.0 ai.djl.pytorch pytorch-engine 0.18.0 ai.djl.pytorch pytorch-native-auto 1.9.1 ai.djl.pytorch pytorch-native-cpu linux-x86_64 1.11.0 ai.djl.pytorch pytorch-jni 1.11.0-0.18.0 ``` My code is: ``` Engine engine = Engine.getEngine("PyTorch");...

bug
more-information-needed

The change detection model we are currently using is to input two images for detection, but how can we input two images simultaneously using DJL?

enhancement
more-information-needed

## Description When dealing with multithreaded tasks, the following three Predictor objects should not be created as singletons. If I use singletons when testing multithreaded tasks, my application will crash...

bug
more-information-needed

## Description I see the configuration as follows at this [link](https://docs.djl.ai/engines/pytorch/pytorch-engine/index.html#linux-gpu) ``` ai.djl.pytorch pytorch-native-cu118 linux-x86_64 2.1.1 runtime ai.djl.pytorch pytorch-jni 2.1.1-0.26.0 runtime ``` I couldn't find version 2.1.1 at the following...

bug

## Description It seems that the official already support the mac arm arch. Hope it will be supported. ## References - list reference and related literature - list known implementations...

enhancement

An error occurs when running the onnx model ``` 2024-01-09 11:06:18.1575181 [E:onnxruntime:, sequential_executor.cc:346 onnxruntime::SequentialExecutor::Execute] Non-zero status code returned while running ReduceMax node. Name:'ReduceMax_1745' Status Message: D:\a\_work\1\s\onnxruntime\core\providers\cpu\reduction\reduction_ops.cc:747 onnxruntime::ValidateKeepDims keepdims was false....

bug

## Description 请问我在CentOS Linux release 7.9.2009 (Core)的环境中,测试一个在我个人电脑中调试成功的demo,但是在运行时候,提示这个错误,请问这个是怎么处理,是不是不同操作系统的,需要配置很多不同的环境? 我执行的命令是:java -jar paddledjl-demo.jar ./89-inference-clas.zip ./1.jpg 收到的返回结果是 Exception in thread "main" java.lang.UnsatisfiedLinkError: /root/.djl.ai/paddle/2.3.2-cpu-linux-x86_64/libpaddle_inference.so: /lib64/libstdc++.so.6: version `CXXABI_1.3.8' not found (required by /root/.djl.ai/paddle/2.3.2-cpu-linux-x86_64/libpaddle_inference.so) at java.lang.ClassLoader$NativeLibrary.load(Native Method)...

enhancement

## Description 目前以YoloV8为基座开发了一系列的模型,用来支持分割、POSE检测等场景,djl的YoloV8Translator能否进行相关增强,以适配不同的下游任务

enhancement

i want to train image keypoints model,Already have labeled data,how to train

enhancement

As of today (2023-12-22), paddlepaddle is at version 2.5.2, and the paddle engine in djl has not been updated

enhancement