-
## Description
Possible cause: can't load AMD 64-bit .so on a MIPS64 LE-bit platform)
-
## Description
NDManager 的 arange 方法在 pytorch-engine、tensorflow-engine 和 mxnet-engine 的输出不一样。
pytorch：
shape [12]: ND: (12) cpu() int32
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
tensor…
-
Python's UNET model works very well. How can I load it into djl
-
Will DJL support time series preprocess ?
Hi, there are some preprocess implementations for cv, nlp in ai.djl.modality,
and currently I am working on time series model inference with djl, so I wonde…
-
ai.djl.translate.TranslateException: ai.djl.engine.EngineException: isTuple()INTERNAL ASSERT FAILED at "C:\\actions-runner\\_work\\pytorch\\pytorch\\builder\\windows\\pytorch\\aten\\src\\ATen/core/iva…
-
## Description
When PyTorch is installed on my system (an up-to-date ArchLinux), djl is trying to load the library `pytorch/1.13.0-cpu-precxx11-linux-x86_64` which does not work. But if I uninstall P…
-
IdEmbedding userEmbedding = new IdEmbedding.Builder()
.setDictionarySize(userCount)
.setEmbeddingSize(64)
.build();
userEmbedding.initializ…
-
I don't know if this is a bug. I'm trying to follow the [official tutorial](https://docs.djl.ai/jupyter/rank_classification_using_BERT_on_Amazon_Review.html) using pytorch engine.
Here is my code …
-
## Description
During training on GPU I experience
- 4 MiB memory leak on GPU per epoch (looks constant)
- duration increase about 1 min per epoch (looks linear)
### Expected Behavior
no memory…
-
## Description
Methods of NDArray, which use the getScalar()-method internally are causing a memory leak (of native memory), because getScalar() creates an NDResource, which is not closed internally …