-
## Description
使用paddle.jit.save导出模型时指定了两个入参input_ids、token_type_ids，但加载模型时JniUtils.getInputNames(predictor)似乎只识别到input_ids。
### Error Message
Caused by: java.lang.IllegalArgumentException: Input…
-
## Description
(A clear and concise description of what the bug is.)
While packaging OpenJDK for Homebrew, it was noticed that DJL Serving's tarball, downloaded from https://publish.djl.ai/djl-ser…
-
Hi!
I'm getting error on my java application when I trying to start it.
`detailMessage -> Model "sentence-transformers/all-mpnet-base-v2" on the Hub doesn't have a tokenizer`
I'm ussing `redis-…
-
## Description
We have our ONNX model, and we're trying to use it in our app. Running the project from the IDE works just fine, but running the jar file produces this error. We debugged this issue an…
-
I was wondering if this supports the tokenizer for XLM as well? Seems like the original tokenizer uses sacremoses - and can we expect that to be incorporated in this repo?
-
## Description
(A clear and concise description of what the feature is.)
Will this change the current api? How?
Who will benefit from this enhancement?
## References
- list reference an…
-
When I used djl-serving 0.19.0, I found that the predicted result of the model was garbled in Chinese.
Later, I found that the image of ubuntu does not support Chinese. Later, I added the installat…
-
Hi,
I tried running the code in Sagemaker studio with the same instance and kernel. But the endpoint creation is failing after around 19 minutes. I tried using `"InstanceType": "ml.g4dn.16xlarge"`…
-
I got an error message as the issue: https://github.com/deepjavalibrary/djl/issues/1618. In DJL, I used cpu to load the model, and the djl has successfully recognized the `libtorch_cpu.so`. However th…
-
## Description
(A clear and concise description of what the bug is.)
I want to use DJL to write a cat and dog classification, I used `RandomAccessDataset`。
But when I read the image and convert it …