djl
djl copied to clipboard
An Engine-Agnostic Deep Learning Framework in Java
How to define BPRLoss and MarginLoss by DJL? Does there have some samples? For PairWise data in recommender systems, like user, pos_item, neg_item I am already see the source of...
IdEmbedding userEmbedding = new IdEmbedding.Builder() .setDictionarySize(userCount) .setEmbeddingSize(64) .build(); userEmbedding.initialize(manager, DataType.FLOAT32, usersND.getShape()); userEmbedding.forward(ps, new NDList(manager.create(new int[] {1,2,3,4})), true).singletonOrThrow(); Exception in thread "main" java.lang.UnsupportedOperationException: Not supported! at ai.djl.ndarray.BaseNDManager.invoke(BaseNDManager.java:285) at ai.djl.nn.transformer.MissingOps.gatherNd(MissingOps.java:31) at ai.djl.nn.transformer.IdEmbedding.forwardInternal(IdEmbedding.java:72)...
## Description 1. MxNet shape uses `int` and then casts results into `long` which crashes when the data has a shape larger than `int`. This does not happen for `PyTorch`...
I want to use bert with DJL. Before finetune, it's ok, while I finetune bert model, it has an error: org.tensorflow.exceptions.TFInvalidArgumentException: Input to reshape is a tensor with 128 values,...
## Description When using the framework I found that constructing an empty NDArray then using the NDArray.set function to set values one by one was a bottleneck in the code....
## Description I implement a ssd. It can run without Exception, but zhe gradients are allways zeros. djl version is 0.12.0 ### Expected Behavior there should be some gradients which...
hey i m following pikachu detection for creating my own dataset ...Dogs Breed Detection + claasification heres my Data prepearation code: **//OBJECT DETECTION DATASET public class GestureDataset extends ObjectDetectionDataset {...
I'm researching on deep learning frameworks on mobile phone recently.Does this framework support traing on Android os?Thanks.
…Class" style check. See issue #937. ## Description ## This change was discussed in issue #937 I am working on data with labels but do not use those labels for...