djl

djl copied to clipboard

WIP: Make MPS default device for macOS M1

1

Currently MPS has many limitations: 1. mapLocation cannot be used to load model onto MPS 2. module that serialized with MPS device cannot be loaded as is, it can only...

frankfliu

## Description ## Brief description of what this PR is about - If this change is a backward incompatible change, why must this change be made? - Interesting edge cases...

xyang16

## Description ## Brief description of what this PR is about - If this change is a backward incompatible change, why must this change be made? - Interesting edge cases...

frankfliu

Updated llama.cpp engine to version b2581

2

## Description ## The llama engine code is now compatible with the b2581 release of llama.cpp repository, upgrading from b1696. - Note : This is just a compatibility update, some...

MatPere

[llama.cpp] Update llama.cpp to latest release b2581 (2024-03-30)

1

In order to get support for BERT based sentence embedding models like [BAAI/bge-base-en-v1.5](https://huggingface.co/BAAI/bge-base-en-v1.5), [mixedbread-ai/mxbai-embed-large-v1](https://huggingface.co/mixedbread-ai/mxbai-embed-large-v1), or others, update [llama.cpp](https://github.com/ggerganov/llama.cpp) from [b1696](https://github.com/ggerganov/llama.cpp/releases/tag/b1696) (2023-12-12) to the current latest release [b2581](https://github.com/ggerganov/llama.cpp/releases/tag/b2581) (2024-03-30). BERT support...

howlger

This feature translates DJL blocks into tensorflow layers. Functional pointers cannot be translated and would need amendment, either in the DJL model (use Blocks.reluBlock() instead of Blocks::relu) or in the...

patins1

Support for Radeon GPUs?

6

## Description Support Radeon GPUs to be used for accelerating inferencing and training Will this change the current api? How? No Who will benefit from this enhancement? All users of...

viggy96

Logistic regression : Accuracy of 1.0 (is there an error in my code ?)

7

## Description Hello. I'm trying to build a simple logistic model with DJL. I have an accuracy of 1 with my logistic regression model. ### Expected Behavior I expected to...

MohamedLEGH

How works the function EasyTrain.evaluateDataset() ?

3

## Description When I run `EasyTrain.fit(trainer, numEpochs, trainingSet, null);` I get the metrics in the standard output like this: ```bash Training: 100% |████████████████████████████████████████| Accuracy: 1,00, SigmoidBinaryCrossEntropyLoss: 121,25 Validating: 100% |████████████████████████████████████████|...

MohamedLEGH

Memory Leak

4

## Description In a SpringBoot Web application, the memory is not released after the method execution is completed. ### Expected Behavior The memory is not released after the method execution...

BOOMXIKA