nnstreamer/nntrainer

Issues

• [LLaMA] uncaught error while running! details: [LayerImpl] Unknown Layer Properties count 2

  #2782 opened 25 days ago by dyna-bytes
  6

• [ Unittest / GPU ] GPU Unittest failure (12 tests ) in unittest_layers

  #2774 opened a month ago by EunjuYang
  2

• [ GPU ] Enable x86 OpenCL with some fixes

  #2776 opened a month ago by EunjuYang
  3

• Enable Android Test with meson build

  #2775 opened a month ago by DonghakPark
  1

• [ Tensor ] more types of Tensors are required.

  #2733 opened a month ago by EunjuYang
  4

• CLPlatform self-assignment

  #2763 opened a month ago by piotrrak
  2

• [ GPU ] GPU Kernel creation time

  #2723 opened 3 months ago by EunjuYang
  4

• [Mixed Precision] Apply Gradient Clipping on Mixed Precision

  #2746 opened 2 months ago by DonghakPark
  2

• CacheLoaderTest fails occasionally in GitHub Action CI

  #2486 opened 2 months ago by djeong20
  2

• Main branch error after merging #2704

  #2715 opened 3 months ago by djeong20
  2

• Remove Cblas params from `tensor` related files

  #2682 opened 3 months ago by skykongkong8
  3

• ConcatLayer performance issue

  #2701 opened 3 months ago by djeong20
  1

• Issues and Questions about Execution of LLaMA using NNTrainer

  #2561 opened 7 months ago by Deeksha-20-99
  13

• Using Bfloat16 GEMM from OpenBlas

  #2668 opened 4 months ago by skykongkong8
  2

• Abusive usage of meson dependency system.

  #2673 opened 4 months ago by myungjoo
  1

• Need to check version function

  #2435 opened 5 months ago by SeoHyungjun
  3

• Support Convolution&Batchnorm Fusing for Optimized Inference Mode

  #2646 opened 5 months ago by DonghakPark
  3

• input-order-mismatch issue in multi-input layer.

  #2660 opened 5 months ago by EunjuYang
  2

• [ HGEMM ] Half-Precision GEMM Roadmap

  #2583 opened 6 months ago by skykongkong8
  4

• Add `-Denable-fp16=true` build to CI

  #2560 opened 5 months ago by heka1024
  6

• Support encoder on Ubuntu & Tizen

  #2632 opened 5 months ago by baek2sm
  1

• Add the calc_derivative and unittest of quick gelu

  #2626 opened 6 months ago by baek2sm
  1

• Bug in `max_abs()` function in FP16 Tensor

  #2593 opened 6 months ago by skykongkong8
  3

• [ Tensor ] Accelerate fp16 matrix transpose with SIMD

  #2582 opened 6 months ago by skykongkong8
  3

• Issue in running the resnet18 example.

  #2600 opened 6 months ago by udit01
  4

• Support loading weights from pytorch model

  #2616 opened 6 months ago by baek2sm
  1

• Support hyper parameter for activation layer

  #2614 opened 6 months ago by baek2sm
  1

• Running examples on PC issue

  #2602 opened 6 months ago by udit01
  3

• Q&A] How to solve build failure for flatbuffers's Table.

  #2592 opened 6 months ago by Hyunjun85
  2

• [ GEMM ] HGEMM noTrans case

  #2488 opened 6 months ago by skykongkong8
  6

• DRL algorithm with api

  #2528 opened 8 months ago by eightreal
  8

• Normalization Layer: FP32/FP16 should NOT be determined at compile-time

  #2408 opened 10 months ago by myungjoo
  5

• Add Depthwise 2D Convolution Layer

  #2520 opened 8 months ago by DonghakPark
  1

• Some confusion about random dataset

  #2525 opened 8 months ago by eightreal
  8

• Knowledge Embedding Interface Specification for RAG in NNTrainer

  #2519 opened 8 months ago by hokeun
  2

• Build fails with `Dplatform=android`

  #2517 opened 8 months ago by dkjung
  1

• Random Idea for Future Features: G-LoRA on NNTrainer.

  #2496 opened 9 months ago by myungjoo
  1

• Channel Last Tensor save/read fails occasionally

  #2494 opened 9 months ago by djeong20
  1

• Further optimization ideas (for faster calculation)

  #2485 opened 9 months ago by myungjoo
  1

• SIGSEGV in Unit Test (occasional): multiply_04_n

  #2459 opened 9 months ago by myungjoo
  4

• [CI] migrate TAOS CI --> gitaction CI

  #2463 opened 9 months ago by DonghakPark
  1

• Consider Git Repo Separation (Application) After 2023 Release

  #2314 opened a year ago by myungjoo
  2

• Dependency Clean Up Request

  #2333 opened 10 months ago by myungjoo
  3

• Multi-head attention bug: _MASK_NUM determined at build time, not at run time.

  #2407 opened 10 months ago by myungjoo
  1

• Half-Precision SGEMV() and SGEMM() not support CblasColMajor

  #2399 opened a year ago by djeong20
  1

• [ layer ] Erasing `_FP16` code block in each layer

  #2362 opened a year ago by skykongkong8
  2

• [TF Lite Export] Layer's Props Inconsistency issue

  #2371 opened a year ago by DonghakPark
  1

• [blas/neon] Support various data types for `scopy`

  #2332 opened a year ago by djeong20
  4

• Tizen 7.0 build

  #2337 opened a year ago by eightreal
  7

• List of embedded layers that need to be implemented to support the diffusion model

  #2338 opened a year ago by baek2sm
  1