wangshuai09

Training loss does not converge

4

Hi, i have problem in training process of Pytorch version. When using hed_loss2/bdcn_loss/bdcn_loss2 in losses，training loss is large and does not converge，model can predict edge map but the result is...

Add training support and change lspci for Ascend NPU

3

## Description - `gpu_info=$(lspci 2>/dev/null | grep -E "VGA|Display")` shows the iBMC chip VGA support for Ascend NPU because Ascend NPU has no ability of VGA or Display. So "npu-smi...

I've verified ChatGLM2-6B in HUAWEI Ascend device, ```python from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("THUDM/chatglm2-6b", trust_remote_code=True) model = AutoModel.from_pretrained("THUDM/chatglm2-6b", trust_remote_code=True, device='npu') model = model.eval() response, history = model.chat(tokenizer, "你好",...

train on own dataset

3

I convert my dataset to the coco format，the dataset has no segmentation，so i set the "segmentation" =[] in annotations，and set the cfg.TRAIN.MRCNN_LABELS_TO_KEEP = ( )，i also convert the pretrained coco...

Fix import error

1

This PR fixes `NameError: name 'IterableWrapper' is not defined`. Before, After,

Add NPU test case

3

In order to use torchtext with `NPU` backend, i've run example and unit test cases. All test cases can be work fine but gpu tests are skipped. This PR adds...

[CANN] Add Ascend NPU support

1

### Motivation Hi, thanks for your great work! I've run insightface on my **Ascend NPU Device** and then want to make this contribution for other users using the same hardware....

[CANN] Add doc and docker image

1

- [x] I have read the [contributing guidelines](https://github.com/ggerganov/llama.cpp/blob/master/CONTRIBUTING.md) - Self-reported review complexity: - [x] Low - [ ] Medium - [ ] High

Fix version check for using -mavxvnni

11

### Description Change the `CMAKE_CXX_COMPILER_VERSION` greater than `11` for using '-mavxvnni'. ### Motivation and Context `CMakeFiles/onnxruntime_mlas.dir/root/Git.d/onnxruntime/onnxruntime/core/mlas/lib/x86_64/QgemmU8S8KernelAvx2.S.o cc: error: unrecognized command-line option ‘-mavxvnni’; did you mean ‘-mavx512vnni’?` using `gcc (GCC) 10.3.1`....

[Hardware][Ascend] Add Ascend NPU backend

56

As mentioned in https://github.com/vllm-project/vllm/issues/7692, this PR make Ascend NPU backend available in VLLM. ### RoadMap: - [x] Ascend Executor - [x] Ascend Worker - [x] Ascend Model Runner - [...