Crystal Liu

when I try to run download_and_convert_data.py , then Traceback (most recent call last): File "download_and_convert_data.py", line 41, in tf.app.run() File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/platform/app.py", line 48, in run _sys.exit(main(_sys.argv[:1] + flags_passthrough)) File "download_and_convert_data.py",...

Hi, @mystic123 Thanks for the great work. I found it may be a bug in line 220 of yolo_v3.py: ``` # transpose the inputs to NCHW if data_format == 'NCHW':...

Thanks for the great work. I trained it on Titan V, but it is very slow. Do you have any idea of how to train it faster?

Hi, @fsx950223 thank you very much for the great work. When I run the code, I found some bugs, such as (1) line 33 in voc_text.py, rand may cause some...

non_max_suppression耗时过长，最后死循环

1

我自己补充了一下缺失的prune_util.py来训练，但是发现non_max_suppression计算的特别慢，计算到99%的时候死循环，也就是detections.size(0)一直保持不变。请问您有遇到过这个问题么? eriklindernoren工程下有人讨论过这个问题，但是没有什么解决办法。

assertionerror

8

在运行train.py的时候，执行到evaluate，得到assert sample_metrics 为空怎么办呢？为什么会有这个问题呢？ 有时候 `outputs = model(imgs)`的outputs不为空，但是经过`outputs = non_max_suppression(outputs, conf_thres=conf_thres, nms_thres=nms_thres)`之后outputs就为空了。

What is the meaning of CBL_idx? What is the difference between pruned_model and compact_model in test_prune.py?

5

有点看不懂CBL_idx，这是什么index？ What is the difference between pruned_model and compact_model in test_prune.py? 谢谢

Yuan2-2B-hf is slow in MTL iGPU

2

Test code: ``` import torch, transformers from transformers import AutoTokenizer,LlamaTokenizer from bigdl.llm.transformers import AutoModelForCausalLM import torch import intel_extension_for_pytorch as ipex import time print("Creat tokenizer...") model_path1 = r"C:\Users\test\Downloads\YuanLLM\Yuan2-2B-Janus-hf" tokenizer = LlamaTokenizer.from_pretrained(model_path1,...

RuntimeError: Native API failed. Native API returns: -999 (Unknown PI error) -999 (Unknown PI error)

2

MTL 32GB mem baichuan13B got error. The input prompt and input token can be found following： ``` token count input: 1036 input: 根据以下信息回答问题。 回答的语种与问题的语种一致。 ADS 使用向导 通过 ADS 考试 为了确保您能充分理解...

speech_paraformer-large_asr_nat-zh-cantonese-en-16k-vocab8501-online 如何导出onnx模型？

5

``` from funasr_onnx import Paraformer,CT_Transformer import time asr_model_path = "./speech_paraformer-large_asr_nat-zh-cantonese-en-16k-vocab8501-online" print("model: ", asr_model_path) asr_model = Paraformer(asr_model_path, batch_size=1, quantize=True) wav_path = ['./chatglm2.wav', 'chengyu.wav', 'chinese_10s_16k.wav', 'hongqiao.wav', 'Sheldon_10s.wav','jfk.wav'] for wavfile in wav_path: st...