sophgo/sophon-demo

Issues

• 可以支持多模态模型么，比如InternVL 等

  #49 opened 7 months ago by 17656178609
  3

• 您好，请问什么时候能支持glm-4v-9B这个模型呢？

  #57 opened 5 months ago by tang799319844
  4

• sorry for open issue in wrong repo

  #56 opened 5 months ago by XiaotaoChen
  0

• wenetspeech decoder模型转换报错

  #54 opened 6 months ago by l99500
  3

• 无法复现yolov10s结果

  #55 opened 6 months ago by nobody-cheng
  1

• minicpm能否提供api访问的方式

  #53 opened 6 months ago by l99500
  0

• DeepSort中FeatureExtractor中的转换系数与模型转换脚本不一致?

  #51 opened 7 months ago by crouchggj
  8

• SDXL模型转Bmodel失败

  #52 opened 7 months ago by Qs315490
  3

• Resnet模型转成bmodel之后，检测精度下降比较严重

  #50 opened 7 months ago by xc597
  1

• 运行llama2-7b_int8_1dev.bmodel模型时报错

  #43 opened 8 months ago by 17656178609
  2

• samples里面没有Qwen2

  #48 opened 7 months ago by tzhang2014
  0

• 转换模型中出现Cannot open libusercpu.so, disable user cpu layer.

  #47 opened 8 months ago by xc597
  1

• 运行yolov5-fuse的问题

  #46 opened 8 months ago by Onlydrink
  3

• TPU_MLIR可以安装在aarch64架构的系统上吗

  #45 opened 8 months ago by xc597
  3

• 跑python yolov5例程的问题

  #44 opened 8 months ago by Onlydrink
  1

• qwen2推理异常

  #42 opened 8 months ago by storyxlx
  1

• 运行Bmodel失败

  #41 opened 8 months ago by SongIdea
  1

• yolov8 cls 啥时候能支持？

  #38 opened 8 months ago by heqq895
  1

• ChatGLM3转bmodel时文件缺失

  #40 opened 8 months ago by 394988736
  0

• 没有PaddleLite的demo吗

  #39 opened 8 months ago by 394988736
  0

• ChatGLM3转bmodel

  #37 opened 9 months ago by githubzjj1
  1

• BOOM and FAILED

  #36 opened 9 months ago by szxysdt
  4

• 是否有支持dynamic_axes输入的计划？

  #34 opened 10 months ago by szxysdt
  2

• yolov8 error in se5 upper segment

  #35 opened 10 months ago by ZhaoMonica
  1

• 能不能在之后的版本增加语音合成的例子

  #17 opened a year ago by XMUEDUx
  2

• AVFrame 转 BM_IMAGE 遇到了问题

  #33 opened a year ago by ijiami-01
  1

• 我拿着AVFrame的指针实例, 怎么转换成bm_image?

  #32 opened a year ago by ijiami-01
  0

• sophon-demo的yolov5_demo里，如果同一个模型重复加载多次，会在内存里存在多份吗？还是算能SDK底层会判断有重复模型，就不重复加载了。

  #25 opened a year ago by ijiami-01
  1

• yolov8n单通道图，batch>1执行in8,f16,f32demo推理均存在问题

  #30 opened a year ago by jiwenfei
  10

• 导出yolov8n-pose模型精度问题

  #31 opened a year ago by jiangzhiwei2018
  1

• 使用代理无法通过dfss下载文件

  #21 opened a year ago by CHNtentes
  2

• YoloV5怎么用2个模型对1个图片(视频)做推理并输出推理后的图片。

  #22 opened a year ago by ijiami-01
  2

• SE7运行demo报错

  #28 opened a year ago by GCwithH2O
  1

• 有没有max_batch大于1个的yolov5模型？

  #27 opened a year ago by ijiami-01
  1

• yolov8 pose 模型支持吗

  #23 opened a year ago by xiaoxiaodetian
  1

• 如何利用代码达到，让一个模型支持多个视频流？

  #24 opened a year ago by ijiami-01
  1

• 这个文生图的性能测试的单位是秒吗？

  #29 opened a year ago by yuyun2000
  1

• yolov8 demo使用的是ultralytics哪一个版本

  #26 opened a year ago by nobody-cheng
  0

• 使用samples里提供的chatglm2及llama2刷机包刷镜像后灯没有闪烁，没刷成功；使用官网0.4.6的镜像可以刷机成功

  #20 opened a year ago by shanchenjie
  5

• Yolov5 的导出 guide 中的修改 yolo.py 是导出时还是训练就要这样改？

  #19 opened a year ago by tianhanghai
  1

• YOLOv8 在SE5上推理结果异常

  #16 opened a year ago by Snowba1l
  3

• SE7盒子 直接使用sophon-demo推理报错

  #18 opened a year ago by SweatLin
  1

• 请问有转换Mask RCNN模型的例子吗

  #10 opened a year ago by coo113
  1

• sophon-demo/sample/CenterNet/scripts/download.sh中下载链接失效

  #12 opened a year ago by suncheng-guihub
  1

• 官方llama2-7b模型转换bmodel推理结果异常

  #14 opened a year ago by tensorflowt
  17

• 请问一下，如果我目前使用中文微调模型Llama2-Chinese-7b-Chat算法模型，是否当前的llama2模型转换及推理代码也适用？

  #15 opened a year ago by tensorflowt
  0

• 下载Llama2-7B失败

  #13 opened a year ago by tensorflowt
  5

• yolov8支持哪些系列模型？

  #11 opened a year ago by 294978174
  1

• ff_decode.hpp 函数和 ff_decode.cpp 不一致

  #7 opened a year ago by joeduff
  1

• ff_decode.cpp该例程在业务逻辑上有问题

  #8 opened a year ago by kunknu
  1