-
你好，我看官方提供了TensorFlow ，Pytorch，ONNX的模型文件，然后MNN也提供了用MNNConvert转模型的方法。
请问是如何转成rvm_mobilenetv3_fp32-480-640.mnn这种480*640的模型的呢，不知道是模型转换的哪个参数呢？
然后我看模型转换里面还有--fp16 可以将将conv/matmul/LSTM的float32参数保存为float16，…
-
how about the results of some lightweight backones such as mobbilenetv3。I used resnet50 GFL as teacher and mobilenetv3 GFL as student but got very small improvement in my own datasets. Any guideline …
-
Thanks for source code ,
- Can you tell me the performance on coco or voc on backbone mobilenetv2?
- I want to add backbone mobilenetv3, I have to do it.
- I think the backbone DLA is quite good…
-
First of all, thanks for sharing the excellent models. I've tested the resnet101 based models, it works perfect. And I noticed that you guys also provide the mobilenet-v3-large model (mobilenetv3-larg…
-
mobilenetv3加LiteR-ASPP的分割网络。网络模型已经从pytorch转换到onnx再转换到ncnn压缩到几百K左右，但是在海思3519上测试，发现实时性性不太行需要200ms左右。所以尝试进行int8量化，但是量化后结果，相较于以前存在很多误检 不知道如何解决。现在提供fp16和int8的模型 以及测试代码，希望大佬能帮忙解决一下，谢谢。
链接: https://pan.baid…
-
Hi,
When I use the [benchmark script](https://www.tensorflow.org/lite/performance/measurement#native_benchmark_binary) and [benchmark apk](https://www.tensorflow.org/lite/performance/measurement#andr…
-
First of all, thank you for your new work,I'd like to know a few questions
1. How does the speed compare to mobilenetv3, real time (FPS) ?
2. Whether the aspp module in CNN can be replaced?
3. The…
-
我在pytorch->onnx->ncnn对mobileNetv3模型转换的过程中，遇到了：1、onnx2ncnn步骤有不支持问题，使用了onnxsim对onnx模型进行了处理后，再进行onnx2ncnn时转换OK，得到解决；2、我在构建应用，使用extractor.extract()时需要blob的name或index，当我使用ncnn2mem将文件转换成*.id.h打开查看最后的index并使…
-
Hi, I tried to train mobilenet v3 following the configuration in [page]( https://github.com/tensorflow/models/tree/master/research/slim/nets/mobilenet). However, some of the configurations are not def…
-
这是我的model的forward
这是报错
![CI4BHW JTC5 XECR%Q QGKM](https://user-images.gi…