siriusmehta

Can we fine-tune on custom dataset

1

Hi Thanks for sharing the amazing repo. I was wondering if we can fine-tune resnet for any other custom dataset except UCF and others. Can any one throw light on...

Example or any Support for Tensorflow_datasets

1

Hi, I would highly appreciate any sample code for how can we use tf Explain GradCam feature for tensorflow dataset. Best Regard Vishal

Training retinanet with a different backend

2

Hi, I am exploring how to train retinanet with a different backend (mobilenet) . Can someone throw some light on it how it can be done ?

Retinanet with mobilenet backend

6

@srihari-humbarwadi Can we adapt to different backend for RetinaNet such as mobilenet ?

Normalizing sensor data

1

Hi @all , I just tried training a model for Punch/Flex gesture and was not able to understand why are we adding 4 and dividing by 8 similarly for 2000...

Hi @CaoWGG , In file included from /home/sirius/Desktop/TensorRT-CenterNet/onnx-tensorrt/OnnxAttrs.cpp:25:0: /home/sirius/Desktop/TensorRT-CenterNet/onnx-tensorrt/onnx2trt_utils.hpp:246:11: warning: ‘nvinfer1::Dims::type’ is deprecated [-Wdeprecated-declarations] shape.type[0] = nvinfer1::DimensionType::kCHANNEL; ^~~~ In file included from /usr/include/aarch64-linux-gnu/NvInferRuntime.h:59:0, from /usr/include/aarch64-linux-gnu/NvInfer.h:53, from /home/sirius/Desktop/TensorRT-CenterNet/onnx-tensorrt/OnnxAttrs.hpp:26, from /home/sirius/Desktop/TensorRT-CenterNet/onnx-tensorrt/OnnxAttrs.cpp:23: /usr/include/aarch64-linux-gnu/NvInferRuntimeCommon.h:225:47:...

Hi @zhreshold I tried fine tuning the slowfast model on two classes and trying to load the export_tvm model .so, .param, json all the files but getting the below error:...

IndexError: Index out of range. size: 4, got index 4

3

Hi, I am trying to convert slowfast_4x16_resnet50_kinetics400 to TVM using the below notebook: https://github.com/zhreshold/cvpr2020-videomodeling-deployment/blob/master/03_deploy_video_model_to_tvm.ipynb But getting below error in line: gcv.utils.export_tvm(model_name, net, data_shape=(1, 3, 224, 224), preprocess=TvmPreprocess(), target='cuda', ctx=mx.gpu(0), opt_level=3,...