search

Ghustwb / MobileNet-SSD-TensorRT

Accelerate mobileNet-ssd with tensorRT
188 stars 91 forks source link

train.prototxt of MobileNet-SSD-TensorRT for custom training #4

Closed PiyalGeorge closed 5 years ago

PiyalGeorge commented 5 years ago

Hi @Ghustwb , I've been looking into this repo, its an awesome work, i've been looking for the exact same. But I'm trying to train a caffe model with my custom dataset. Please help me

Ghustwb commented 5 years ago

Hi How can I help you

On 10/29/2018 15:28，PiyalGeorgenotifications@github.com wrote：

Hi @Ghustwb , I've been looking into this repo, its an awesome work, i've been looking for the exact same. But I'm trying to train a caffe model with my custom dataset. Please help me

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.

PiyalGeorge commented 5 years ago

@Ghustwb Thanks for the response. i've trained my model with dataset using https://github.com/chuanqi305/MobileNet-SSD . I've seen the deploy.prototxt file in your repo. i dont know how to modify the train.prototxt , test.prototxt files. Can you please help me providing train.prototxt and test.prototxt to train on my custom dataset?

Ghustwb commented 5 years ago

OK I think you should read the repo( https://github.com/chuanqi305/MobileNet-SSD )carefully,there is a script file,help you generate the train.prototxt,test.prototxt,deploy.prototxt

On 10/29/2018 17:08，PiyalGeorgenotifications@github.com wrote：

Thanks for the response. i've trained my model with dataset using https://github.com/chuanqi305/MobileNet-SSD . I've seen the deploy.prototxt file in your repo. i dont know how to modify the train.prototxt , test.prototxt files. Can you please help me providing train.prototxt and test.prototxt to train on my custom dataset?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.

PiyalGeorge commented 5 years ago

@Ghustwb , Yeah i got it, and using that i have trained a model with my custom dataset for myself. but in the deploy.prototxt file in your repo has fields like 'IPlugin', which is causing me this error- Check failed: registry.count(type) == 1 (0 vs. 1) Unknown layer type: IPlugin

PiyalGeorge commented 5 years ago

Hey, @Ghustwb When i use my caffemodel generated using custom dataset, do i need to run cmake commands again inorder to get the output?

yiboliu31 commented 5 years ago

@PiyalGeorge IPlugin means the layer that TensorRT doesn't support ，if you don't use TensorRT ,change it to your original layer name

PiyalGeorge commented 5 years ago

Hi, @yiboliu31 , thanks for responding. I'm quite new to deep learning. I have trained a model using my custom dataset in mobileNet-SSD. Its working successfully. My problem is - I'm trying to implement the same in TensorRT in Jetson Tx2. But i don't know how to do that. Please kindly help me. How to make use of your repo. Please.