xLuge

> Hi, > Could you share your training settings in detail? > To reproduce the result in pascal voc leader board, you need to follow these steps: > > 1....

- > how r u fine tuning the model i just fine tuning model like that https://hangzhang.org/PyTorch-Encoding/model_zoo/segmentation.html#pascal-voc-dataset but according to the deeplab3, we should freeze bn after coco pretrain.

> You may refer to the original paper. FYI, gluoncv has the benchmark results. The performance should be similar to this repo https://cv.gluon.ai/model_zoo/segmentation.html#pascal-voc-dataset thanks! In addition, I submitted the results...

> Actually I don't know how to make dateset_mead.json and dataset_exp.json, i use an dataset.json which contains 25 dimensional information for every image to replace. finally it can train, but...

hi，I have the same doubt. Have you trained yet?

> Hello, a quick question, I am trying to train this model from scratch, and it seems the first-stage training is ok, and I have gotten a promising result. While...

> > > > Hello, a quick question, I am trying to train this model from scratch, and it seems the first-stage training is ok, and I have gotten a...

> > > > > > Hello, a quick question, I am trying to train this model from scratch, and it seems the first-stage training is ok, and I have...

> > > > > > > Hello, a quick question, I am trying to train this model from scratch, and it seems the first-stage training is ok, and I...

> > > > > > > > > Hello, a quick question, I am trying to train this model from scratch, and it seems the first-stage training is ok,...