simple-faster-rcnn-pytorch

simple-faster-rcnn-pytorch copied to clipboard

训练好的模型在哪里保存啊？

1

wmrenr

could anyone write the Dependencies versions that successfully run training?

Qasem84

I wonder how to convert the pretrained model in checkpoints to onnx format.

JugramAlucard

Dujuniorrr

i am confused and i cant give my answer during any research. roi pooling involves non-differentiable operations like indexing(quantizing the coordinate(like 3.5) to integers(3)). However why we detaching the proposals?...

sanhai77

step by step understanding approximate joint training method #192

5

i don't understand exactly approximate joint training method. i know RPN and detector merged as a one network during training. the forward path is started pre trained conv network and...

sanhai77

out of memory 训练的时候显存一直在增长

4

已经尝试的方法： loss.item()没问题 dataloader加载数据也没有增长数据。

songhat

About concatenate roi and bbox

3

in ProposalTargetCreator function in [creator_tool.py](https://github.com/chenyuntc/simple-faster-rcnn-pytorch/blob/master/model/utils/creator_tool.py), why do we need to concatenate ROI and bbox at the beginning? `roi = np.concatenate((roi, bbox), axis=0)`

txytju

train my own data

2

if i train my own data,where should i modify

njustymk

怎么打印fps

1

怎么打印fps

yapengLI1