MagicSource

Hi, I don't for what reason, by calling your API got wrong charset... ``` {Code: 0, Message: 获取数据成功, Data: [{id: 1, sort: 161598, title: 知乎}, {id: 2, sort: 59138, title:...

Does there any scripts convert coco to image list format?

Message type "caffe.SolverParameter" has no field named "eval_type".

4

``` Message type "caffe.SolverParameter" has no field named "eval_type". ```

Wrong detection result?

3

I download the 73000 iter caffemodel and converted into no_bn model. when run demo (changed the net and model of course), the result are wrong:  The result are messed...

does tensorflow 2.0 finished?

3

does tensorflow 2.0 finished?

Just for enchancement, this repo are helpful for build realtime segmentation applications, but now that Pytorch official has been depracted cffi, would that be a good idea to upgrade into...

Loss is not decreasing

19

I have trained ssd with mobilenetv2 on VOC but after almost 500 epochs, the loss is still like this: ``` 517/518 in 0.154s [##########] | loc_loss: 1.4773 cls_loss: 2.3165 ==>Train:...

All the ssds method are based on 300, can it changed into 500 or 512?

12

Can the input image size be changed into another?

trained model there are no detections

2

Training from scratch without pretrained model, the model can not predict anything:

memory leak when run inference

3

 the memory are increasing massively..... ``` python scripts/demo_inference.py --indir ../supp_assets/bilibili/images --outdir ../supp_assets/bilibili/images_res --cfg configs/coco/resnet/256x192_res50_lr1e-3_1x-duc.yaml --checkpoint pretrained_models/fast_421_res50-shuffle_256x192.pth --save_video --save_img --flip --min_box_area 300 ```