xiaoxiongli

it seems the inference is very slow on my linux server?

5

Hi, Dear NJU-Jet my linux server: several 2.6GHz CPU + several V100, and I run the **generate_tflite.py** to got a quantized model. and then in function **evaluate**, I add below...

I want to train the model and then I need a int8 onnx model, what should I do step by step?

6

Dear @NJU-Jet : I feel confused that if I use the command "**python train.py --opt options/train/base7.yaml --name base7_D4C28_bs16ps64_lr1e-3 --scale 3 --bs 16 --ps 64 --lr 1e-3 --gpu_ids 0**" to train...

How to deal with QuantizeLinear and DequantizeLinear node when I do qutization using openvino/tnn/mnn?

1

I train a x2 model, and after that I finetune it using QAT training using below command: >"python train.py --opt options/train/base7_qat.yaml --name base7_D4C28_bs16ps64_lr12-3_qat_x2 --scale 2 --bs 16 --ps 64 --lr...

During QAT training for scale = 2, do I need change below "3" to "2" ?

1

I have already trained a model that scale = 2, and after that I wanna to do QAT training for it(scale = 2) using command "**python train.py --opt options/train/base7_qat.yaml --name...

a bug that during training the x2 model, 0811.pt is not generated?

6

Hi Dear @NJU-Jet : it seems have a bug, please see below figure:   and I check the folder data/DIV2K/bin/DIV2K_train_HR/, below figure:  you can see that the it...

it seems have a bug when I train a x2(scale=2) model?

1

Hi Dear @NJU-Jet: I wanna to train a model that scale=2. (as you know, the defaut scale is 3) it seems have a bug here:  you can see that...

How to convert tflite model to onnx model?

3

it is good work and congratulation! I have to two questions that: 1) which framwork do you use: tensorflow or tensorflow_lite? 2) if you use tensorflow_lite, How to convert the...

failed to reproduce your lfw 99.3% precision.

3

Dear happynear, @happynear i failed to reproduce your lfw 99.3% precision. I use your below prototxt and caffemodel download from BaiDu Yun disk: net = caffe.Net('caffe_proto/center_loss_ms.prototxt','caffe_proto/center_loss_ms.caffemodel', 'test'); and run getPairs.m,...

why wider_face_train.mat 's file size is different with the widerface official one?

1

Dear @Seanlinx I download the anotation files from the widerface's official website, and i find that your wider_face_train.mat in folder "wider_annotations" is different with the official one. i do not...

how much time when training P-Net in GPU?

1

it seems very slow to train P-Net in GPU? please see below, is it normal? INFO:root:Epoch[0] Batch [4400] Speed: **27.19 samples/sec** Accuracy=0.885761 LogLoss=0.304886 BBOX_MSE=1.203766 INFO:root:Epoch[0] Batch [4600] Speed: **26.81 samples/sec**...