jacinto-ai-devkit

jacinto-ai-devkit copied to clipboard

when i traing segmentation i meet a problem: when i set batch_size=8 and iter_size=1,program case a problem "isinstance(x, (list,tuple)) and len(x)

skoomn

caffe-jacinto training memory problem

2

Hello! I try to train ssd using the caffe-jacinto with a Geforce 940MX GPU(2002MB available). At first, I ran the training script(voc0712, 256*256) with batch_size == 16 and i failed....

KurtKoo

Bounding box information in output tensor of onnx model

1

Hi, I trained a model with Regnet backbone with regnetx_800mf arch and RetinanetLite head and converted it to an onnx model. There are 10 output tensors in the onnx model...

PrithviKambhampati

pytorch-jacinto-ai-devkit train time problem

17

hello,I have a question. when I train my model without xnn.quantize.QuantTrainModule, it costs 30minutes one epoch. but when I add xnn.quantize.QuantTrainModule in my train code,it costs 4hours one epoch. they...

WangGangUCAS

about channel_pitch

2

hello, in tidl_import tool, there is a confusing problem: for data layer , when calculate channel pitch, example: data_input 224*224*3, pad_w = 2, pad_h = 2, ti calculate channel_pitch =...

wuzhiyang2016

about clip layer

7

hello, when i use quantize training to train a quantized model, the training code will insert **clip** layer to some layers, there is some problems: 1. for original float model...

wuzhiyang2016

the quantize training, is only decrease the loss of activation quantize loss without weights quantize loss?

11

hello, in **quant_train_module.py** file, i saw a line of code : y.data.copy_(yq.data), this code change the data of relu's output data.data, in order to use it in backword for calculate...

wuzhiyang2016

QuantizeDequantizeG's backward() function doesn't be executed!!!!

7

hello, when i train quantized model by pytorch-jacinto-ai-devkit, a class called **QuantizeDequantizeG**, when doing loss.backward(), it's backword() function is not used,,, the code is in xnn/layers/funtion.py Look forward to your...

wuzhiyang2016

Does Jacinto7 family supports Tflite, caffee?

4

TIDL documentation is bit confusing . I'm trying to do a custom object detection and classification, could you advise me 1) should I use pytorch? 2)if tf lite supports, could...

ghost