phil
phil
when i try to train the faster rcnn model with mobilenetv1 as it's backbone, i found the rpn can find anchor_target correctly，i saved the anchor target imgs， but the rcnn...
@auroua hi： the tricks as follows： in eval_data_reader.py “if flip == 1: img = np.fliplr(img) ” and in verification.py "embeddings = embeddings_list[0] + embeddings_list[1] embeddings = sklearn.preprocessing.normalize(embeddings) " so ,...
@wy1iu hi， 1 have two problems： 1）when i train the model with “type: SINGLE”，the loss can not converge，it is always 9.xx 2）then i use the model trained by 1），to train...
at the beginning，i use m as “type：SINGLE”，and the loss converge，acc can is about 0.98. but when i change the m as “type：QUADRUPLE”，the loss become larger and larger，and acc come down...
If i had a image，but i don't know the noise type of it, can it be denoised? and which model should i use
When using a generator for validation data, you must specify a value for `validation_steps` I trained the model successfully before，but i when i train it again， i met this problem，how...
sorry i make a mistake
@xiaolonw if(tloss1 > 0) { for(int k = 0; k < dim; k ++) { fori_diff[k] += (fneg[k] - fpos[k]); // / (pairNum * 1.0 - 2.0); fpos_diff[k] += -fori[k];...