speeding motor

> HI, when I read you code , it make me refused here, in the the LIstNet loss function: > > ``` > def get_loss(self, x_t, y_t): > # ----...

have you slove it now, I can not get it too

why need to make a distinction between coord_mask, conf_mask, and class_mask , use the y_true[..., 4] directly is work well ,right? hope to receive you reply, thanks in advance

the code : ``` class_wt = tf.ones(num_classes) class_mask = y_true[..., 4] * tf.gather(self.class_wt, true_box_class) ``` is total equals to : `class_mask = y_true[..., 4]` so why still have to do...

> why need to make a distinction between coord_mask, conf_mask, and class_mask , use the y_true[..., 4] directly is work well ,right? > > hope to receive you reply, thanks...

`maybe use tf.exp(tf.sofrmax(pred[..., 2:4])) instead of tf.exp(pred[..., 2:4]) ?` because tf.exp(pred[..., 2:4]) is always NAN in training

> > I m very confused with this, I still read know the loss_code but still don't know why, > > what directly different with this, > > > #####...

> This section: > > ### adjust confidence > ``` > true_wh_half = true_box_wh / 2. > true_mins = true_box_xy - true_wh_half > true_maxes = true_box_xy + true_wh_half > >...

how to understand the conf_mak here, `loss_conf = tf.reduce_sum(tf.square(true_box_conf - pred_box_conf) * conf_mask)` can you please help me ...thank you very much

I have a little confuse with the loss function: why not only use detector_mask to judge the box have box or not, why still need to get the object_detections, any...