Open FSet89 opened 3 years ago
The variable scores in the predict function is never used and class scores are computed based on confs. Is that crrect?
def predict(imgs, default_boxes): confs, locs = ssd(imgs) confs = tf.squeeze(confs, 0) locs = tf.squeeze(locs, 0) confs = tf.math.softmax(confs, axis=-1) classes = tf.math.argmax(confs, axis=-1) scores = tf.math.reduce_max(confs, axis=-1) # never used boxes = decode(default_boxes, locs) out_boxes = [] out_labels = [] out_scores = [] for c in range(1, NUM_CLASSES): cls_scores = confs[:, c] # correct?
The variable scores in the predict function is never used and class scores are computed based on confs. Is that crrect?