Why my loss have no changes?

[pixixiaonaogou]

I have train the pretrained_model on my own datasets by using this command

DATASET_DIR=./tfrecords TRAIN_DIR=./log/ CHECKPOINT_PATH=./checkpoints/vgg_16.ckpt python train_ssd_network.py \ --train_dir=${TRAIN_DIR} \ --dataset_dir=${DATASET_DIR} \ --dataset_name=pascalvoc_2007 \ --dataset_split_name=train \ --model_name=ssd_300_vgg \ --checkpoint_path=${CHECKPOINT_PATH} \ --checkpoint_model_scope=vgg_16 \ --checkpoint_exclude_scopes=ssd_300_vgg/conv6,ssd_300_vgg/conv7,ssd_300_vgg/block8,ssd_300_vgg/block9,ssd_300_vgg/block10,ssd_300_vgg/block11,ssd_300_vgg/block4_box,ssd_300_vgg/block7_box,ssd_300_vgg/block8_box,ssd_300_vgg/block9_box,ssd_300_vgg/block10_box,ssd_300_vgg/block11_box \ --trainable_scopes=ssd_300_vgg/conv6,ssd_300_vgg/conv7,ssd_300_vgg/block8,ssd_300_vgg/block9,ssd_300_vgg/block10,ssd_300_vgg/block11,ssd_300_vgg/block4_box,ssd_300_vgg/block7_box,ssd_300_vgg/block8_box,ssd_300_vgg/block9_box,ssd_300_vgg/block10_box,ssd_300_vgg/block11_box \ --save_summaries_secs=60 \ --save_interval_secs=600 \ --weight_decay=0.0005 \ --optimizer=adam \ --learning_rate=0.001 \ --learning_rate_decay_factor=0.94 \ --batch_size=32 Some problems in this process is solved by me . But the loss have no changes in the process of training So i want to ask some help for this question

[pixixiaonaogou]

And I provide some detail for my datasets NUM_CLASSES = 1 and the loss is 1.3292 all the time .

[jhgfkdj]

Me too, and my total_loss always wave at 100 up and down and never devline , have you solved this problem?

[yyc9268]

I guess the NUM_CLASSES should be 2 because it includes background class. You can confirm that default NUM_CLASSES is set to 21 although PASCAL VOC includes only 20 classes.