FIFOQueue '_2_batch_join/fifo_queue ' is closed and has insufficient elements (requested 90, current size 0) [[Node: batch_join = QueueDequeueUpToV2[component_types=[DT_FLOAT, DT _INT64], timeout_ms=-1, _device="/job:localhost/replica:0/task:0/cpu:0"](batch_join/fifo_queue, _arg_batch_size_0_0)]]

[superYong2020]

Hello, guys! Recently I got some issues while running facenet code. After nearly ten epochs training, it will show the following mistakes. I have tried the solutions such as change decode_images(...,channels = 3) or set smaller max_nrof_epochs. I am really confused for that. I am using a subset of Ms_celebA_1M dataset, about 1.3 million images. I'd really appreciated for any of your suggestion.Thanks a lot!

This is my parameters: python src/train_softmax.py --logs_base_dir ~/logs/facenet/ --models_base_dir ~/models/facenet/ --data_dir /home/facenet-master/data/MS_Celeb_1M_clean_aligned --image_size 160 --model_def models.inception_resnet_v1
--optimizer RMSPROP --learning_rate -1 --max_nrof_epochs 100 --keep_probability 0.8 --random_crop --random_flip --learning_rate_schedule_file data/learning_rate_schedule_classifier_casia.txt --weight_decay 5e-5 --center_loss_factor 1e-2 --center_loss_alfa 0.9 --lfw_dir /home/facenet-master/data/lfw_aligned

[turboLIU]

I met the same problem!!!!!!!!someone help

[406747925]

+1

[tjpulfn]

I met the problem again , did you solved????