WIKI

代码无法使用！！！！！

4

各种环境准备好后，执行CUDA_VISIBLE_DEVICES='0' python3 run_pretraining.py --input_file=./data/tf*.tfrecord --eval_batch_size=64 --output_dir=./checkpoint --do_train=True --bert_config_file=./model/albert_config_xlarge.json --train_batch_size=4096 --max_seq_length=512 --max_predictions_per_seq=20 --num_train_steps=125000 --num_warmup_steps=12500 --learning_rate=0.00176 --save_checkpoints_steps=2000 --init_checkpoint=./model/albert_model.ckpt 打印一大堆错误，所有参数应该设置正确！！！ 20-01-15 17:46:55.428436: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7 2020-01-15 17:46:55.436464: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1746]...

I make a better one [https://github.com/WIKI2020/FacePose_pytorch]

No boxes to NMS!!!!!!!!!!!!!!!!!!!

27

The code can not be used at all. It can be trained and tested. However, demo.py prompt:"no boxes to NMS", and eval. Py prompt:"AVG map: 0.0"

what the ***k!

5

my number_class is 3(1bg+2classes)。eval is error and demo.py not work! which para error? Loading checkpoint: ./weights/Final_efficientdet-d4.pth ... Loaded pretrained weights for efficientnet-b4 Traceback (most recent call last): File "eval.py", line...

I make a better one [https://github.com/WIKI2020/FacePose_pytorch]

6

I make a better one [https://github.com/WIKI2020/FacePose_pytorch]

有没有高手，改成用百度查找资料，用中文使用，适合中国使用

15

### Duplicates - [X] I have searched the existing issues ### Summary 💡 _No response_ ### Examples 🌈 _No response_ ### Motivation 🔦 _No response_