梦里茶

Results 57 comments of 梦里茶

@bqxu1024 mix.list与论文中的实验无关,只是一个额外的测试。

@rintiunse 肯定是图像模型训练出问题了,需要提供更多信息才能判断具体原因。

@rintiunse 所有的纯图像结果都在renew_pid.log中,所有融合结果都在cross_filter_pid.log中,根据你之前的日志,你的纯图像直接迁移(market_grid-cv0-test/renew_pid.log)就已经出问题了,所以问题肯定是在图像模型训练上。

In real scene application, the number of people in target scene is unlimited, so we only care the number of the data we collect, which is easy to get by...

Can you provide the toolkit names about converting pytorch model to trt? I guess you convert torch to onnx with pytorch basic api and load onnx model in tensorRT?

In net2, we need top1 loss for label_local, in net3, we don't need the losses. If that's right, there exists a conflict when training the align net containing net2 and...

@tzengwei 暂时是酱:smile:

@chenjun0210 如果使用pycharm,需要把src目录设置为source directory,才能找到not_mnist包下的代码

@chenjun0210 需要先安装tensorflow,否则直接import tensorflow会出错的, 因为直接运行python seq2seq.py的时候,源代码目录就相当于是在rnn这个目录下了 可以试一下cd 到src目录下,然后 ``` /home/cwh/.conda/envs/tensorflow/bin/python convnet/conv_mnist.py ``` 其中 ``` /home/cwh/.conda/envs/tensorflow/bin/python ``` 是安装了tensorflow的python的路径