zhuantouer

tf.train.import_meta_graph

1

Hi, @dennybritz when I use `tf.train.import_meta_graph` to eval the trained model for new samples, it is very slow to load the graph, do you known why, is there other method...

pip安装失败

7

服务器是ubuntu 20.04, python3.8， python3.7, python2都试过了，报下面的错误： ``` 2021-01-30 17:12:50 (63.8 MB/s) - ‘kcptun-v20210103.tar.gz’ saved [6367092/6367092] + sha1sum -c + echo '31ced1ecd7adefdc62d1f311e88d9d307c229dc8 kcptun-v20210103.tar.gz' kcptun-v20210103.tar.gz: OK + tar -zxf kcptun-v20210103.tar.gz -C /usr/local/kcptun +...

0.14版本下 08.machine_translation infer无法运行

2

``` Traceback (most recent call last): File "infer.py", line 210, in main(use_cuda) File "infer.py", line 205, in main decode_main(False) # Beam Search does not support CUDA File "infer.py", line 147,...

ImportError: cannot import name 'default_gui_banner'

3

env: mac: 10.13.6 python: Python 3.6.0 (v3.6.0:41df79263a11, Dec 22 2016, 17:23:13) all the pip, jupyter, notebook are latest. ``` WARNING:root:kernel ae65b357-ee9b-418e-9e20-951e8f7e5877 restarted Traceback (most recent call last): File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/runpy.py", line...

oc引导windows10蓝屏

5

cpu: i5-10500 gpu: rx580 4G 主板: b460m-mortar 非wifi 网卡: BCM94360/2cs免驱网卡 安装过程：先安装的windows，然后压缩200G的硬盘；pe中重建efi分区，用release的oc efi安装苹果； 将U盘oc cp到硬盘efi分区（此时无法启动windows，用pe修复引导）；修改config，添加win10引导；通过oc选win10，启动后蓝屏，提示system_thread_exception_not_handled； bios里目前两个启动，一个windows启动，一个oc启动，开机直接选择windows启动没问题，但是从oc进入就蓝屏

It will be perfect that adding save and load model using c++

11

Very instructive work

android.app.ServiceConnectionLeaked

1

test sample with android 4.4.2 , this is the log : E/ActivityThread﹕ Service me.tatarka.support.internal.job.JobSchedulerService has leaked ServiceConnection me.tatarka.support.internal.job.JobSchedulerService$JobServiceConnection@41d7c648 that was originally bound here android.app.ServiceConnectionLeaked: Service me.tatarka.support.internal.job.JobSchedulerService has leaked ServiceConnection me.tatarka.support.internal.job.JobSchedulerService$JobServiceConnection@41d7c648...

java.lang.IllegalArgumentException: Service not registered

1

test your sample with android 4.4.2 , this is the log: java.lang.IllegalArgumentException: Service not registered: me.tatarka.support.internal.job.JobSchedulerService$JobServiceConnection@42055c48 at android.app.LoadedApk.forgetServiceDispatcher(LoadedApk.java:961) at android.app.ContextImpl.unbindService(ContextImpl.java:1690) at android.content.ContextWrapper.unbindService(ContextWrapper.java:529) at me.tatarka.support.internal.job.JobSchedulerService.finishJob(JobSchedulerService.java:212) at me.tatarka.support.internal.job.JobSchedulerService.access$000(JobSchedulerService.java:29) at me.tatarka.support.internal.job.JobSchedulerService$JobServiceConnection$1.acknowledgeStopMessage(JobSchedulerService.java:143) at me.tatarka.support.job.JobService$JobHandler.ackStopMessage(JobService.java:158)...

when i test your sample application i found this problem , this is the log: Process: me.tatarka.support.job.sample, PID: 29234 java.lang.RuntimeException: Unable to start service me.tatarka.support.job.sample.service.TestJobService@41e617c8 with null: java.lang.NullPointerException at android.app.ActivityThread.handleServiceArgs(ActivityThread.java:2885)...

do not need to find service when service finded

2

in `org.zpcat.ble.ui.central.CentralDetailsActivity.displayGattServices(): displayGATTServices` : when connected to a GATT server and get the services and characteristics, it is no need to re-find the service again. I test it on nexus...