-
Hello Andrea,
really nice project! I searched quite a long time to find a programming solution which avoids the (driver-) buggy USBi Interface (e.g. USB 3).
I'm really interested in programming my…
-
There are quite a few tools Shuffle can't natively connect to receive and control information. These are things like:
- RabbitMQ
- Syslog
- Kafka
- Random monitors
The thought is to use somethi…
-
#### Describe the problem
The logstash pipeline for sysmon (pipeline/1531-winevent-sysmon-filter.conf) changes the name of the "Image" field to "process_path". As the sigma rules are looking for a fi…
-
I run a Doorbird camera since 2 years without problems via Scrypted and HKSV, but since 2 weeks the camera stops responding in HK after a few hours. It can be resolved through reloading the plugin in …
-
I have added a custom rule to helk-elastalert that tests blacklist hashes.
This rule has been tested against sample data using elastalert-test-rule and returns the sample hit as seen below in the fin…
-
Hello,
my configuration parameters are as follows
{
"train_config": {
"output_directory": "checkpoints",
"epochs": 100000,
"learning_rate": 1e-5,
"sigma":…
-
James Fehlinger has reported a couple of problems attempting to run this package on simh-V3 and open-simh. Here is what he said, beginning with comments concerning V3.
`So the first clue I had tha…
-
I trained waveglow from scratch without loading pre-trained weights for 100K steps.The audio quality is not getting better and it has some noise with it.
The loss keeps oscillating between 5 and 6
…
-
I am training a Waveglow model from scratch with 8k sampling rate dataset and got this error after 1 epoch
![ảnh](https://user-images.githubusercontent.com/43977734/83830927-84599480-a710-11ea-97f9-8…
-
Hi,
I tried to train my own model with different 'n_mel_channels', i.e. 160 instead of 80.
Since my autoencoder's output dimension is 160.
The other settings are the same as the original setting.
…