András Gyácsok's blog

 Recently I found NATS server which can be used as message bus in service choreography architecture implementations. Python client can be found https://github.com/nats-io/nats.py

 The paper  describes impressive result for dealing with inpatient temporal data.

I would like to recommend this very good video on Multipass. I lot of topics are covered in it, e.g ssh between host and vm, installation of GUI on vm and more.

  Intro There 2 options in distributed training - the first one distributed data training and the second one - distributed model training. The distributed model training is used when model won't fit into one GPU (e.g. model requires to much memory) this way we split models into multiple GPUs. The distributed data training is used when we have large amount of data and training on one GPU could be too long. For our experiment we would like to use  LUNA dataset  for langue nodule analysis. We plan to use also already written code for classification model training. The code can be found in  this repository . The code is not prepared for distributed model training on multiple nodes. Task To run our training on HPC, as well as other tasks, we have to use  Slurm . Slurm manages cluster and tasks queue. To interact with Slurm we use file with special syntaxis. In our experiment we will acquire 4 nodes with GPU and will look for the way how to utiles correctly this 4 nod...

Very interesting video about AI models training and inference profiling

 Нещодавно перший раз в житті склав повністю кубіка Рубіка, а допоміг мені в цьому допис  Як скласти кубик Рубіка 3х3 інструкція для новачків

 Казка про Снігову королеву передає нам цікаву психологічну модель нарцистичних відносин. Нещодавно зустрів чудову статтю на цю тему. Статтю можна знайти за цим посиланням .

Here is a list of articles about big leap of Alexa into a hospital room: https://www.fastcompany.com/90689082/amazon-alexa-echo-in-hospitals https://www.mobihealthnews.com/news/amazon-vocera-team-new-alexa-skill-patients-hospitals https://www.businessinsider.com/amazon-alexa-hospitals-echo-next-to-us-hospital-bed-2021-10 Interesting competitor: https://www.k4connect.com/

 The title of the post is the same as a title of the same article  Performance Evaluation of Offline Speech Recognition on Edge Devices  published by Santosh Gondi and Vineel Pratap. The guys tested to ASR models on RSP 4 and Jetson Nano. In the article they provide super interesting results for those who is interested in AI on EDGE devices, especially speech to text. The main sense of the test is to try transformers deployment on EDGE. They use PyTorch models one from Huggingface (Word2Vec) and other from fairseq  (speech2text), and TorchScript to quantize the models.

 Very interesting article about detecting human speech in hive recordings can be found under the link . In general article explains usage of Siamese networks  + KNN for audio classification .