theikkila

5 ms·

1.
by theikkila 3y ago
Does WarpStream guarantee correct order inside partition only for acknowledged messages or also among the acknowledged messages (in different batches)? If so how do you keep clocks synchronized between the agents?
2.
by theikkila 3y ago
Related to 1. If I understood corrently the agent generates single object per each flushing interval containing all data accross all topics it has received. Does this mean that when reading the consumer needs to read multiple partition data
3.
by theikkila 4y ago
I have somehow feeling that this does not answer the original question. Could you give some examples of impossible or not easy tasks this achieves and other solutions don’t?
4.
Face recognition on broadcast of Finnish independence day(blog.emblica.fi)
1 points by theikkila 4y ago|0 comments
5.
by theikkila 6y ago
wasn't that 4k requests per minute?
6.
by theikkila 6y ago
I'm little bit troubled by the model created for classifying the vein patterns. With only corpus of size 40 and using two classes without augmentation will most likely end up overfitting the model. I would say the model is currently le
7.
Design of Roller Coasters [pdf](aaltodoc.aalto.fi)
2 points by theikkila 7y ago|0 comments
8.
Official CNCF/Kubernetes meetup Helsinki – livestream(youtube.com)
1 points by theikkila 8y ago|0 comments
9.
by theikkila 9y ago
Not at all, deep nets are difficult to train and they need lot's of processing before they learn same kind of classifying features than ie. SVM has. So yes you can simulate SVM with deep networks but usually it's not very good sol
10.
by theikkila 9y ago
Hmm ok, yeah apparently you can leverage their API's and teach new labels for your own data. https://azure.microsoft.com/en-us/services/cognitive-service... That is great and will definetly help with problems
11.
by theikkila 9y ago
That sounds interesting, it's funny if you really can use their own models as base and do that. For the platform sake, Google offers also SaaS where you can train and evaluate your own models but then the base model is something you ha
12.
by theikkila 9y ago
I wanted to try out different types of final layers (SVM, Logistic Reg.) so the direct example didn't suit very well for that kind of testing out. For starters that good though.
13.
by theikkila 9y ago
Hello author of the article here! I will probably publish in some time (when I'm not so busy doing other projects) some code samples too but until that I can give you tips. You should start inspecting how the pre trained models (Incept
14.
Using Tensorflow to Recognize Saimaa Ringed Seals from Webcam(blog.emblica.fi)
137 points by theikkila 9y ago|20 comments
15.
by theikkila 9y ago
I find this interesting to use even with common web applications as a design pattern. It's usually easy to just dispatch events but parsing them is always complicated and if you need to do that in all clients youre essentially decoupli
16.
Edge vs. level triggered logic in Kubernetes(speakerdeck.com)
4 points by theikkila 9y ago|1 comments
17.
Digging into the TurboFan JIT (2015)(v8project.blogspot.com)
5 points by theikkila 9y ago|0 comments
18.
Machine Learning in Production(github.com)
3 points by theikkila 9y ago|0 comments
19.
Clojure editor, Cursive 1.5.0, released(groups.google.com)
2 points by theikkila 9y ago|0 comments
20.
by theikkila 11y ago
Crashed my chrome :D
21.
Show HN: 7000 skill-keywords scraped from Linkedin(gist.github.com)
5 points by theikkila 12y ago|2 comments