yoseph.won

i swear i am getting confused a lot... is the music in the background is it with me weys hulum ga new ?

wait is this a rap battle weys Box... 😄

after that i’ve been digging through my code and writing better tests and now y’all got me hooked on these ETFC thing... and my brain is saying "ለሊቱ የኛ ነው give me some እረፍት"... so time to join the fun 🥊💪🏻

thought i handled every edge cases.... then one edge case i completely forgot abt came to my mind and tested my code with that… and my code still ran like everything was fine 💀 silent failures are scary.... ✍️ don’t confuse “no error” with “correct behavior.” #silent_failure @yosephwon

Deadline😭

they are hyping it too much🙄

The only way out is through.... @MrBereketK

5hr is actually insane 😭 #dhh #lex #podcast

DAMN ma bb😭 all cores are running at 100%... performance still unavailable... same as my brain😭

remember when i used to yap about regex? back then, i only saw how powerful they were for validation... but now they’ve saved me a ton of work once again turns out they’re also a damn good way to structure and chunk documents. give it a shot next time. #regex #chunking #rag @yosephwon

And in general this is the goto book for RL might feel outdated if you want fake RL/LLM RL, the ones everyone was doing the past couple of years for LLMs but now real RL is being used a lot. pdf: https://web.stanford.edu/class/psych209/Readings/SuttonBartoIPRLBook2ndEd.pdf

We're releasing the first batch of Dataset.ET open speech data for Amharic. 22.7 hours. 7,405 recordings. 320 speakers. Free, CC BY 4.0, on Hugging Face. Amharic has close to no open speech data. That's the gap we're trying to close, and this is the first step rather than the finished thing. Some honesty about what this is: it's a first batch. The audio isn't preprocessed yet. We're not going to tell you it's the best quality out there. We don't think it is. What we want is for people to actually use it and tell us where it falls short. The next release will be different: properly preprocessed audio, a real data pipeline with testing and validation built in. This exists because 320 people in Ethiopia gave their voices to it, and because contributors reviewed each other's recordings. Thank you to all of them. If you're working on Ethiopian languages, low-resource ASR, or you just want to talk about the data, DM us. We'd genuinely like to hear from you. 🔗 https://huggingface.co/datasets/snapwre/amharic-speech

uploading btw😎😋

W for my brother! Couldn’t be prouder! 🔥💪🏻 #G12 #National_Exam_Result

take care of yourselves you guys. like be careful with which roads you take, and what time you’re out, and make sure you’re not alone and also with not jema... ena don’t stay out too long kemeshe behuala just head home Stay safe fr. #staysafe

it’s not funny anymore yewnet like i’ve been joking abt ts for the past few days and people are making jokes abt it too... gn it’s actually a serious thing. this has to stop somehow... I freaked the hell out today and now i keep thinking abt what those dudes are thinking and what their families are gonna say. it's scary ena betam debari neger someone seriously needs to do something about this.

just saw two dudes መታፈሲንግ🏃 #run_for_ur_life

i really need to give this a try... what a relief especially for docker networking setup😮‍💨

W #fastapi