Samri’s Log
Kanalga Telegram’da o‘tish
Hey, I’m Samri 🤗. I’m a junior developer hooked on AI and ML. This channel’s my logbook , where I drop resources, share little finds, jot down thoughts, and track my progress along the way.
Ko'proq ko'rsatishMamlakat belgilanmaganToif belgilanmagan
253
Obunachilar
Ma'lumot yo'q24 soatlar
+17 kunlar
+130 kunlar
Postlar arxiv
253
Repost from DOT_RUTH
🚀 Founder Academy Applications Now Open!
Are you ready to take your business idea to the next level?
📅 Application Deadline: October 8
🗓 Programme Start Date: October 13
This 4-week intensive programme is designed to help founders sharpen their vision, build strong business foundations, and connect with a community of innovators.
👉 Apply today using the pinned steps in the channel and start building the future: https://shorturl.at/6EBf3
#ALXVentures #ALXAfrica #LifeAtALX #ALXEthiopia #DoHardThings
253
What's your 2017 takeaways? drop it here anonymously or share in the comment.
Let's learn from each other's 1 year journey
253
2017 Takeaway
1. When you set a goal or anything focus on the input. If it’s not working, just change the input. You won't be discouraged. There are many ways to get something just be flexible 😊.
2. Morning hours especially from 10am onward are the most productive hours. Be greedy with that time , use it for hard or challenging tasks 🙃
3. Some people’s sh*t is their problem. I repeat it’s not about you 🤷♀
4. Inner peace is priceless. If where you are or something you hold on to, makes you lose your peace, let it go. Just do it for the sake of your peace 🤗. Your happiness, focus and productivity is all sit on top of inner peace ✌️
5. Avoid social media right after you wake up. Else you will just get cheap dopamine and waste your fresh mind 😌.
@samli_documentation
253
Overfitting is not a problem nowadays, coz we have techniques like early stopping, learning rate scheduling with exponential decay, gradient descent optimization, dropout.....
When we comes to underfitting I think it's more of data and skill issues 🤷♀
I remember how I was struggling with my first deep learning task the model didn’t even beat the baseline accuracy.😓
I don't know where I get this phrase but
Garbage in garbage outDirty data ~ bad model Know your model well like very well(plus if it is good fit for the objective or whatever) and clean ur data 🧹 @samli_documentation
253
https://competitions.langflow.org/
If you are up to agentic workflows , check it out.
@samli_documentation
Endi mavjud! Telegram Tadqiqoti 2025 — yilning asosiy insaytlari 
