TildAlice – Tech Blog for Developers
Python, AI, Deep Learning, Data Analysis, Computer Vision, Reinforcement Learning โ tips and tutorials
Latest Posts
-
BFS/DFS ์์ ์ ๋ณต – ๊ทธ๋ํ ํ์ ์๊ณ ๋ฆฌ์ฆ ๊ตฌํ๋ถํฐ ์ค์ ์์ฉ๊น์ง
BFS vs DFS: when to use each, how they differ in time complexity, and 5 graph problems you can't solve without mastering both algorithms.
-
๊ทธ๋ฆฌ๋ ์๊ณ ๋ฆฌ์ฆ ์๋ฒฝ ์ ๋ฆฌ – ํ๋ ์ ํ, ํ์์ค ๋ฐฐ์ ๋ฑ ์ฝ๋ฉํ ์คํธ ํ์ ์ ํ ํ์ด
Greedy algorithms decoded: solve activity selection, meeting rooms, and coin change in O(n log n). Spot the 3 cases where greedy fails.
-
SQL๊ณผ Python ์ฐ๋ ์๋ฒฝ ๊ฐ์ด๋: SQLAlchemy + Pandas ์ค์ ์ํฌํ๋ก์ฐ
Connect SQL to Python with SQLAlchemy + Pandas: extract 1M rows in 30 seconds, automate reports, and avoid the 3 common connection leaks.
-
Policy Gradient ์๋ฒฝ ๊ฐ์ด๋: REINFORCE๋ถํฐ PPO๊น์ง ์ํ ์๋ฆฌ์ PyTorch ๊ตฌํ
Policy Gradient from scratch: REINFORCE to PPO with PyTorch code. Reduce variance 100x using baselines and see why PPO beats vanilla PG.
-
Python ํ์ ํํธ ๊ณ ๊ธ ํ์ฉ: Generic, Protocol, TypeVar๋ก ํ์ ์์ ํ ๋ผ์ด๋ธ๋ฌ๋ฆฌ ์ค๊ณํ๊ธฐ
Build type-safe Python libraries using Generic, Protocol, and TypeVar. Catch bugs at edit-time, not runtimeโhere's the exact pattern.
-
์ฌ๋ผ์ด๋ฉ ์๋์ฐ ํจํด: O(Nยฒ)์์ O(N)์ผ๋ก ์ต์ ํ
Speed up subarray problems from O(nยฒ) to O(n) using sliding window. Fixed vs variable size, two pointers, and 4 LeetCode patterns solved.
-
Python ์ ๋๋ ์ดํฐ์ yield ์์ ์ ๋ณต: ๋ฉ๋ชจ๋ฆฌ ํจ์จ์ ์ธ ๋์ฉ๋ ๋ฐ์ดํฐ ์ฒ๋ฆฌ ํจํด๊ณผ ์ค์ ์์
Python generators cut memory usage by 4 million times. Process giant files, build data pipelines, and handle infinite sequences with yield.
-
Switch Transformers ๋ฆฌ๋ทฐ: 1.6์กฐ ํ๋ผ๋ฏธํฐ MoE ๋ชจ๋ธ
Switch Transformers paper decoded: how Top-1 routing scales MoE to 1.6T parameters, load balancing tricks, and distillation speed hacks.