TildAlice – Tech Blog for Developers
Python, AI, Deep Learning, Data Analysis, Computer Vision, Reinforcement Learning โ tips and tutorials
Latest Posts
-
RoPE vs ALiBi: 32K Context LLaMA Perplexity Beats MPT
RoPE vs ALiBi performance at 32K context: LLaMA's perplexity wins vs MPT. Position encoding comparison reveals surprising scaling differences.
-
SciPy FFT vs NumPy FFT: 2.3x Speed Gap at 50kHz
NumPy FFT ran 2.3x slower than SciPy on 50kHz bearing data. Here's the benchmark, the gotchas, and a one-line fix for your vibration pipeline.
-
Python list vs tuple vs set: Read/Write Speed Benchmark
Compare Python list, tuple, and set performance with benchmarks revealing read/write speeds and when each data structure shines in real code.
-
Pillow to OpenCV: 3.4x Faster Image Load in Production
Pillow's imread took 800ms per image. OpenCV cut it to 240ms. Here's the color space trap, dtype bug, and EXIF hell that break every migration.
-
Pandas GroupBy 2x Faster: Categorical Dtypes on 1M Rows
Speed up Pandas GroupBy 2x with categorical dtypes on large datasets. Real benchmark on 1M rows shows dramatic performance gains and memory savings.
-
Cloud vs Edge PHM: $47K/Year Gap in 100-Sensor Setup
Compare TCO of cloud vs edge PHM architectures for 100-sensor deployments. $47K annual cost gap revealed through real-world analysis.
-
Seed Phrase Backup: Paper vs Metal vs NFC (Real Tests)
I tested paper wallets, steel plates, titanium tags, and NFC cards for seed phrase backup. Here's what survived fire, water, and everyday use.
-
LangChain vs LlamaIndex: 1M Document Query Speed Test
LangChain hit 47s query latency at 1M documents while LlamaIndex stayed under 100ms. Here's the architectural difference that causes this 500x gap.