TildAlice – Tech Blog for Developers
Python, AI, Deep Learning, Data Analysis, Computer Vision, Reinforcement Learning โ tips and tutorials
Latest Posts
-
5 NumPy Tricks That Cut Portfolio Backtest Time by 10x
Cut backtest runtime from 45 minutes to 4 with stride tricks, vectorized position logic, and array broadcasting. Real before/after benchmarks.
-
Why Excel Still Beats Python for 80% of Data Analysis
Excel beats Python for most business data tasks. Here's when to use each โ based on real timing comparisons, cognitive load, and iteration speed.
-
Chocolate Sales: EDA to ML in Two Kaggle Notebooks
A complete walkthrough of the Kaggle Chocolate Sales dataset โ from interactive Plotly EDA to predicting revenue with Ridge, Random Forest, XGBoost, and a Keras neural network.
-
W&B Artifacts and Team Collaboration: Model Version Control
W&B Artifacts version models like Git commits. Real lineage tracking, team workflows, and the one metadata trick that saved a 3AM rollback.
-
W&B Sweeps: Bayesian vs Grid Search Benchmark (ResNet-18)
Bayesian search hit 91% in 28 runs. Grid needed 64. Real benchmarks, ResNet-18, CIFAR-10โsee which strategy wins.
-
ML Experiment Tracking with W&B: Lessons from 50 Failed Runs
W&B cuts experiment chaos in half โ but only if you log the right metrics. Here's what 50 runs taught me.
-
QLoRA Review: 4-Bit Finetuning for 65B Models on One GPU
QLoRA fits a 65B model on one 48GB GPU with zero quality loss. Here's exactly how NF4 quantization, double quant, and LoRA combine โ with code.
-
Real-Time License Plate Reader: YOLOv8 + PaddleOCR
YOLOv8 + PaddleOCR pipeline runs at 21 FPS with 94% accuracy. Here's the preprocessing trick that fixed OCR failures.