标签：#alpr

Cancellations have accelerated.

The last post was about making a number trustworthy. Leakage geometry, purge widths, de-overlap, a baseline that could not cheat. It ended with a minute-scale ceiling that held at 52% across seven configurations and a model family swap. This one is about what happens after you trust the number. Because a probability you are going to bet on is a different object from a probability you are going to report. The probabilities are not decorative The path-passage classifier is a three-class LightGBM. It returns p_up , p_down , p_none . Those go straight into the expected-value score that decides whether to take a trade and how big: long_score = p_up * ( B - C ) + p_down * ( - B - C ) + p_none * ( - C ) short_score = p_up * ( - B - C ) + p_down * ( B - C ) + p_none * ( - C ) B is the barrier, C the cost. Read the arithmetic. Every term is linear in a probability. Scale p_up by 1.2 and you scale the long score by very nearly 1.2. So miscalibration does not stay in the model. It becomes a bet-sizing error, in proportion, in the bins where the gate actually fires. A classifier that is right 70% of the time while claiming 90% is not 20 points wrong. It is sizing every position in that bin as though the edge were far larger than it is. Boosted trees are known for uncalibrated softmax output. I had been consuming it as if it were a probability. The audit Seven live assets. For each one, fit an Inductive Venn-Abers wrapper on the time-ordered older 80% of that model's training data, 6,988 rows, and evaluate against a 500-row uniform-random sample of the newer 20%, seed 42. The LightGBM models are reloaded from disk and left alone. Only the wrapper is fit. Measure Expected Calibration Error and log-loss, before and after. Asset ECE before → after ECE Δ Log-loss Δ BTC 0.1272 → 0.0621 -51.2% -5.5% ETH 0.1795 → 0.0298 -83.4% -11.5% SOL 0.1680 → 0.0386 -77.0% -10.6% XRP 0.2219 → 0.0645 -70.9% -17.7% ADA 0.1419 → 0.0369 -74.0% -8.0% LINK 0.1260 → 0.0737 -41.5% -1.2% LTC 0.1508 → 0.0603

Flock to lock out cops who stalk exes, but agencies can still hide abuse.

I once spent nearly a week trying to fix a web scraper that, on paper, had absolutely no reason to fail. The target website wasn't using aggressive, visible defense walls. My script spaced out requests naturally, rotated common user agents, and used browser automation configured to mimic human interactions down to mouse movements. Yet, the results were an absolute nightmare. Some batches of requests would go through cleanly, while others immediately triggered CAPTCHAs or returned 403 Forbidden errors. Every single time I thought I had patched the logic, the failure rate climbed right back up. Like most developers, my default instinct was to assume the application layer was broken. I went down a rabbit hole optimization sprint checking request headers, browser fingerprints, cookies, and session persistence. Nothing explained the wild inconsistency until I noticed a strange clue: some proxy pools performed beautifully, while others crashed on the exact same codebase. The code wasn’t the issue. The culprit was a fundamental misunderstanding of proxy network architecture. Looking Beyond the IP Address: Enter the ASN For a long time, I treated proxies as interchangeable commodities. An IP address was just an IP address, and if one got blocked, you simply rotated to the next. Modern anti-bot solutions like Cloudflare, Akamai, and PerimeterX don't look at IPs in a vacuum. They analyze network layer characteristics, specifically the ASN (Autonomous System Number). An ASN is a unique identifier assigned to a network operator that defines who owns and routes an IP range. When your scraper hits a website, the target's security system looks up your ASN to check your network identity. If your traffic originates from a commercial hosting provider or data center ASN, it carries an automatic penalty score for sensitive endpoints. To build reliable systems, you have to move past basic rotation and understand the two core proxy frameworks that mask this identity: ISP Proxies and Resi