---
title: "Machine Learning"
description: "622 posts about Machine Learning, summarised, each linking to the original."
---

# Machine Learning
> 622 posts about Machine Learning, summarised, each linking to the original.

## Articles

### [Unlocking Real-time Predictions with Shopify's Machine Learning Platform](https://yomu.fyi/post/unlocking-real-time-predictions-with-shopify-s-machine-learning-platfo.md)
- Company: [Shopify](https://yomu.fyi/company/shopify.md)
- Author: 2023-10-18
- Published: Mar 10, 2023
- Type: Explainer

Shopify’s Merlin machine learning platform was expanded to support online inference, allowing trained models to return predictions as input arrives rather than through periodic batch jobs. The design targets varied Shopify use cases and requirements, including support for TensorFlow, PyTorch and XGBoost, low latency, autoscaling, rolling deployments, observability and model updates. Each use case runs as a dedicated service on Google Kubernetes Engine, with a Python serving layer that loads models and artifacts, exposes an inference endpoint, and can use the Comet ML model registry and Pano feature store. Separate staging and production configurations let teams tune resources and test model versions before release; Shopify reports that online inference is providing the intended low latency, scalability and faster iteration, while ensemble inference, platform-based monitoring and continuous training remain planned areas.


### [Storing OpenAI embeddings in Postgres with pgvector](https://yomu.fyi/post/storing-openai-embeddings-in-postgres-with-pgvector.md)
- Company: [Supabase](https://yomu.fyi/company/supabase.md)
- Author: gregnr
- Published: Feb 6, 2023
- Type: Tutorial

Post describes how embeddings represent semantic relatedness as vectors and how OpenAI’s text-embedding-ada-002 model produces 1536-dimensional embeddings for text. It addresses storing and searching large or dynamically changing collections that may not fit comfortably in memory, using PostgreSQL and the pgvector extension instead of a small CSV-and-NumPy workflow. The implementation enables the vector extension, creates a documents table with a vector(1536) column, and defines match\_documents() with cosine distance, thresholds, counts, and ordering. It also shows JavaScript code for generating embeddings, a search flow that retrieves matching documents for prompt context, and token limiting before a completion request, while noting that documents can be chunked. For larger tables, the post recommends an ivfflat index with vector\_cosine\_ops and gives 4 \* sqrt(table\_rows) as a starting point for lists.


### [Monte Carlo Simulations: Separating Signal from Noise in Sampled Success Metrics](https://yomu.fyi/post/monte-carlo-simulations-separating-signal-from-noise-in-sampled-succes.md)
- Company: [Shopify](https://yomu.fyi/company/shopify.md)
- Author: 2023-10-18
- Published: Jan 3, 2023
- Type: Tutorial

Sampled success metrics can be difficult to interpret when the underlying population is too expensive or time-consuming to measure completely. The method presents a Monte Carlo simulation for estimating how sample size and an assumed trend affect the reliability of Shopify’s sampled “shop issue rate,” which estimates requirement violations for an average installed app. Its workflow establishes simulation metrics such as 1mDO, 2mDO, and 1qDO, measures MAPE, defines a sampling distribution, and runs repeated simulations across parameter combinations using Python and pandas. For the example, the issue count is modeled with a Poisson distribution, while sample size, lambda, and monthly decrease are varied to quantify uncertainty and collection cost. With 100 audits per month, an initial mean of 10 issues, and a 5-percent monthly decrease, the simulations predict monthly decreases 83 percent of the time and decreases over two months 97 percent of the time.


### [How KartaCam powers GrabMaps](https://yomu.fyi/post/how-kartacam-powers-grabmaps.md)
- Company: [Grab](https://yomu.fyi/company/grab.md)
- Author: Shuangquan Hou
- Published: Dec 1, 2022
- Type: Problem & solution

Grab's Geo team required an efficient, low-cost way to collect fresh street-level imagery across Southeast Asia, where professional mapping equipment is prohibitively expensive and smartphone crowdsourcing yields inconsistent quality. To address this, Grab built KartaCam, a custom mapping device equipped with a 12MP sensor, dual-band GNSS, 4G LTE, and onboard edge AI. Edge machine learning models evaluate scene suitability, check image quality, filter for map-relevant objects, and blur personal data directly on the device prior to upload. Deploying a four-camera KartaCam 360 array delivers panoramic coverage and point-of-interest data comparable to commercial mapping rigs at roughly one-twentieth the hardware cost.


### [Graph for fraud detection](https://yomu.fyi/post/graph-for-fraud-detection.md)
- Company: [Grab](https://yomu.fyi/company/grab.md)
- Author: Min Chen
- Published: Nov 24, 2022
- Type: Problem & solution

Rapid expansion across multiple business verticals introduced diverse, evolving fraud patterns at Grab, making traditional rule engines and decision trees inadequate due to their heavy reliance on extensive feature engineering and labeled data. To address cold starts and detect unknown fraud, the engineering team implemented a semi-supervised Relational Graph Convolutional Network (RGCN) across millions of connected ecosystem entities. The model propagates information along neighborhood nodes through graph convolutional layers to produce node embeddings and output fraud probabilities. Trained on graphs where only a small percentage of nodes are labeled, the RGCN achieved an AUROC close to 1 and distinctly separated fraudulent embeddings from genuine ones. The approach also offers high explainability by surfacing dense clusters of shared physical devices and addresses ongoing challenges in real-time prediction and noisy connections.


### [How to Structure Your Data Team for Maximum Influence](https://yomu.fyi/post/how-to-structure-your-data-team-for-maximum-influence.md)
- Company: [Shopify](https://yomu.fyi/company/shopify.md)
- Author: 2023-10-18
- Published: Oct 20, 2022
- Type: Problem & solution

Embedded data science leaders may support multiple product and business areas, creating competing work streams, specializations, and stakeholder relationships. The post compares swim lanes and stochastic process assignments against principles of efficiency, influence, stakeholder clarity, stability, growth, and flexibility. It recommends a hybrid “diamond defense” in which scientists are loosely assigned to zones for specialization while a bullpen, including the manager and an additional scientist, supplies reserve capacity. Resources can be redirected to overloaded areas, and team members can rotate into vacancies or cover leave. The framework is presented as combining clear ownership and subject-matter depth with the flexibility to handle change, though the post warns that excessive rotation can create volatility and turnover.


### [What is a Full Stack Data Scientist?](https://yomu.fyi/post/what-is-a-full-stack-data-scientist.md)
- Company: [Shopify](https://yomu.fyi/company/shopify.md)
- Author: 2023-10-18
- Published: Sep 15, 2022
- Type: Explainer

Shopify defines full stack data science as an ownership-oriented, end-to-end approach to solving business problems across the data science lifecycle. Rather than specializing in one stage, practitioners participate in discovery and analysis, acquisition, data modeling, pipeline development, and production delivery. The role combines business acumen, communication, Python and SQL programming, exploratory data analysis, data engineering, and machine learning, while treating machine learning as a tool rather than a default and starting simple before adding complexity. Shopify describes the model as generalist with T-shaped development: broad capability is paired with deeper expertise in a specific area, enabling individual strengths within teams to address complex challenges.


### [Data-Centric Machine Learning: Building Shopify Inbox’s Message Classification Model](https://yomu.fyi/post/data-centric-machine-learning-building-shopify-inbox-s-message-classif.md)
- Company: [Shopify](https://yomu.fyi/company/shopify.md)
- Author: 2023-10-18
- Published: Jul 21, 2022
- Type: Problem & solution

Shopify Inbox needed to classify millions of monthly customer messages so merchants could prioritize responses and guide product decisions. The team adopted a data-centric approach, improving the dataset rather than primarily changing model architectures. An initial prototype used Universal Sentence Encoder embeddings, UMAP and HDBScan clustering, semi-supervised labeling, and a neural classifier, but reached about 70% accuracy with 35% high-confidence coverage. For version 2.0, the team created a more explicit taxonomy, used human-labeled data and fine-tuned DistilBERT, while addressing privacy and class imbalance through data handling and targeted annotation. In production, version 2.0 achieved about 90% accuracy and 80% high-confidence coverage, although the taxonomy and training data require ongoing updates as conversation trends change.


### [Policy Gradient with PyTorch](https://yomu.fyi/post/policy-gradient-with-pytorch.md)
- Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md)
- Author: Thomas Simonini
- Published: Jun 30, 2022
- Type: Tutorial

Policy gradient algorithms optimize reinforcement learning policies directly without learning intermediate action-value functions. Instead of assigning discrete Q-values to actions, these methods parameterize a stochastic policy and apply gradient ascent on an objective score function. This setup eliminates manual exploration tuning, resolves perceptual aliasing in identical states, and naturally accommodates continuous or high-dimensional action spaces. The Reinforce Monte Carlo policy gradient algorithm operationalizes this approach by collecting full episodic trajectories and adjusting parameters along the gradient of the log-action probabilities scaled by return. Practitioners implement this algorithm using PyTorch to evaluate agent robustness across environments like CartPole-v1, PixelCopter, and Pong.


### [Liftoff! How to get started with your first ML project 🚀](https://yomu.fyi/post/liftoff-how-to-get-started-with-your-first-ml-project.md)
- Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md)
- Author: Nima Boscarino
- Published: Jun 29, 2022
- Type: Tutorial

Beginners in machine learning frequently face difficulties when selecting a framework and defining the scope of their initial hands-on project. Sentence Transformers provides an accessible starting point by computing dense vector representations for sentences, paragraphs, and images, enabling semantic search through cosine similarity. A structured four-step strategy guides practitioners through listing library capabilities, identifying compelling datasets, selecting a familiar secondary tool, and brainstorming project concepts. Applying this framework, a song lyrics dataset was paired with Gradio Blocks to build a prompt-based playlist generator. Developing this application required evaluating pre-trained models, hosting generated embeddings on Hugging Face Spaces, and utilizing multi-processor support to accelerate embedding generation.


### [Accelerate Large Model Training using DeepSpeed](https://yomu.fyi/post/accelerate-large-model-training-using-deepspeed.md)
- Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md)
- Author: Sourab Mangrulkar, Sylvain Gugger
- Published: Jun 28, 2022
- Type: Tutorial

Training large models on hardware with limited GPU memory frequently causes out-of-memory errors when using standard Distributed Data Parallel. To solve this bottleneck, Hugging Face Accelerate integrates DeepSpeed ZeRO data parallelism to shard optimizer states, gradients, and model parameters across workers. For a 900-million-parameter DeBERTa model, ZeRO Stage 2 increased the maximum per-device batch size from eight to forty while achieving a 3.5-fold training speedup over DDP without degrading accuracy or F1 score. Advanced configurations allow sequence-to-sequence chatbot finetuning and full ZeRO Stage 3 CPU offloading for models like the 1.5-billion-parameter GPT-XL, enabling batch size 16 training where DDP fails. Accelerate enables these memory optimizations through simple configuration files and the accelerate launch command with minimal or zero code modifications.


### [Announcing Evaluation on the Hub](https://yomu.fyi/post/announcing-evaluation-on-the-hub.md)
- Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md)
- Author: Lewis Tunstall, Abhishek, Tristan Thrush, Sasha Luccioni, Leandro von Werra, Nazneen Rajani, Aleksandra Piktus, Omar Sanseviero, Douwe Kiela
- Published: Jun 28, 2022
- Type: Announcement

Hugging Face introduced Evaluation on the Hub, a no-code tool designed to streamline the evaluation of machine learning models across diverse datasets. Traditional evaluation approaches often suffer from reproducibility issues, implementation inconsistencies, and cumbersome workflows across varied metrics and datasets. Powered by AutoTrain and Hugging Face Spaces, the service allows users to configure tasks, map dataset columns, select evaluation metrics, and run tests directly from dataset pages. Completed evaluations automatically open pull requests on the respective model cards to encode standardized verification metadata and update public dataset leaderboards. This system establishes consistent benchmarking pipelines across tasks like image classification, text summarization, and named entity recognition without requiring local code execution.


### [Getting Started With Embeddings](https://yomu.fyi/post/getting-started-with-embeddings.md)
- Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md)
- Author: Omar Espejel
- Published: Jun 23, 2022
- Type: Tutorial

Embeddings represent unstructured information such as text and images as numerical vectors in a shared semantic space. To demonstrate their utility, a simple semantic search engine is built over US Social Security Medicare frequently asked questions. The system generates 384-dimensional vector representations for thirteen FAQ entries by dispatching POST requests to the Hugging Face Inference API using the sentence-transformers/all-MiniLM-L6-v2 model. Incoming user queries are converted into matching vector representations and evaluated against stored dataset vectors using the util.semantic\_search function from the Sentence Transformers library. By calculating cosine similarity scores, the system retrieves and ranks the five most semantically relevant questions without requiring custom keyword rules or massive labeled training sets.


### [Convert Transformers to ONNX with Hugging Face Optimum](https://yomu.fyi/post/convert-transformers-to-onnx-with-hugging-face-optimum.md)
- Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md)
- Author: Philipp Schmid
- Published: Jun 22, 2022
- Type: Tutorial

Exporting Hugging Face Transformers models to the ONNX format can be accomplished through three different abstraction levels. The low-level approach utilizes torch.onnx.export, requiring manual specification of dummy inputs, input names, output names, opset versions, and dynamic axes configurations. At the intermediate level, the transformers.onnx package simplifies the conversion process by relying on FeaturesManager and prebuilt configuration objects to handle dynamic axis definitions automatically. The high-level method uses Hugging Face Optimum classes such as ORTModelForSequenceClassification by setting the from\_transformers flag to True inside from\_pretrained. This Optimum export leverages transformers.onnx internally and produces a model ready for immediate inference execution or integration into pipelines.


### [Intel and Hugging Face Partner to Democratize Machine Learning Hardware Acceleration](https://yomu.fyi/post/intel-and-hugging-face-partner-to-democratize-machine-learning-hardwar.md)
- Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md)
- Author: Julien Simon
- Published: Jun 15, 2022
- Type: Announcement

Intel has joined Hugging Face's Hardware Partner Program to accelerate Transformer training, fine-tuning, and inference on Intel platforms. Large Transformer models introduce latency bottlenecks in production workloads like search and chatbots, where hardware-level optimization typically requires tedious trial and error. To streamline model optimization, the collaboration introduces Optimum Intel, an open-source library integrating the Intel Neural Compressor for automated quantization, pruning, and distillation. A demonstration applies post-training dynamic quantization to a fine-tuned DistilBERT classification model using a CPU-only PyTorch setup. The quantized model converted 38 Linear and 2 Embedding operators to 8-bit integers, reducing evaluation duration by 1.34x while keeping the accuracy drop within a 5% threshold.


### [Director of Machine Learning Insights \[Part 3: Finance Edition\]](https://yomu.fyi/post/director-of-machine-learning-insights-part-3-finance-edition.md)
- Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md)
- Author: Britney Muller
- Published: Jun 14, 2022
- Type: Explainer

Machine learning leaders in financial institutions face significant operational hurdles when integrating automated models into production environments. Practitioners must navigate complex legacy architectures, strict regulatory oversight, and privacy mandates tied to personally identifiable financial records. Integrating models often fails when organizations lack clear communication buffers, deploy models using improper prediction windows, or treat systems as opaque black boxes without verifying underlying mechanics. Despite these challenges, financial applications increasingly rely on machine learning for anti-money laundering compliance, automated fraud screening, and loan underwriting. Success in this regulated sector demands comprehensive evaluation across representative input spaces, robust ongoing model monitoring, and explainable decision outputs.


### [8 Data Conferences Shopify Data Thinks You Should Attend](https://yomu.fyi/post/8-data-conferences-shopify-data-thinks-you-should-attend.md)
- Company: [Shopify](https://yomu.fyi/company/shopify.md)
- Author: 2023-10-18
- Published: Jun 10, 2022
- Type: Explainer

Shopify’s data scientists and engineers curated eight upcoming data conferences for 2022, covering virtual, hybrid, and in-person formats. The selection supports learning and development by helping practitioners hear from peers about applications, techniques, use cases, and opportunities in the wider data community. Hybrid options include Data + AI Summit, Transform, RecSys, and ODSC West, with topics spanning analytics, machine learning, open source technologies, recommender systems, MLOps, natural language processing, and big data. In-person choices—PyData London, KDD, Re-Work Deep Learning Summit, and Crunch—feature tutorials, research, business applications, AI challenges, data-team practices, and machine learning at scale, with locations including London, Washington D.C., Toronto, and Budapest.


### [How facial recognition technology keeps you safe](https://yomu.fyi/post/how-facial-recognition-technology-keeps-you-safe.md)
- Company: [Grab](https://yomu.fyi/company/grab.md)
- Author: Kai Feng Tee
- Published: Jun 9, 2022
- Type: Explainer

Grab utilizes facial recognition technology across its platform for driver authentication, passenger verification, and digital electronic Know Your Customer (e-KYC) processes. The core pipeline consists of image preprocessing through face detection and alignment, anti-spoofing checks, feature extraction into high-dimensional vector embeddings, and downstream verification or search. To counter spoof attacks like screen replays, synthetic moiré patterns are generated and cropped face patches are used during training and inference to focus on local structures rather than global semantic noise. Face verification challenges involving shallow ID datasets and masked faces are resolved using semi-Siamese training architectures and masked data augmentations.


### [The Annotated Diffusion Model](https://yomu.fyi/post/the-annotated-diffusion-model.md)
- Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md)
- Author: Niels Rogge, Kashif Rasul
- Published: Jun 7, 2022
- Type: Tutorial

Denoising Diffusion Probabilistic Models generate data samples by learning to reverse a discrete-time forward diffusion process that incrementally adds Gaussian noise to inputs. The forward process adds noise across a predefined schedule of variance parameters up to a fixed number of steps, transforming data into an isotropic Gaussian distribution. Because the true reverse conditional probability distribution is intractable, a neural network optimizes the variational lower bound to approximate the distribution mean at each step while keeping the variance fixed. Sampling subsequently proceeds by iteratively denoising pure Gaussian noise through successive network evaluations to reconstruct a clean data sample. Although diffusion models demonstrate high generation quality, their primary drawback remains the requirement of multiple forward passes during inference compared to models like generative adversarial networks.


### [Deep Q-Learning with Space Invaders](https://yomu.fyi/post/deep-q-learning-with-space-invaders.md)
- Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md)
- Author: Thomas Simonini
- Published: Jun 7, 2022
- Type: Explainer

Tabular Q-learning fails to scale to complex environments like Atari games due to massive observation spaces that make maintaining discrete state-action tables impractical. Deep Q-learning addresses this limitation by using a neural network to approximate Q-values for each possible action from input states. To process visual inputs effectively and capture temporal motion, raw game screens are grayscaled, resized to 84x84 pixels, and stacked in groups of four consecutive frames before passing through convolutional and fully connected layers. The training process stabilizes learning through experience replay buffers, fixed target networks that update periodically, and Double DQN architectures that decouple action selection from target evaluation. Agents can then be trained on environments like Space Invaders using frameworks such as RL-Zoo without tabular memory constraints.


[Newer posts](https://yomu.fyi/topic/machine-learning/page/24.md) · [Older posts](https://yomu.fyi/topic/machine-learning/page/26.md)