Dataset
apple-patents-embeddings
Vector embeddings of Apple patent text totalling about 4 million entries, stored as parquet shards with a 768-dimensional float64 feature per record.
Data preview
Dataset card
The apple-patents-embeddings dataset, published by Sutro, packages more than four million Apple patent records as precomputed vector embeddings stored in Parquet format. It is released under an MIT license and is intended for machine-learning workflows over patent text.
What this dataset contains
The dataset contains roughly 4,039,988 rows, all assigned to a single train split, sourced from Apple's patent filings. Each row pairs a piece of patent text with a dense embedding vector generated from that text, enabling similarity search, clustering, and other downstream NLP tasks. The reported dataset size on disk is approximately 37 GB, with a compressed download size near 7.4 GB.
Structure and fields
The schema is intentionally minimal and consists of two columns. The first column, text, holds the raw patent passage as a large_string value. The second column, named job-76844041-b2bf-4248-9603-b7f750231b34, stores the corresponding float64 embedding vector for that passage. Data is distributed as Parquet and optimized Parquet files, and the dataset is compatible with the Hugging Face datasets, Dask, Polars, and ML Croissant libraries.
How analysts use it
Quantitative and intellectual-property analysts can use the embeddings to map Apple's technology footprint, identify thematic clusters such as optics or chip design, and track patenting intensity over time as an indicator of R&D direction. Buy-side and corporate-strategy teams may also incorporate the vectors into competitive-intelligence dashboards, monitoring similarity between new filings and existing technology areas to inform investment or partnership decisions in Apple-exposed names.
Documentation & references 1
Related datasets
Browse moreParsed eligibility criteria from 13,229 ClinicalTrials.gov studies representing the candidate pool gathered by two first-stage retrievers during TREC Clinical Trials 2021–2023, formatted as typed entity-relation graphs for reranking models.
The clinical-trials-patient-graphs dataset is a small tabular collection of patient-level clinical records distributed in Parquet format by the 2001jdev publisher. It organizes entities and relations extracted from trial data into structured rows spanning 2021 through 2023.
This dataset is a parsed snapshot of ClinicalTrials.gov records, distributed in Parquet format on the Hugging Face Hub under an unspecified license. It contains roughly 52,000 trial entries drawn from public registry filings.
clinical-trials-trec-qrels is a tabular relevance-judgment file distributed by the 2001jdev user on the Hugging Face Hub. It maps clinical-trial topic identifiers to NCT registry IDs with graded relevance scores.
The clinical-trials-trec-topics dataset is a small tabular collection of clinical-trial topic records distributed in Parquet format on the Hugging Face Hub. It appears to be derived from TREC clinical-trial retrieval topics, with rows partitioned by year.
SEC EDGAR 10-K & 10-Q financial statements for 2,600+ NASDAQ companies.
Discussion
Questions or observations about this dataset? Join the conversation on X.
No comments yet. Start the conversation on X.