MLSysBook: Foundations of Machine Learning Systems
An open textbook on machine-learning systems, covering the hardware and software stack for training and serving ML models, distributed training, inference optimisation and systems design.
Search titles, authors, subjects and dates from across the DARE African Knowledge Repository.
102 resources found in the repository.
An open textbook on machine-learning systems, covering the hardware and software stack for training and serving ML models, distributed training, inference optimisation and systems design.
A practical four-chapter introduction to AI, NLP and machine learning with Python: LSTMs, tokenisation, stopwords, lemmatisation, bag-of-words, supervised learning and text classification.
Open textbook for an introductory machine-learning course in mechanical engineering, with runnable Python examples covering supervised learning, gradient descent, generative adversarial networks, variational autoencoders, normalising flows, diffusion models, reinforcement learning and transformers.
31,102 verse-aligned Bible verses in isiNdebele (Zimbabwe) with full book/chapter/verse metadata; 66 books.
7,957 verse-aligned New Testament verses in chiShona with book/chapter/verse metadata, 27 books, Parquet format.
Translations of a subset of XNLI into 16 African languages with validation and test splits for natural language inference evaluation.
Translations of a subset of MMLU into 15 African languages to benchmark multitask knowledge and reasoning.
A 0.4B-parameter transformer language model trained on African languages by Lelapa AI, released with associated research (arXiv:2408.17024).
~803k aligned four-way parallel sentences across English, Nuer, Dinka and Swahili for multilingual machine translation and cross-lingual transfer.
Evaluation set of GSM8K translations into 16 African languages with 250 test problems per language for multilingual mathematical reasoning.
110,000+ annotated tweets for sentiment analysis across 14 under-represented African languages.
Largest publicly available topic-classification dataset based on Flores-200, covering 205 languages and dialects with a compact seven-class label set.