curriculum
stage2_mixed_curriculum_v1
Stage 2 Mixed Text/Phoneme TTS Dataset
This dataset contains mixed text/phoneme sequences for TTS training with curriculum learning.
Curriculum Learning
The probability of converting words to phonemes increases over the dataset:
Start: p = 0.3 (more text, less phonemes)
End: p = 1.0 (all phonemes)
Transition: Linear over 500,000 rows
Each row uses p(i) for ALL its words/spaces, then i increments for the next row.
Features
Column
Description
text… See the full description on the dataset page: https://huggingface.co/datasets/AdoCleanCode/stage2_mixed_curriculum_v1.stratified_10m_curriculum
Dataset Card for Stratified 10M Curriculum
This is a stratified split by domain of the raw datasets used by the 2024 BabyLM challange.
Sampled from the original datasets, equal amounts of tokens for each stage (C1,...C5).
Child-directed speech accounts for nearly half of the original dataset by word count.
In preliminary experiments using a training data influence estimation method, this category was by far the most influential.
This dataset enables us to investigate whether this… See the full description on the dataset page: https://huggingface.co/datasets/babylm-anon/stratified_10m_curriculum.llasa_curriculum_dataset_englishbabylm_2024_10m_curriculum
Dataset Card for BabyLM 2024 10M Curriculum
The documents from the 10M dataset provided by the 2024 BabyLM challange.
We add a validation split we with additional documents from the 100M dataset.
The order in which to load this dataset is provided as .pt files (e.g., curriculum.pt or random.pt).
Pretraining split (train)
Stage
Words
Documents
C1: Child Directed Speech
2839591
28.53%
580000
49.19%
C2: Unscripted Dialogue
1079286
10.84%
108000
9.16%
C3:… See the full description on the dataset page: https://huggingface.co/datasets/babylm-anon/babylm_2024_10m_curriculum.capstone-curriculumvision-opd-vqa14k-fullimage-curriculum-v8
Vision-OPD VQA14K Full-Image Curriculum v8
Private single-image visual-question-answering dataset.
Split
Rows
Train
14,000
Diagnostic validation
609
The repository contains 14,609 content-addressed media files (4,580,273,467 bytes). Paths in both Parquet files are relative to the repository root and follow media/<sha256-prefix>/<filename>.
from pathlib import Path
import pyarrow.parquet as pq
from huggingface_hub import snapshot_download
root =… See the full description on the dataset page: https://huggingface.co/datasets/yyy051007/vision-opd-vqa14k-fullimage-curriculum-v8.