A supervised training dataset for machine learning and AI systems that learn to identify lithology from well-log data. This dataset contains exactly 400 wells with standardized wireline-log measurements and corresponding lithological labels. https://huggingface.co/datasets/NoraResearchLab/Lithology-Training-Dataset This dataset is intended for training and model development. It is separate from the independent evaluation benchmark. - SHALE - SANDSTONE - LIMESTONE - DOLOMITE - ANHYDRITE - SALT - COAL - UNKNOWN/MIXED UNKNOWN/MIXED represents intervals where the available log evidence does not provide sufficient confidence for assigning one of the principal lithological classes. Depth-indexed…