The following are modifications or improvements to original notebooks. Please refer to the authors' models for the published primary work.

Hugging Face Pipelines 02Jul24

Here, 32 Hugging Face pipelines were combined into one Google Colab notebook. Pipelines allow for speed, simplicity, and screening of many Deep Learning workflows with minimal code prior to further incorporation into more specific workflows. Broad utility of these Generative and machine learning approaches accelerate R&D far beyond utilizing only ai chat methods. Here, ChatGPT 4o and Hugging Face were used to help assemble the pipelines into a python notebook and were tested with user .rtf, .mp3, .png, and .mp4 files - manually specifying models when not included, on a single NVIDIA A100 40GB GPU. All pipelines ran, with some requiring additional application specific attention. In general, pipeline code was most reliable from individual Hugging Face model pages. Drive 6MB Python Notebook.

Hugging Face pipelines mentioned above: 'text-classification', 'ner', 'question-answering', 'translation_XX_to_YY', 'translation', 'summarization', 'text-generation', 'fill-mask', 'text2text-generation', 'zero-shot-classification', 'conversational', 'feature-extraction', 'image-to-text', 'image-classification', 'object-detection', 'image-segmentation', 'document-question-answering', 'automatic-speech-recognition', 'audio-classification', 'depth-estimation', 'image-feature-extraction', 'mask-generation', 'sentiment-analysis', 'table-question-answering', ‘token-classification’, 'visual-question-answering'(vqa), 'zero-shot-image-classification', 'zero-shot-object-detection', 'zero-shot-audio-classification', 'text-to-speech'(text-to-audio), 'video-classification', 'image-to-image'.