{"name":"T1DMAI","repo":"0xdeadf1sh/T1DMAI","url":"https://midorreal.com/project/0xdeadf1sh-t1dmai","repository":"https://github.com/0xdeadf1sh/T1DMAI","description":"A model zoo of blood glucose forecasters trained & tested on my own data","facts":{"license":{"value":"MIT","source":"https://github.com/0xdeadf1sh/T1DMAI","checked":"21 Nov 2026"},"language":{"value":"Python","source":"https://github.com/0xdeadf1sh/T1DMAI","checked":"21 Nov 2026"},"releases":{"value":"0","source":"https://github.com/0xdeadf1sh/T1DMAI/releases","checked":"21 Nov 2026"},"contributors":{"value":"1","source":"https://github.com/0xdeadf1sh/T1DMAI/graphs/contributors","checked":"21 Nov 2026"},"archived":{"value":"no","source":"https://github.com/0xdeadf1sh/T1DMAI","checked":"21 Nov 2026"}},"stars":88.0,"writeup":{"what_it_is":"T1DMAI is a research repository that trains and tests blood glucose forecasting models on the author's own data and the public MetaboNet dataset. Its main model is a small transformer that forecasts glucose 2 hours ahead, in two variants trained with DILATE and MSE loss, with an optional LoRA adapter for fine-tuning.","audience":"Researchers and developers working on blood glucose forecasting for type 1 diabetes.","claims":[{"kind":"specific","claim":"The main model forecasts blood glucose 2 hours ahead.","excerpt":"forecasts blood glucose 2 hours ahead","status":"not_checked"},{"kind":"specific","claim":"The model has 31k parameters and 16 layers of width 16.","excerpt":"16 layers of width 16, 31k parameters","status":"not_checked"},{"kind":"specific","claim":"The LoRA adapter adds 3771 parameters.","excerpt":"adds a frozen-base rank 1 adapter [23] (3771 parameters)","status":"not_checked"},{"kind":"specific","claim":"On the author's data, D16 DILATE + LoRA ranks 3rd of 26 models by score.","excerpt":"D16 DILATE + LoRA comes 3rd of 26 (22.8)","status":"not_checked"},{"kind":"broad","claim":"The project is described as research and educational use only, not a medical device.","excerpt":"Research and educational use only - not a medical device.","status":"too_broad"}],"alternatives":[{"name":"FluidVoice","url":"https://midorreal.com/project/altic-dev-fluidvoice"},{"name":"transformers","url":"https://midorreal.com/project/huggingface-transformers"},{"name":"opal","url":"https://midorreal.com/project/xingjian-bai-opal"},{"name":"daily_stock_analysis","url":"https://midorreal.com/project/zhulinsen-daily-stock-analysis"},{"name":"CLM","url":"https://midorreal.com/project/contrastive-lm-clm"}],"written_by":"AI, from the project's README","date":"2026-11-12"},"why_now":null,"owner_supplied":null,"score":{"verdict":"too_early","hype":92,"reality":46,"gap":46,"momentum":null,"confidence":60,"version":"score-2.0","rule":"verdict-2.0","calculated_at":"2026-11-22T08:00:00+00:00","verdict_source":"rule"},"early_read":{"estimate":true,"lean":"too_close","gap_low":20,"gap_high":62,"rule":"early-2.0"},"changes":[{"date":"2026-07-29T05:07:53+00:00","text":"Repository created","source":"https://github.com/0xdeadf1sh/T1DMAI"}],"cite":"T1DMAI is a GitHub project at https://github.com/0xdeadf1sh/T1DMAI written mainly in Python, described there as \"A model zoo of blood glucose forecasters trained & tested on my own data\".","attribution":"Data from Mid or Real, https://midorreal.com/project/0xdeadf1sh-t1dmai, checked 22 November 2026. Attribution with a link is required.","still_real":null}