Next, in roofAI 🏛️, a giza 🔻 fix, @option::subset x=3,z=4 x=1,y=2 giza update push move into a script: @code roofAI-train @source roofAI-train help

Next, in roofAI 🏛️, train with a dataset generated on the fly, on GPU, and register, using https://github.com/kamangir/roofAI/blob/main/roofAI/semseg/README.md, options=om=0,profile=VALIDATION,~register echo $( @option::subset \ profile=FULL,register,suffix=gpu_v1 \ "$options" ) At the gallery, booting up sagemaker,

Next, in roofAI 🏛️,

Next, in roofAI 🏛️, boot up SageMaker with GPU, from https://aws.amazon.com/sagemaker/pricing/

Next, in roofAI 🏛️, AWS CloudShell @seed cloudshell screen

Next, in 🛠️ OpenAI Vision, function runme() { local tag=2023-11-25-tag-v1 @tag set x1 $tag @tag set x2 $tag @tag set x3 $tag @tag set x4 $tag @tag set x5 $tag @tag set x6 $tag @tag search $tag } runme 6 object(s): x1, x2, x3, x4, x5, x6 @tag search $tag …

Next, in 🛠️ OpenAI Vision, review cost, vanwatch vision \ "these images show the same intersection at the same time \ from different angles, write a short story that happens \ in this intersection at this time." \ area=vancouver,offset=0 \ Davie,Bute fix the issue with finish_reason. ✅ 🤖 success: On…

Next, in 🛠️ OpenAI Vision, log URLs also when not verbose. ✅ start from 1 ✅

Next, in 🛠️ OpenAI Vision, openai vision \ "you are a police offier, what do you see in these images?" \ - Davie,Bute,.jpg \ $(vanwatch list area=vancouver,ingest,published \ --log 0 \ --count 1 \ --offset 0) \ --max_count 10 \ --verbose 1