Koan-Sin Tan
Koan-Sin Tan
Amazon price list: https://aws.amazon.com/device-farm/pricing/
@swasson488 Let's reopen this. To to have CI/CD running on service providers who are already running device farm.
close for now since there is no such issue with current deivice.
Does moving large files to some S3 buckets help? It depends on the cloud provider we use (note that S3 is not Amazon). Maybe it's the limitation on client side...
To split a file large then a certain threshold, a simple shell script serve well. It event works on Android. Assume that we want to split all the files larger...
which link is deleted? I know two ways 1. what is used in tinygrad 2. what I did at https://github.com/freedomtan/coreml_to_ane_hwx/blob/main/ane%2FREADME.md
which link is deleted? I know two ways 1. what is used in tinygrad 2. what did at https://github.com/freedomtan/coreml_to_ane_hwx/blob/main/ane%2FREADME.md
@anhappdev to investigate what we should do and to see if we can get the "experimental access" to Google's Tensor ML SDK.
max input tokens: 1,024 max output tokens: 1,024/2,048 (input tokens, output tokens) = ((1024, 2048), (1024, 512)) query == sample == prompt, here @farook-edev: please verify that given a list...
Why we cannot get these numbers with Core ML tools? It seems, for either the [palette ](https://apple.github.io/coremltools/docs-guides/source/opt-palettization-perf.html) or [linear quantization](https://apple.github.io/coremltools/docs-guides/source/opt-quantization-perf.html) schemes, the models are actually using float16 as we knew...