Gautam Mittal
Gautam Mittal
Justin’s requirements.txt will install ray. If the version of ray does not equal our ray version, it will cause some problems. Maybe we should encourage the user to use conda...
Currently Sky only supports TPU nodes, but an investigation by @infwinston and myself with our JAX experiments showed that Google is pushing all support to TPU VMs (see: https://cloud.google.com/blog/products/compute/introducing-cloud-tpu-vms) for...
We currently do not support provisioning multiple TPUs for a cluster, only the head node will have access to the provisioned TPU.
I have turned `geqo` off and increased `from_collapse_limit` and `join_collapse_limit` to appropriate values. I am attempting to plan the following: ``` /*+ HashJoin(ct t chn cn ci mc rt) HashJoin(t...
I have multiple GCP accounts/configurations (TPU research cloud, personal account, various organization accounts, etc.) and would like to be able to switch between them and have SkyPilot automatically use the...
(low priority) based on personal experience. If a user moves laptops/machines/etc., there is no way for their state to transfer with them (sky status is now empty on the new...
This is a (very rough) first pass at implementing a node provider for [Lambda Labs](https://lambdalabs.com/) which has some of the cheapest cloud GPUs available. ### TODO - [x] Single-node support...
``` python -m training.main \ --save-frequency 1 \ --zeroshot-frequency 1 \ --report-to tensorboard \ --hf-tokenizer-name gpt2 \ --train-data=~/flava_cc/Train_GCC-training_output.csv \ --val-data=~/flava_cc/Validation_GCC-1.1.0-Validation_output.csv \ --csv-img-key filepath \ --csv-caption-key title \ --imagenet-val=~/imagenet_validation \ --warmup...
This code used to break from every so often mainly because the API layer around the Xcode toolchain was written with some very bad Node.js code. Some of that was...