Merge pull request #293 from gridai/dev

Docs Release

Author: alexandercort

docs/features/runs/1_Creating Runs/1_Basic Runs/3_sweep-syntax.md

@@ -9,7 +9,9 @@ Check the [Runs FAQ](https://docs.grid.ai/features/runs/faq.md) for common quest
 
 # Hyperparameter Sweeps
 
-Grid Runs offer the abilty to launch [hyperparameter sweeps](https://www.grid.ai/what-are-hyperparameter-sweeps-and-why-are-they-important-to-production-machine-learning) without changing a single line of code! Just make sure your script can take arguments from the CLI:
+Grid Runs offer the abilty to launch [hyperparameter sweeps](https://www.grid.ai/what-are-hyperparameter-sweeps-and-why-are-they-important-to-production-machine-learning) without changing a single line of code!
+This feature will allow you to run multiple permutations of your training script (i.e. an experiment) on different machines in parallel.
+Just make sure your training script can take arguments from the CLI:
 
 ```bash
 grid run main.py --layers 32 --learning_rate 0.01

docs/features/runs/1_README.md

@@ -80,14 +80,16 @@ grid run hello.py --number "[1, 2]" --food_item "['pizza', 'hotdog']"
 The above is equivalent to running each of the following lines on a separate machine:
 
 ```bash
-python hello.py --number 1 --food_item 'pizza'
-python hello.py --number 2 --food_item 'pizza'
-python hello.py --number 1 --food_item 'hotdog'
-python hello.py --number 2 --food_item 'hotdog'
+python hello.py --number 1 --food_item 'pizza'  # will run on machine 1
+python hello.py --number 2 --food_item 'pizza'  # will run on machine 2
+python hello.py --number 1 --food_item 'hotdog' # will run on machine 3
+python hello.py --number 2 --food_item 'hotdog' # will run on machine 4
 ```
 
 :::note
-A RUN is a collection of experiments (the run has 4 experiments in this example).
+A RUN is a collection of EXPERIMENTS (the run has 4 experiments in this example).
+
+Each experiment will execute on it's own machine!
 :::
 
 To see the status of your Run and all associated experiments, run the `grid status <my-run-name>` command. (More details can be found [here](https://docs.grid.ai/cli#grid-status)).

docs/platform/10_known-issues.md

@@ -9,7 +9,7 @@ sidebar_label: Known Issues
 
 ### [BYOC](https://docs.grid.ai/platform/custom-cloud-credentials/byoc)
 * Creating runs with the same name across two cluster is currenly supported, but prohibits certain operations against runs/experiments.
-* BYOC can only support instance types with the x86_64 architecture. The instance families of the following instance types are known to not be supported due to their architecture: A1, T4g, M6g, C6g, and R6g.
+* BYOC can only support instance types with the x86_64 architecture. The instance families of the following instance types are known to not be supported due to their architecture: A1, T4g, M6g, C6g, G5g, and R6g.
 
 ### [Collaboration](https://docs.grid.ai/platform/collaboration)
 * Downloading artifacts from a teammate's experiments is not supported from the CLI. Please use the UI as a workaround.
@@ -27,3 +27,6 @@ sidebar_label: Known Issues
 
 ### [Session](https://docs.grid.ai/features/sessions)
 * Cloning a Github repo with SSH is currenlty not supported. Insead, use Github's HTTPS clone method.
+
+### [Artifacts](https://docs.grid.ai/features/runs/Analyzing%20Runs/artifacts)
+* Currently, artifact downloads are not optimized for a large number of files. This can lead to excessively long download times.