introduce ilab profile set doc
#130
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: Charlie Doern <cdoern@redhat.com>
khaledsulayman
approved these changes
Sep 3, 2024
| 1. User needs to set: | ||
| a. Model Path for SDG, Eval, Serving, Training | ||
| b. GPUs for SDG, Eval, Serving, Training | ||
| c. Training Config Per-GPU (based on vram) |
There was a problem hiding this comment.
Suggested change
| c. Training Config Per-GPU (based on vram) | |
| c. Training Config Per-GPU (based on vRAM) |
Comment on lines
+6
to
+7
| a. Model Path for SDG, Eval, Serving, Training | ||
| b. GPUs for SDG, Eval, Serving, Training |
There was a problem hiding this comment.
Serving just for vLLM, or vLLM and Llama CPP?
|
|
||
| ### Workflow | ||
|
|
||
| The user will run `ilab profile set` |
There was a problem hiding this comment.
Will this be required before running other ilab commands? If not, what is the path for a user that does not do this?
|
|
||
| The user will run `ilab profile set` | ||
| Alongside the various model paths and GPU amounts, this command will | ||
| set the train profile for the following scenarions: |
There was a problem hiding this comment.
Suggested change
| set the train profile for the following scenarions: | |
| set the train profile for the following scenarios: |
Comment on lines
+18
to
+21
| 2. single consumer GPU | ||
| 3. multi consumer GPU | ||
| 4. single server GPU | ||
| 5. multi server GPU |
There was a problem hiding this comment.
Are terms like server and consumer familiar to users? I myself don't fully understand what this means.
| 5. multi server GPU | ||
| 6. MacOS (once MPS support is added) | ||
|
|
||
| There is also a Choose for me option which reads the Nvidia cards on the system and assigns a cfg+train profile based off the amount of vRAM |
There was a problem hiding this comment.
Suggested change
| There is also a Choose for me option which reads the Nvidia cards on the system and assigns a cfg+train profile based off the amount of vRAM | |
| There is also a "Choose for me" option which reads the Nvidia cards on the system and assigns a cfg+train profile based off the amount of vRAM |
What if I don't have a Nvidia card?
Comment on lines
+33
to
+38
| [1] CPU | ||
| [2] Single Consumer GPU | ||
| [3] Single Server GPU | ||
| [4] Multi Consumer GPU | ||
| [5] Multi Server GPU | ||
| [6] MacOS |
There was a problem hiding this comment.
Is the assumption here systems 0-5 are Linux?
|
@cdoern what is going on with this? |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.