## What type of PR is this? (check all applicable)
- [ ] Refactor
- [ ] Feature
- [X] Bug Fix
- [ ] Optimization
- [ ] Documentation Update
- [ ] Community Node Submission
## Have you discussed this change with the InvokeAI team?
- [X] Yes
- [ ] No, because:
## Have you updated all relevant documentation?
- [X] Yes
- [ ] No
## Description
At install time, when the user's config specified "auto" precision, the
installer was downloading the fp32 models even when an fp16 model would
be appropriate for the OS.
## Related Tickets & Documents
<!--
For pull requests that relate or close an issue, please include them
below.
For example having the text: "closes #1234" would connect the current
pull
request to issue 1234. And when we merge the pull request, Github will
automatically close the issue.
-->
- Closes#4127
## What type of PR is this? (check all applicable)
- [ ] Refactor
- [x] Feature
- [ ] Bug Fix
- [ ] Optimization
- [ ] Documentation Update
- [ ] Community Node Submission
## Have you discussed this change with the InvokeAI team?
- [x] Yes
- [ ] No, because:
## Have you updated all relevant documentation?
- [ ] Yes
- [x] No
## Description
Add lora loading for sdxl.
NOT TESTED - I run only 2 loras, please check more(including lycoris if
they already exists).
## QA Instructions, Screenshots, Recordings
https://civitai.com/models/118536/voxel-xl
## Added/updated tests?
- [ ] Yes
- [x] No
## What type of PR is this? (check all applicable)
- [ ] Refactor
- [ ] Feature
- [x] Bug Fix
- [ ] Optimization
- [ ] Documentation Update
- [ ] Community Node Submission
## Have you discussed this change with the InvokeAI team?
- [x] Yes
- [ ] No, because:
## Have you updated all relevant documentation?
- [ ] Yes
- [ ] No
## Description
## Related Tickets & Documents
<!--
For pull requests that relate or close an issue, please include them
below.
For example having the text: "closes #1234" would connect the current
pull
request to issue 1234. And when we merge the pull request, Github will
automatically close the issue.
-->
- Related Issue #
- Closes #
## QA Instructions, Screenshots, Recordings
<!--
Please provide steps on how to test changes, any hardware or
software specifications as well as any other pertinent information.
-->
## Added/updated tests?
- [ ] Yes
- [ ] No : _please replace this line with details on why tests
have not been included_
## [optional] Are there any post deployment tasks we need to perform?
## What type of PR is this? (check all applicable)
- [X] Feature
## Have you discussed this change with the InvokeAI team?
- [X] Yes
- [ ] No, because:
## Have you updated all relevant documentation?
- [X] Yes
- [ ] No
## Description
This PR adds execution time and VRAM usage reporting to each graph
invocation. The log output will look like this:
```
[2023-08-02 18:03:04,507]::[InvokeAI]::INFO --> Graph stats: c7764585-9c68-4d9d-a199-55e8186790f3
[2023-08-02 18:03:04,507]::[InvokeAI]::INFO --> Node Calls Seconds VRAM Used
[2023-08-02 18:03:04,507]::[InvokeAI]::INFO --> main_model_loader 1 0.005s 0.01G
[2023-08-02 18:03:04,508]::[InvokeAI]::INFO --> clip_skip 1 0.004s 0.01G
[2023-08-02 18:03:04,508]::[InvokeAI]::INFO --> compel 2 0.512s 0.26G
[2023-08-02 18:03:04,508]::[InvokeAI]::INFO --> rand_int 1 0.001s 0.01G
[2023-08-02 18:03:04,508]::[InvokeAI]::INFO --> range_of_size 1 0.001s 0.01G
[2023-08-02 18:03:04,508]::[InvokeAI]::INFO --> iterate 1 0.001s 0.01G
[2023-08-02 18:03:04,508]::[InvokeAI]::INFO --> metadata_accumulator 1 0.002s 0.01G
[2023-08-02 18:03:04,508]::[InvokeAI]::INFO --> noise 1 0.002s 0.01G
[2023-08-02 18:03:04,508]::[InvokeAI]::INFO --> t2l 1 3.541s 1.93G
[2023-08-02 18:03:04,508]::[InvokeAI]::INFO --> l2i 1 0.679s 0.58G
[2023-08-02 18:03:04,508]::[InvokeAI]::INFO --> TOTAL GRAPH EXECUTION TIME: 4.749s
[2023-08-02 18:03:04,508]::[InvokeAI]::INFO --> Current VRAM utilization 0.01G
```
On systems without CUDA, the VRAM stats are not printed.
The current implementation keeps track of graph ids separately so will
not be confused when several graphs are executing in parallel. It
handles exceptions, and it is integrated into the app framework by
defining an abstract base class and storing an implementation instance
in `InvocationServices`.
