* Update dream.py. k_euler_a and k_dpm_2_a M1 fix
Make results reproducible (so runs with the same seed produce the same result).
Implements fix by @wbowling referenced in https://github.com/lstein/stable-diffusion/issues/397#issuecomment-1240679294
* Update dream.py. Remove import torch from dream.py
* generate.py: k_euler_a and k_dpm_2_a M1 fix#579
Co-authored-by: Lincoln Stein <lincoln.stein@gmail.com>
* Refactor generate.py and dream.py
* config file path (models.yaml) is parsed inside Generate() to simplify
API
* Better handling of keyboard interrupts in file loading mode vs
interactive
* Removed oodles of unused variables.
* move nonfunctional inpainting out of the scripts directory
* fix ugly ddim tqdm formatting
* fix embiggen breakage, formatting fixes
* fix web server handling of rel and abs outdir paths
* Can now specify either a relative or absolute path for outdir
* Outdir path does not need to be inside the stable-diffusion directory
* Closes security hole that allowed user to read any file within
stable-diffusion (eek!)
* Closes#536
* revert inadvertent change of conda env name (#528)
* Refactor generate.py and dream.py
* config file path (models.yaml) is parsed inside Generate() to simplify
API
* Better handling of keyboard interrupts in file loading mode vs
interactive
* Removed oodles of unused variables.
* move nonfunctional inpainting out of the scripts directory
* fix ugly ddim tqdm formatting
Code cleanup and attention.py einsum_ops update for M1 16-32GB performance.
Expected: On par with fastest ever from 8 to 128GB for 512x512. Allows large images.
When running on just cpu (intel), a call to torch.layer_norm would error with RuntimeError: expected scalar type BFloat16 but found Float
Fix buggy device handling in model.py.
Tested with scripts/dream.py --full_precision on just cpu on intel laptop. Works but slow at ~10s/it.
* Add Embiggen automation
* Make embiggen_tiles masking more intelligent and count from one (at least for the user), rewrite sections of Embiggen README, fix various typos throughout README
* drop duplicate log message
* This moves the call to half() before model.to(device) to avoid GPU
copy of full model. Improves speed and reduces memory usage dramatically
* This fix contributed by @mh-dm (Mihai)
