Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[SD] unified val file names #685

Merged
merged 1 commit into from
Feb 29, 2024
Merged

[SD] unified val file names #685

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion stable_diffusion/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -207,7 +207,7 @@ We achieved this subset by following these steps:

1. download coco-2014 validation dataset: `scripts/datasets/coco-2014-validation-download.sh --output-dir /datasets/coco2014`
2. create the validation subset, and resize the images to 512x512: `scripts/datasets/coco-2014-validation-split-resize.sh --input-images-path /datasets/coco2014/val2014 --input-coco-captions /datasets/coco2014/annotations/captions_val2014.json --output-images-path /datasets/coco2014/val2014_512x512_30k --output-tsv-file /datasets/coco2014/val2014_30k.tsv`
3. generate the FID statistics: `scripts/datasets/generate-fid-statistics.sh --dataset-dir /datasets/coco2014/val2014_512x512_30k --output-file /datasets/coco2014/val2014_512x512_30k_stats.npz`
3. generate the FID statistics: `scripts/datasets/generate-fid-statistics.sh --dataset-dir /datasets/coco2014/val2014_512x512_30k --output-file /datasets/coco2014/val2014_30k_stats.npz`

## The Model
Stable Diffusion v2 is a latent diffusion model which combines an autoencoder with a diffusion model that is trained in the latent space of the autoencoder. During training:
Expand Down
2 changes: 1 addition & 1 deletion stable_diffusion/configs/train_01x08x08.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@ model:
enabled: True
inception_weights_url: https://github.com/mseitzer/pytorch-fid/releases/download/fid_weights/pt_inception-2015-12-05-6726825d.pth
cache_dir: /checkpoints/inception
gt_path: /datasets/coco2014/val2014_512x512_30k_stats.npz
gt_path: /datasets/coco2014/val2014_30k_stats.npz
clip:
enabled: True
clip_version: "ViT-H-14"
Expand Down
2 changes: 1 addition & 1 deletion stable_diffusion/configs/train_32x08x02.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@ model:
enabled: True
inception_weights_url: https://github.com/mseitzer/pytorch-fid/releases/download/fid_weights/pt_inception-2015-12-05-6726825d.pth
cache_dir: /checkpoints/inception
gt_path: /datasets/coco2014/val2014_512x512_30k_stats.npz
gt_path: /datasets/coco2014/val2014_30k_stats.npz
clip:
enabled: True
clip_version: "ViT-H-14"
Expand Down
2 changes: 1 addition & 1 deletion stable_diffusion/configs/train_32x08x02_raw_images.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@ model:
enabled: True
inception_weights_url: https://github.com/mseitzer/pytorch-fid/releases/download/fid_weights/pt_inception-2015-12-05-6726825d.pth
cache_dir: /checkpoints/inception
gt_path: /datasets/coco2014/val2014_512x512_30k_stats.npz
gt_path: /datasets/coco2014/val2014_30k_stats.npz
clip:
enabled: True
clip_version: "ViT-H-14"
Expand Down
2 changes: 1 addition & 1 deletion stable_diffusion/configs/train_32x08x04.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@ model:
enabled: True
inception_weights_url: https://github.com/mseitzer/pytorch-fid/releases/download/fid_weights/pt_inception-2015-12-05-6726825d.pth
cache_dir: /checkpoints/inception
gt_path: /datasets/coco2014/val2014_512x512_30k_stats.npz
gt_path: /datasets/coco2014/val2014_30k_stats.npz
clip:
enabled: True
clip_version: "ViT-H-14"
Expand Down
2 changes: 1 addition & 1 deletion stable_diffusion/configs/train_32x08x08.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@ model:
enabled: True
inception_weights_url: https://github.com/mseitzer/pytorch-fid/releases/download/fid_weights/pt_inception-2015-12-05-6726825d.pth
cache_dir: /checkpoints/inception
gt_path: /datasets/coco2014/val2014_512x512_30k_stats.npz
gt_path: /datasets/coco2014/val2014_30k_stats.npz
clip:
enabled: True
clip_version: "ViT-H-14"
Expand Down
2 changes: 1 addition & 1 deletion stable_diffusion/scripts/datasets/coco-2014-validation-split-resize.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@

: "${INPUT_IMAGES_PATH:=/datasets/coco2014/val2014}"
: "${INPUT_COCO_CAPTIONS:=/datasets/coco2014/annotations/captions_val2014.json}"
: "${OUTPUT_IMAGES_PATH:=/datasets/coco2014/val2014_512x512_30k}"
: "${OUTPUT_IMAGES_PATH:=/datasets/coco2014/val2014_30k}"
: "${OUTPUT_TSV_FILE:=/datasets/coco2014/val2014_30k.tsv}"
: "${NUM_SAMPLES:=30000}"

Expand Down
4 changes: 2 additions & 2 deletions stable_diffusion/scripts/datasets/generate-fid-statistics.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,7 @@
#!/usr/bin/env bash

: "${DATASET_DIR:=/datasets/coco2014/val2014_512x512_30k}"
: "${OUTPUT_FILE:=/datasets/coco2014/val2014_512x512_30k_stats.npz}"
: "${DATASET_DIR:=/datasets/coco2014/val2014_30k}"
: "${OUTPUT_FILE:=/datasets/coco2014/val2014_30k_stats.npz}"

while [ "$1" != "" ]; do
case $1 in
Expand Down
Loading