search

G-U-N / Phased-Consistency-Model

Boosting the performance of consistency models with PCM!
https://g-u-n.github.io/projects/pcm/
Apache License 2.0
296 stars 10 forks source link

Format of the dataset? #10

Open kylechang523 opened 5 days ago

kylechang523 commented 5 days ago

Hi, what an awesome work. Can you please share the dataset you are using to distillate the SD1.5? Or provide some sample data that we can try? Thank you so much!

G-U-N commented 5 days ago

Hi, I use this tool image2dataset to download the data.

command

img2dataset --url_list cc3m.tsv --input_format "tsv"\
         --url_col "url" --caption_col "caption" --output_format webdataset\
           --output_folder cc3m --processes_count 16 --thread_count 64 --resize_mode no \
             --enable_wandb True

Then I extract all tars into one folder.

kylechang523 commented 3 days ago

Hi, I use this tool image2dataset to download the data.

command

img2dataset --url_list cc3m.tsv --input_format "tsv"\
         --url_col "url" --caption_col "caption" --output_format webdataset\
           --output_folder cc3m --processes_count 16 --thread_count 64 --resize_mode no \
             --enable_wandb True

Then I extract all tars into one folder.

Thank you！