Reproducing Results

[tanriverdiege]

@brandontrabucco Thank you for the great research and open-sourcing your code. I was trying to reproduce the plots in your paper (Figure 5). I am using the tokens you have provided in Google Drive. I am making a mistake somewhere and can not reproduce the exact plots. If it is possible can you provide the exact arguments for the train_classifier.py script so that I can obtain the results in the paper. Thank you.

[vaventt]

Have you tried these arguments?

python train_classifier.py --logdir pascal-baselines/textual-inversion-1.0-0.75-0.5-0.25 \
--synthetic-dir "/projects/rsalakhugroup/btrabucc/aug/\
textual-inversion-1.0-0.75-0.5-0.25/{dataset}-{seed}-{examples_per_class}" \
--dataset pascal --prompt "a photo of a {name}" \
--aug textual-inversion textual-inversion textual-inversion textual-inversion \
--guidance-scale 7.5 7.5 7.5 7.5 \
--strength 1.0 0.75 0.5 0.25 \
--mask 0 0 0 0 \
--inverted 0 0 0 0 \
--probs 0.25 0.25 0.25 0.25 \
--compose parallel --num-synthetic 10 --synthetic-probability 0.5 \
--num-trials 8 --examples-per-class 1 2 4 8 16

[Julius-R-UI]

@tanriverdiege where you ever able recreate the results + what drive link?