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

Better defaults for LTXV #198

Merged
merged 2 commits into from
Jan 9, 2025
Merged

Better defaults for LTXV #198

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
4cc5231
update
a-r-r-o-w Jan 9, 2025
8f44536
Merge branch 'main' into better-defaults-ltx
sayakpaul Jan 9, 2025
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
8 changes: 5 additions & 3 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -43,6 +43,8 @@ Then launch LoRA fine-tuning. Below we provide an example for LTX-Video. We refe
<details>
<summary>Training command</summary>

TODO: LTX does not do too well with the disney dataset. We will update this to use a better example soon.

```bash
#!/bin/bash
export WANDB_MODE="offline"
Expand Down Expand Up @@ -75,18 +77,18 @@ dataset_cmd="--data_root $DATA_ROOT \
dataloader_cmd="--dataloader_num_workers 0"

# Diffusion arguments
diffusion_cmd="--flow_resolution_shifting"
diffusion_cmd="--flow_weighting_scheme logit_normal"

# Training arguments
training_cmd="--training_type lora \
--seed 42 \
--mixed_precision bf16 \
--batch_size 1 \
--train_steps 1200 \
--train_steps 3000 \
--rank 128 \
--lora_alpha 128 \
--target_modules to_q to_k to_v to_out.0 \
--gradient_accumulation_steps 1 \
--gradient_accumulation_steps 4 \
--gradient_checkpointing \
--checkpointing_steps 500 \
--checkpointing_limit 2 \
Expand Down
6 changes: 3 additions & 3 deletions docs/training/ltx_video.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,18 +36,18 @@ dataset_cmd="--data_root $DATA_ROOT \
dataloader_cmd="--dataloader_num_workers 0"

# Diffusion arguments
diffusion_cmd="--flow_resolution_shifting"
diffusion_cmd="--flow_weighting_scheme logit_normal"

# Training arguments
training_cmd="--training_type lora \
--seed 42 \
--mixed_precision bf16 \
--batch_size 1 \
--train_steps 1200 \
--train_steps 3000 \
--rank 128 \
--lora_alpha 128 \
--target_modules to_q to_k to_v to_out.0 \
--gradient_accumulation_steps 1 \
--gradient_accumulation_steps 4 \
--gradient_checkpointing \
--checkpointing_steps 500 \
--checkpointing_limit 2 \
Expand Down