Update readme and QwenImageTransformer2DModel (#321)
This commit is contained in:
Bubbliiiing
@@ -10,10 +10,10 @@ Some parameters in the sh file can be confusing, and they are explained in this
|
||||
- `random_frame_crop` is used for random cropping on video frames to simulate videos with different frame counts.
|
||||
- `random_hw_adapt` is used to enable automatic height and width scaling for images and videos. When `random_hw_adapt` is enabled, the training images will have their height and width set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum. For training videos, the height and width will be set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=1024`, the resolution of image inputs for training is `512x512` to `1024x1024`, and the resolution of video inputs for training is `512x512x49` to `1024x1024x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- `training_with_video_token_length` specifies training the model according to token length. For training images and videos, the height and width will be set to `image_sample_size` as the maximum and `video_sample_size` as the minimum.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- The token length for a video with dimensions 512x512 and 49 frames is 13,312. We need to set the `token_sample_size = 512`.
|
||||
- At 512x512 resolution, the number of video frames is 49 (~= 512 * 512 * 49 / 512 / 512).
|
||||
- At 768x768 resolution, the number of video frames is 21 (~= 512 * 512 * 49 / 768 / 768).
|
||||
|
||||
@@ -30,10 +30,10 @@ Some parameters in the sh file can be confusing, and they are explained in this
|
||||
- `random_frame_crop` is used for random cropping on video frames to simulate videos with different frame counts.
|
||||
- `random_hw_adapt` is used to enable automatic height and width scaling for images and videos. When `random_hw_adapt` is enabled, the training images will have their height and width set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum. For training videos, the height and width will be set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=1024`, the resolution of image inputs for training is `512x512` to `1024x1024`, and the resolution of video inputs for training is `512x512x49` to `1024x1024x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- `training_with_video_token_length` specifies training the model according to token length. For training images and videos, the height and width will be set to `image_sample_size` as the maximum and `video_sample_size` as the minimum.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- The token length for a video with dimensions 512x512 and 49 frames is 13,312. We need to set the `token_sample_size = 512`.
|
||||
- At 512x512 resolution, the number of video frames is 49 (~= 512 * 512 * 49 / 512 / 512).
|
||||
- At 768x768 resolution, the number of video frames is 21 (~= 512 * 512 * 49 / 768 / 768).
|
||||
|
||||
@@ -8,10 +8,10 @@ Some parameters in the sh file can be confusing, and they are explained in this
|
||||
- `random_frame_crop` is used for random cropping on video frames to simulate videos with different frame counts.
|
||||
- `random_hw_adapt` is used to enable automatic height and width scaling for images and videos. When `random_hw_adapt` is enabled, the training images will have their height and width set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum. For training videos, the height and width will be set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=1024`, the resolution of image inputs for training is `512x512` to `1024x1024`, and the resolution of video inputs for training is `512x512x49` to `1024x1024x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- `training_with_video_token_length` specifies training the model according to token length. For training images and videos, the height and width will be set to `image_sample_size` as the maximum and `video_sample_size` as the minimum.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- The token length for a video with dimensions 512x512 and 49 frames is 13,312. We need to set the `token_sample_size = 512`.
|
||||
- At 512x512 resolution, the number of video frames is 49 (~= 512 * 512 * 49 / 512 / 512).
|
||||
- At 768x768 resolution, the number of video frames is 21 (~= 512 * 512 * 49 / 768 / 768).
|
||||
|
||||
Executable
+165
@@ -0,0 +1,165 @@
|
||||
## Training Code
|
||||
|
||||
We can choose whether to use deepspeed or fsdp in qwen-image, which can save a lot of video memory.
|
||||
|
||||
Some parameters in the sh file can be confusing, and they are explained in this document:
|
||||
|
||||
- `enable_bucket` is used to enable bucket training. When enabled, the model does not crop the images at the center, but instead, it trains the entire images after grouping them into buckets based on resolution.
|
||||
- `random_hw_adapt` is used to enable automatic height and width scaling for images. When `random_hw_adapt` is enabled, the training images will have their height and width set to `image_sample_size` as the maximum and `512` as the minimum.
|
||||
- For example, when `random_hw_adapt` is enabled, `image_sample_size=1024`, the resolution of image inputs for training is `512x512` to `1024x1024`
|
||||
- `resume_from_checkpoint` is used to set the training should be resumed from a previous checkpoint. Use a path or `"latest"` to automatically select the last available checkpoint.
|
||||
|
||||
Without deepspeed:
|
||||
|
||||
Training qwen-image without DeepSpeed may result in insufficient GPU memory.
|
||||
```sh
|
||||
export MODEL_NAME="models/Diffusion_Transformer/Qwen-Image"
|
||||
export DATASET_NAME="datasets/internal_datasets/"
|
||||
export DATASET_META_NAME="datasets/internal_datasets/metadata.json"
|
||||
# NCCL_IB_DISABLE=1 and NCCL_P2P_DISABLE=1 are used in multi nodes without RDMA.
|
||||
# export NCCL_IB_DISABLE=1
|
||||
# export NCCL_P2P_DISABLE=1
|
||||
NCCL_DEBUG=INFO
|
||||
|
||||
accelerate launch --mixed_precision="bf16" scripts/qwenimage/train.py \
|
||||
--pretrained_model_name_or_path=$MODEL_NAME \
|
||||
--train_data_dir=$DATASET_NAME \
|
||||
--train_data_meta=$DATASET_META_NAME \
|
||||
--train_batch_size=1 \
|
||||
--image_sample_size=1024 \
|
||||
--gradient_accumulation_steps=1 \
|
||||
--dataloader_num_workers=8 \
|
||||
--num_train_epochs=100 \
|
||||
--checkpointing_steps=50 \
|
||||
--learning_rate=2e-05 \
|
||||
--lr_scheduler="constant_with_warmup" \
|
||||
--lr_warmup_steps=100 \
|
||||
--seed=42 \
|
||||
--output_dir="output_dir" \
|
||||
--gradient_checkpointing \
|
||||
--mixed_precision="bf16" \
|
||||
--adam_weight_decay=3e-2 \
|
||||
--adam_epsilon=1e-10 \
|
||||
--vae_mini_batch=1 \
|
||||
--max_grad_norm=0.05 \
|
||||
--enable_bucket \
|
||||
--uniform_sampling \
|
||||
--trainable_modules "."
|
||||
```
|
||||
|
||||
With deepspeed zero-2:
|
||||
|
||||
```sh
|
||||
export MODEL_NAME="models/Diffusion_Transformer/Qwen-Image"
|
||||
export DATASET_NAME="datasets/internal_datasets/"
|
||||
export DATASET_META_NAME="datasets/internal_datasets/metadata.json"
|
||||
# NCCL_IB_DISABLE=1 and NCCL_P2P_DISABLE=1 are used in multi nodes without RDMA.
|
||||
# export NCCL_IB_DISABLE=1
|
||||
# export NCCL_P2P_DISABLE=1
|
||||
NCCL_DEBUG=INFO
|
||||
|
||||
accelerate launch --use_deepspeed --deepspeed_config_file config/zero_stage2_config.json --deepspeed_multinode_launcher standard scripts/qwenimage/train.py \
|
||||
--pretrained_model_name_or_path=$MODEL_NAME \
|
||||
--train_data_dir=$DATASET_NAME \
|
||||
--train_data_meta=$DATASET_META_NAME \
|
||||
--train_batch_size=1 \
|
||||
--image_sample_size=1024 \
|
||||
--gradient_accumulation_steps=1 \
|
||||
--dataloader_num_workers=8 \
|
||||
--num_train_epochs=100 \
|
||||
--checkpointing_steps=50 \
|
||||
--learning_rate=2e-05 \
|
||||
--lr_scheduler="constant_with_warmup" \
|
||||
--lr_warmup_steps=100 \
|
||||
--seed=42 \
|
||||
--output_dir="output_dir" \
|
||||
--gradient_checkpointing \
|
||||
--mixed_precision="bf16" \
|
||||
--adam_weight_decay=3e-2 \
|
||||
--adam_epsilon=1e-10 \
|
||||
--vae_mini_batch=1 \
|
||||
--max_grad_norm=0.05 \
|
||||
--enable_bucket \
|
||||
--uniform_sampling \
|
||||
--trainable_modules "."
|
||||
```
|
||||
|
||||
Deepspeed zero-3:
|
||||
|
||||
After training, you can use the following command to get the final model:
|
||||
```sh
|
||||
python scripts/zero_to_bf16.py output_dir/checkpoint-{our-num-steps} output_dir/checkpoint-{your-num-steps}-outputs --max_shard_size 80GB --safe_serialization
|
||||
```
|
||||
|
||||
Training shell command is as follows:
|
||||
```sh
|
||||
export MODEL_NAME="models/Diffusion_Transformer/Qwen-Image"
|
||||
export DATASET_NAME="datasets/internal_datasets/"
|
||||
export DATASET_META_NAME="datasets/internal_datasets/metadata.json"
|
||||
# NCCL_IB_DISABLE=1 and NCCL_P2P_DISABLE=1 are used in multi nodes without RDMA.
|
||||
# export NCCL_IB_DISABLE=1
|
||||
# export NCCL_P2P_DISABLE=1
|
||||
NCCL_DEBUG=INFO
|
||||
|
||||
accelerate launch --zero_stage 3 --zero3_save_16bit_model true --zero3_init_flag true --use_deepspeed --deepspeed_config_file config/zero_stage3_config.json --deepspeed_multinode_launcher standard scripts/qwenimage/train.py \
|
||||
--pretrained_model_name_or_path=$MODEL_NAME \
|
||||
--train_data_dir=$DATASET_NAME \
|
||||
--train_data_meta=$DATASET_META_NAME \
|
||||
--train_batch_size=1 \
|
||||
--image_sample_size=1024 \
|
||||
--gradient_accumulation_steps=1 \
|
||||
--dataloader_num_workers=8 \
|
||||
--num_train_epochs=100 \
|
||||
--checkpointing_steps=50 \
|
||||
--learning_rate=2e-05 \
|
||||
--lr_scheduler="constant_with_warmup" \
|
||||
--lr_warmup_steps=100 \
|
||||
--seed=42 \
|
||||
--output_dir="output_dir" \
|
||||
--gradient_checkpointing \
|
||||
--mixed_precision="bf16" \
|
||||
--adam_weight_decay=3e-2 \
|
||||
--adam_epsilon=1e-10 \
|
||||
--vae_mini_batch=1 \
|
||||
--max_grad_norm=0.05 \
|
||||
--enable_bucket \
|
||||
--uniform_sampling \
|
||||
--trainable_modules "."
|
||||
```
|
||||
|
||||
With FSDP:
|
||||
|
||||
```sh
|
||||
export MODEL_NAME="models/Diffusion_Transformer/Qwen-Image"
|
||||
export DATASET_NAME="datasets/internal_datasets/"
|
||||
export DATASET_META_NAME="datasets/internal_datasets/metadata.json"
|
||||
# NCCL_IB_DISABLE=1 and NCCL_P2P_DISABLE=1 are used in multi nodes without RDMA.
|
||||
# export NCCL_IB_DISABLE=1
|
||||
# export NCCL_P2P_DISABLE=1
|
||||
NCCL_DEBUG=INFO
|
||||
|
||||
accelerate launch --mixed_precision="bf16" --use_fsdp --fsdp_auto_wrap_policy TRANSFORMER_BASED_WRAP --fsdp_transformer_layer_cls_to_wrap=QwenImageTransformerBlock --fsdp_sharding_strategy "FULL_SHARD" --fsdp_state_dict_type=SHARDED_STATE_DICT --fsdp_backward_prefetch "BACKWARD_PRE" --fsdp_cpu_ram_efficient_loading False scripts/qwenimage/train.py \
|
||||
--pretrained_model_name_or_path=$MODEL_NAME \
|
||||
--train_data_dir=$DATASET_NAME \
|
||||
--train_data_meta=$DATASET_META_NAME \
|
||||
--train_batch_size=1 \
|
||||
--image_sample_size=1024 \
|
||||
--gradient_accumulation_steps=1 \
|
||||
--dataloader_num_workers=8 \
|
||||
--num_train_epochs=100 \
|
||||
--checkpointing_steps=50 \
|
||||
--learning_rate=2e-05 \
|
||||
--lr_scheduler="constant_with_warmup" \
|
||||
--lr_warmup_steps=100 \
|
||||
--seed=42 \
|
||||
--output_dir="output_dir" \
|
||||
--gradient_checkpointing \
|
||||
--mixed_precision="bf16" \
|
||||
--adam_weight_decay=3e-2 \
|
||||
--adam_epsilon=1e-10 \
|
||||
--vae_mini_batch=1 \
|
||||
--max_grad_norm=0.05 \
|
||||
--enable_bucket \
|
||||
--uniform_sampling \
|
||||
--trainable_modules "."
|
||||
```
|
||||
Executable
+153
@@ -0,0 +1,153 @@
|
||||
## Lora Training Code
|
||||
|
||||
We can choose whether to use deepspeed or fsdp in qwen-image, which can save a lot of video memory.
|
||||
|
||||
Some parameters in the sh file can be confusing, and they are explained in this document:
|
||||
|
||||
- `enable_bucket` is used to enable bucket training. When enabled, the model does not crop the images at the center, but instead, it trains the entire images after grouping them into buckets based on resolution.
|
||||
- `random_hw_adapt` is used to enable automatic height and width scaling for images. When `random_hw_adapt` is enabled, the training images will have their height and width set to `image_sample_size` as the maximum and `512` as the minimum.
|
||||
- For example, when `random_hw_adapt` is enabled, `image_sample_size=1024`, the resolution of image inputs for training is `512x512` to `1024x1024`
|
||||
- `resume_from_checkpoint` is used to set the training should be resumed from a previous checkpoint. Use a path or `"latest"` to automatically select the last available checkpoint.
|
||||
|
||||
Without deepspeed:
|
||||
|
||||
Training qwen-image without DeepSpeed may result in insufficient GPU memory.
|
||||
```sh
|
||||
export MODEL_NAME="models/Diffusion_Transformer/Qwen-Image"
|
||||
export DATASET_NAME="datasets/internal_datasets/"
|
||||
export DATASET_META_NAME="datasets/internal_datasets/metadata.json"
|
||||
# NCCL_IB_DISABLE=1 and NCCL_P2P_DISABLE=1 are used in multi nodes without RDMA.
|
||||
# export NCCL_IB_DISABLE=1
|
||||
# export NCCL_P2P_DISABLE=1
|
||||
NCCL_DEBUG=INFO
|
||||
|
||||
accelerate launch --mixed_precision="bf16" scripts/qwenimage/train_lora.py \
|
||||
--pretrained_model_name_or_path=$MODEL_NAME \
|
||||
--train_data_dir=$DATASET_NAME \
|
||||
--train_data_meta=$DATASET_META_NAME \
|
||||
--train_batch_size=1 \
|
||||
--image_sample_size=1024 \
|
||||
--gradient_accumulation_steps=1 \
|
||||
--dataloader_num_workers=8 \
|
||||
--num_train_epochs=100 \
|
||||
--checkpointing_steps=50 \
|
||||
--learning_rate=1e-04 \
|
||||
--seed=42 \
|
||||
--output_dir="output_dir_lora" \
|
||||
--gradient_checkpointing \
|
||||
--mixed_precision="bf16" \
|
||||
--adam_weight_decay=3e-2 \
|
||||
--adam_epsilon=1e-10 \
|
||||
--vae_mini_batch=1 \
|
||||
--max_grad_norm=0.05 \
|
||||
--enable_bucket \
|
||||
--uniform_sampling
|
||||
```
|
||||
|
||||
With deepspeed zero-2:
|
||||
|
||||
```sh
|
||||
export MODEL_NAME="models/Diffusion_Transformer/Qwen-Image"
|
||||
export DATASET_NAME="datasets/internal_datasets/"
|
||||
export DATASET_META_NAME="datasets/internal_datasets/metadata.json"
|
||||
# NCCL_IB_DISABLE=1 and NCCL_P2P_DISABLE=1 are used in multi nodes without RDMA.
|
||||
# export NCCL_IB_DISABLE=1
|
||||
# export NCCL_P2P_DISABLE=1
|
||||
NCCL_DEBUG=INFO
|
||||
|
||||
accelerate launch --use_deepspeed --deepspeed_config_file config/zero_stage2_config.json --deepspeed_multinode_launcher standard scripts/qwenimage/train_lora.py \
|
||||
--pretrained_model_name_or_path=$MODEL_NAME \
|
||||
--train_data_dir=$DATASET_NAME \
|
||||
--train_data_meta=$DATASET_META_NAME \
|
||||
--train_batch_size=1 \
|
||||
--image_sample_size=1024 \
|
||||
--gradient_accumulation_steps=1 \
|
||||
--dataloader_num_workers=8 \
|
||||
--num_train_epochs=100 \
|
||||
--checkpointing_steps=50 \
|
||||
--learning_rate=1e-04 \
|
||||
--seed=42 \
|
||||
--output_dir="output_dir_lora" \
|
||||
--gradient_checkpointing \
|
||||
--mixed_precision="bf16" \
|
||||
--adam_weight_decay=3e-2 \
|
||||
--adam_epsilon=1e-10 \
|
||||
--vae_mini_batch=1 \
|
||||
--max_grad_norm=0.05 \
|
||||
--enable_bucket \
|
||||
--uniform_sampling
|
||||
```
|
||||
|
||||
Deepspeed zero-3:
|
||||
|
||||
After training, you can use the following command to get the final model:
|
||||
```sh
|
||||
python scripts/zero_to_bf16.py output_dir/checkpoint-{our-num-steps} output_dir/checkpoint-{your-num-steps}-outputs --max_shard_size 80GB --safe_serialization
|
||||
```
|
||||
|
||||
Training shell command is as follows:
|
||||
```sh
|
||||
export MODEL_NAME="models/Diffusion_Transformer/Qwen-Image"
|
||||
export DATASET_NAME="datasets/internal_datasets/"
|
||||
export DATASET_META_NAME="datasets/internal_datasets/metadata.json"
|
||||
# NCCL_IB_DISABLE=1 and NCCL_P2P_DISABLE=1 are used in multi nodes without RDMA.
|
||||
# export NCCL_IB_DISABLE=1
|
||||
# export NCCL_P2P_DISABLE=1
|
||||
NCCL_DEBUG=INFO
|
||||
|
||||
accelerate launch --zero_stage 3 --zero3_save_16bit_model true --zero3_init_flag true --use_deepspeed --deepspeed_config_file config/zero_stage3_config.json --deepspeed_multinode_launcher standard scripts/qwenimage/train_lora.py \
|
||||
--pretrained_model_name_or_path=$MODEL_NAME \
|
||||
--train_data_dir=$DATASET_NAME \
|
||||
--train_data_meta=$DATASET_META_NAME \
|
||||
--train_batch_size=1 \
|
||||
--image_sample_size=1024 \
|
||||
--gradient_accumulation_steps=1 \
|
||||
--dataloader_num_workers=8 \
|
||||
--num_train_epochs=100 \
|
||||
--checkpointing_steps=50 \
|
||||
--learning_rate=1e-04 \
|
||||
--seed=42 \
|
||||
--output_dir="output_dir_lora" \
|
||||
--gradient_checkpointing \
|
||||
--mixed_precision="bf16" \
|
||||
--adam_weight_decay=3e-2 \
|
||||
--adam_epsilon=1e-10 \
|
||||
--vae_mini_batch=1 \
|
||||
--max_grad_norm=0.05 \
|
||||
--enable_bucket \
|
||||
--uniform_sampling
|
||||
```
|
||||
|
||||
With FSDP:
|
||||
|
||||
```sh
|
||||
export MODEL_NAME="models/Diffusion_Transformer/Wan2.2-Fun-A14B-InP"
|
||||
export DATASET_NAME="datasets/internal_datasets/"
|
||||
export DATASET_META_NAME="datasets/internal_datasets/metadata.json"
|
||||
# NCCL_IB_DISABLE=1 and NCCL_P2P_DISABLE=1 are used in multi nodes without RDMA.
|
||||
# export NCCL_IB_DISABLE=1
|
||||
# export NCCL_P2P_DISABLE=1
|
||||
NCCL_DEBUG=INFO
|
||||
|
||||
accelerate launch --mixed_precision="bf16" --use_fsdp --fsdp_auto_wrap_policy TRANSFORMER_BASED_WRAP --fsdp_transformer_layer_cls_to_wrap=QwenImageTransformerBlock --fsdp_sharding_strategy "FULL_SHARD" --fsdp_state_dict_type=SHARDED_STATE_DICT --fsdp_backward_prefetch "BACKWARD_PRE" --fsdp_cpu_ram_efficient_loading False scripts/qwenimage/train_lora.py \
|
||||
--pretrained_model_name_or_path=$MODEL_NAME \
|
||||
--train_data_dir=$DATASET_NAME \
|
||||
--train_data_meta=$DATASET_META_NAME \
|
||||
--train_batch_size=1 \
|
||||
--image_sample_size=1024 \
|
||||
--gradient_accumulation_steps=1 \
|
||||
--dataloader_num_workers=8 \
|
||||
--num_train_epochs=100 \
|
||||
--checkpointing_steps=50 \
|
||||
--learning_rate=1e-04 \
|
||||
--seed=42 \
|
||||
--output_dir="output_dir_lora" \
|
||||
--gradient_checkpointing \
|
||||
--mixed_precision="bf16" \
|
||||
--adam_weight_decay=3e-2 \
|
||||
--adam_epsilon=1e-10 \
|
||||
--vae_mini_batch=1 \
|
||||
--max_grad_norm=0.05 \
|
||||
--enable_bucket \
|
||||
--uniform_sampling
|
||||
```
|
||||
@@ -10,10 +10,10 @@ Some parameters in the sh file can be confusing, and they are explained in this
|
||||
- `random_frame_crop` is used for random cropping on video frames to simulate videos with different frame counts.
|
||||
- `random_hw_adapt` is used to enable automatic height and width scaling for images and videos. When `random_hw_adapt` is enabled, the training images will have their height and width set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum. For training videos, the height and width will be set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=1024`, the resolution of image inputs for training is `512x512` to `1024x1024`, and the resolution of video inputs for training is `512x512x49` to `1024x1024x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- `training_with_video_token_length` specifies training the model according to token length. For training images and videos, the height and width will be set to `image_sample_size` as the maximum and `video_sample_size` as the minimum.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- The token length for a video with dimensions 512x512 and 49 frames is 13,312. We need to set the `token_sample_size = 512`.
|
||||
- At 512x512 resolution, the number of video frames is 49 (~= 512 * 512 * 49 / 512 / 512).
|
||||
- At 768x768 resolution, the number of video frames is 21 (~= 512 * 512 * 49 / 768 / 768).
|
||||
|
||||
@@ -8,10 +8,10 @@ Some parameters in the sh file can be confusing, and they are explained in this
|
||||
- `random_frame_crop` is used for random cropping on video frames to simulate videos with different frame counts.
|
||||
- `random_hw_adapt` is used to enable automatic height and width scaling for images and videos. When `random_hw_adapt` is enabled, the training images will have their height and width set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum. For training videos, the height and width will be set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=1024`, the resolution of image inputs for training is `512x512` to `1024x1024`, and the resolution of video inputs for training is `512x512x49` to `1024x1024x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- `training_with_video_token_length` specifies training the model according to token length. For training images and videos, the height and width will be set to `image_sample_size` as the maximum and `video_sample_size` as the minimum.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- The token length for a video with dimensions 512x512 and 49 frames is 13,312. We need to set the `token_sample_size = 512`.
|
||||
- At 512x512 resolution, the number of video frames is 49 (~= 512 * 512 * 49 / 512 / 512).
|
||||
- At 768x768 resolution, the number of video frames is 21 (~= 512 * 512 * 49 / 768 / 768).
|
||||
|
||||
@@ -8,10 +8,10 @@ Some parameters in the sh file can be confusing, and they are explained in this
|
||||
- `random_frame_crop` is used for random cropping on video frames to simulate videos with different frame counts.
|
||||
- `random_hw_adapt` is used to enable automatic height and width scaling for images and videos. When `random_hw_adapt` is enabled, the training images will have their height and width set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum. For training videos, the height and width will be set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=1024`, the resolution of image inputs for training is `512x512` to `1024x1024`, and the resolution of video inputs for training is `512x512x49` to `1024x1024x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- `training_with_video_token_length` specifies training the model according to token length. For training images and videos, the height and width will be set to `image_sample_size` as the maximum and `video_sample_size` as the minimum.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- The token length for a video with dimensions 512x512 and 49 frames is 13,312. We need to set the `token_sample_size = 512`.
|
||||
- At 512x512 resolution, the number of video frames is 49 (~= 512 * 512 * 49 / 512 / 512).
|
||||
- At 768x768 resolution, the number of video frames is 21 (~= 512 * 512 * 49 / 768 / 768).
|
||||
|
||||
@@ -28,10 +28,10 @@ Some parameters in the sh file can be confusing, and they are explained in this
|
||||
- `random_frame_crop` is used for random cropping on video frames to simulate videos with different frame counts.
|
||||
- `random_hw_adapt` is used to enable automatic height and width scaling for images and videos. When `random_hw_adapt` is enabled, the training images will have their height and width set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum. For training videos, the height and width will be set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=1024`, the resolution of image inputs for training is `512x512` to `1024x1024`, and the resolution of video inputs for training is `512x512x49` to `1024x1024x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- `training_with_video_token_length` specifies training the model according to token length. For training images and videos, the height and width will be set to `image_sample_size` as the maximum and `video_sample_size` as the minimum.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- The token length for a video with dimensions 512x512 and 49 frames is 13,312. We need to set the `token_sample_size = 512`.
|
||||
- At 512x512 resolution, the number of video frames is 49 (~= 512 * 512 * 49 / 512 / 512).
|
||||
- At 768x768 resolution, the number of video frames is 21 (~= 512 * 512 * 49 / 768 / 768).
|
||||
|
||||
@@ -28,10 +28,10 @@ Some parameters in the sh file can be confusing, and they are explained in this
|
||||
- `random_frame_crop` is used for random cropping on video frames to simulate videos with different frame counts.
|
||||
- `random_hw_adapt` is used to enable automatic height and width scaling for images and videos. When `random_hw_adapt` is enabled, the training images will have their height and width set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum. For training videos, the height and width will be set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=1024`, the resolution of image inputs for training is `512x512` to `1024x1024`, and the resolution of video inputs for training is `512x512x49` to `1024x1024x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- `training_with_video_token_length` specifies training the model according to token length. For training images and videos, the height and width will be set to `image_sample_size` as the maximum and `video_sample_size` as the minimum.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- The token length for a video with dimensions 512x512 and 49 frames is 13,312. We need to set the `token_sample_size = 512`.
|
||||
- At 512x512 resolution, the number of video frames is 49 (~= 512 * 512 * 49 / 512 / 512).
|
||||
- At 768x768 resolution, the number of video frames is 21 (~= 512 * 512 * 49 / 768 / 768).
|
||||
|
||||
@@ -8,10 +8,10 @@ Some parameters in the sh file can be confusing, and they are explained in this
|
||||
- `random_frame_crop` is used for random cropping on video frames to simulate videos with different frame counts.
|
||||
- `random_hw_adapt` is used to enable automatic height and width scaling for images and videos. When `random_hw_adapt` is enabled, the training images will have their height and width set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum. For training videos, the height and width will be set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=1024`, the resolution of image inputs for training is `512x512` to `1024x1024`, and the resolution of video inputs for training is `512x512x49` to `1024x1024x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- `training_with_video_token_length` specifies training the model according to token length. For training images and videos, the height and width will be set to `image_sample_size` as the maximum and `video_sample_size` as the minimum.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- The token length for a video with dimensions 512x512 and 49 frames is 13,312. We need to set the `token_sample_size = 512`.
|
||||
- At 512x512 resolution, the number of video frames is 49 (~= 512 * 512 * 49 / 512 / 512).
|
||||
- At 768x768 resolution, the number of video frames is 21 (~= 512 * 512 * 49 / 768 / 768).
|
||||
|
||||
@@ -10,10 +10,10 @@ Some parameters in the sh file can be confusing, and they are explained in this
|
||||
- `random_frame_crop` is used for random cropping on video frames to simulate videos with different frame counts.
|
||||
- `random_hw_adapt` is used to enable automatic height and width scaling for images and videos. When `random_hw_adapt` is enabled, the training images will have their height and width set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum. For training videos, the height and width will be set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=1024`, the resolution of image inputs for training is `512x512` to `1024x1024`, and the resolution of video inputs for training is `512x512x49` to `1024x1024x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- `training_with_video_token_length` specifies training the model according to token length. For training images and videos, the height and width will be set to `image_sample_size` as the maximum and `video_sample_size` as the minimum.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- The token length for a video with dimensions 512x512 and 49 frames is 13,312. We need to set the `token_sample_size = 512`.
|
||||
- At 512x512 resolution, the number of video frames is 49 (~= 512 * 512 * 49 / 512 / 512).
|
||||
- At 768x768 resolution, the number of video frames is 21 (~= 512 * 512 * 49 / 768 / 768).
|
||||
|
||||
@@ -8,10 +8,10 @@ Some parameters in the sh file can be confusing, and they are explained in this
|
||||
- `random_frame_crop` is used for random cropping on video frames to simulate videos with different frame counts.
|
||||
- `random_hw_adapt` is used to enable automatic height and width scaling for images and videos. When `random_hw_adapt` is enabled, the training images will have their height and width set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum. For training videos, the height and width will be set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=1024`, the resolution of image inputs for training is `512x512` to `1024x1024`, and the resolution of video inputs for training is `512x512x49` to `1024x1024x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- `training_with_video_token_length` specifies training the model according to token length. For training images and videos, the height and width will be set to `image_sample_size` as the maximum and `video_sample_size` as the minimum.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- The token length for a video with dimensions 512x512 and 49 frames is 13,312. We need to set the `token_sample_size = 512`.
|
||||
- At 512x512 resolution, the number of video frames is 49 (~= 512 * 512 * 49 / 512 / 512).
|
||||
- At 768x768 resolution, the number of video frames is 21 (~= 512 * 512 * 49 / 768 / 768).
|
||||
|
||||
@@ -8,10 +8,10 @@ Some parameters in the sh file can be confusing, and they are explained in this
|
||||
- `random_frame_crop` is used for random cropping on video frames to simulate videos with different frame counts.
|
||||
- `random_hw_adapt` is used to enable automatic height and width scaling for images and videos. When `random_hw_adapt` is enabled, the training images will have their height and width set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum. For training videos, the height and width will be set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=1024`, the resolution of image inputs for training is `512x512` to `1024x1024`, and the resolution of video inputs for training is `512x512x49` to `1024x1024x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- `training_with_video_token_length` specifies training the model according to token length. For training images and videos, the height and width will be set to `image_sample_size` as the maximum and `video_sample_size` as the minimum.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- The token length for a video with dimensions 512x512 and 49 frames is 13,312. We need to set the `token_sample_size = 512`.
|
||||
- At 512x512 resolution, the number of video frames is 49 (~= 512 * 512 * 49 / 512 / 512).
|
||||
- At 768x768 resolution, the number of video frames is 21 (~= 512 * 512 * 49 / 768 / 768).
|
||||
|
||||
@@ -28,10 +28,10 @@ Some parameters in the sh file can be confusing, and they are explained in this
|
||||
- `random_frame_crop` is used for random cropping on video frames to simulate videos with different frame counts.
|
||||
- `random_hw_adapt` is used to enable automatic height and width scaling for images and videos. When `random_hw_adapt` is enabled, the training images will have their height and width set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum. For training videos, the height and width will be set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=1024`, the resolution of image inputs for training is `512x512` to `1024x1024`, and the resolution of video inputs for training is `512x512x49` to `1024x1024x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- `training_with_video_token_length` specifies training the model according to token length. For training images and videos, the height and width will be set to `image_sample_size` as the maximum and `video_sample_size` as the minimum.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- The token length for a video with dimensions 512x512 and 49 frames is 13,312. We need to set the `token_sample_size = 512`.
|
||||
- At 512x512 resolution, the number of video frames is 49 (~= 512 * 512 * 49 / 512 / 512).
|
||||
- At 768x768 resolution, the number of video frames is 21 (~= 512 * 512 * 49 / 768 / 768).
|
||||
|
||||
@@ -28,10 +28,10 @@ Some parameters in the sh file can be confusing, and they are explained in this
|
||||
- `random_frame_crop` is used for random cropping on video frames to simulate videos with different frame counts.
|
||||
- `random_hw_adapt` is used to enable automatic height and width scaling for images and videos. When `random_hw_adapt` is enabled, the training images will have their height and width set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum. For training videos, the height and width will be set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=1024`, the resolution of image inputs for training is `512x512` to `1024x1024`, and the resolution of video inputs for training is `512x512x49` to `1024x1024x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- `training_with_video_token_length` specifies training the model according to token length. For training images and videos, the height and width will be set to `image_sample_size` as the maximum and `video_sample_size` as the minimum.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- The token length for a video with dimensions 512x512 and 49 frames is 13,312. We need to set the `token_sample_size = 512`.
|
||||
- At 512x512 resolution, the number of video frames is 49 (~= 512 * 512 * 49 / 512 / 512).
|
||||
- At 768x768 resolution, the number of video frames is 21 (~= 512 * 512 * 49 / 768 / 768).
|
||||
|
||||
@@ -8,10 +8,10 @@ Some parameters in the sh file can be confusing, and they are explained in this
|
||||
- `random_frame_crop` is used for random cropping on video frames to simulate videos with different frame counts.
|
||||
- `random_hw_adapt` is used to enable automatic height and width scaling for images and videos. When `random_hw_adapt` is enabled, the training images will have their height and width set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum. For training videos, the height and width will be set to `image_sample_size` as the maximum and `min(video_sample_size, 512)` as the minimum.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=1024`, the resolution of image inputs for training is `512x512` to `1024x1024`, and the resolution of video inputs for training is `512x512x49` to `1024x1024x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- For example, when `random_hw_adapt` is enabled, with `video_sample_n_frames=49`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49`.
|
||||
- `training_with_video_token_length` specifies training the model according to token length. For training images and videos, the height and width will be set to `image_sample_size` as the maximum and `video_sample_size` as the minimum.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=1024`, and `image_sample_size=256`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=1024`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x49`.
|
||||
- For example, when `training_with_video_token_length` is enabled, with `video_sample_n_frames=49`, `token_sample_size=512`, `video_sample_size=256`, and `image_sample_size=1024`, the resolution of image inputs for training is `256x256` to `1024x1024`, and the resolution of video inputs for training is `256x256x49` to `1024x1024x9`.
|
||||
- The token length for a video with dimensions 512x512 and 49 frames is 13,312. We need to set the `token_sample_size = 512`.
|
||||
- At 512x512 resolution, the number of video frames is 49 (~= 512 * 512 * 49 / 512 / 512).
|
||||
- At 768x768 resolution, the number of video frames is 21 (~= 512 * 512 * 49 / 768 / 768).
|
||||
|
||||
@@ -711,21 +711,21 @@ class QwenImageTransformer2DModel(ModelMixin, ConfigMixin, PeftAdapterMixin, Fro
|
||||
|
||||
for index_block, block in enumerate(self.transformer_blocks):
|
||||
if torch.is_grad_enabled() and self.gradient_checkpointing:
|
||||
def create_custom_forward(module):
|
||||
def custom_forward(*inputs):
|
||||
return module(*inputs)
|
||||
def create_custom_forward(module):
|
||||
def custom_forward(*inputs):
|
||||
return module(*inputs)
|
||||
|
||||
return custom_forward
|
||||
ckpt_kwargs: Dict[str, Any] = {"use_reentrant": False} if is_torch_version(">=", "1.11.0") else {}
|
||||
encoder_hidden_states, hidden_states = torch.utils.checkpoint.checkpoint(
|
||||
create_custom_forward(block),
|
||||
hidden_states,
|
||||
encoder_hidden_states,
|
||||
encoder_hidden_states_mask,
|
||||
temb,
|
||||
image_rotary_emb,
|
||||
**ckpt_kwargs,
|
||||
)
|
||||
return custom_forward
|
||||
ckpt_kwargs: Dict[str, Any] = {"use_reentrant": False} if is_torch_version(">=", "1.11.0") else {}
|
||||
encoder_hidden_states, hidden_states = torch.utils.checkpoint.checkpoint(
|
||||
create_custom_forward(block),
|
||||
hidden_states,
|
||||
encoder_hidden_states,
|
||||
encoder_hidden_states_mask,
|
||||
temb,
|
||||
image_rotary_emb,
|
||||
**ckpt_kwargs,
|
||||
)
|
||||
|
||||
else:
|
||||
encoder_hidden_states, hidden_states = block(
|
||||
|
||||
Reference in New Issue
Block a user