support auto parallel in dit and largedit

ppdiffusers/examples/class_conditional_image_generation/DiT/0_run_train_dit_trainer_auto.sh

@@ -0,0 +1,97 @@
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
+# 
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+# 
+#     http://www.apache.org/licenses/LICENSE-2.0
+# 
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+export FLAGS_use_cuda_managed_memory=true
+export FLAGS_enable_pir_api=true
+
+TRAINING_MODEL_RESUME="None"
+TRAINER_INSTANCES='127.0.0.1'
+MASTER='127.0.0.1:8080'
+TRAINERS_NUM=1 # nnodes, machine num
+TRAINING_GPUS_PER_NODE=8 # nproc_per_node
+DP_DEGREE=1 # data_parallel_degree
+MP_DEGREE=1 # tensor_parallel_degree
+PP_DEGREE=1 # pipeline_parallel_degree
+SHARDING_DEGREE=8 # sharding_parallel_degree
+
+# real dp_parallel_degree = nnodes * nproc_per_node / tensor_parallel_degree / sharding_parallel_degree
+# Please make sure: nnodes * nproc_per_node >= tensor_parallel_degree * sharding_parallel_degree
+
+config_file=config/DiT_XL_patch2.json
+OUTPUT_DIR=./output_trainer/DiT_XL_patch2_auto_trainer
+feature_path=./data/fastdit_imagenet256
+
+per_device_train_batch_size=32
+gradient_accumulation_steps=1
+
+resolution=256
+num_workers=8
+max_steps=7000000
+logging_steps=1
+save_steps=5000
+image_logging_steps=-1
+seed=0
+
+max_grad_norm=-1
+
+USE_AMP=True
+FP16_OPT_LEVEL="O2"
+
+enable_tensorboard=True
+recompute=True
+enable_xformers=True
+to_static=0  # whether we use dynamic to static training
+
+TRAINING_PYTHON="python -m paddle.distributed.launch --master ${MASTER} --nnodes ${TRAINERS_NUM} --nproc_per_node ${TRAINING_GPUS_PER_NODE} --ips ${TRAINER_INSTANCES}"
+${TRAINING_PYTHON} train_image_generation_trainer_auto.py \
+    --do_train \
+    --feature_path ${feature_path} \
+    --output_dir ${OUTPUT_DIR} \
+    --per_device_train_batch_size ${per_device_train_batch_size} \
+    --gradient_accumulation_steps ${gradient_accumulation_steps} \
+    --learning_rate 1e-4 \
+    --weight_decay 0.0 \
+    --resolution ${resolution} \
+    --max_steps ${max_steps} \
+    --lr_scheduler_type "constant" \
+    --warmup_steps 0 \
+    --image_logging_steps ${image_logging_steps} \
+    --logging_dir ${OUTPUT_DIR}/tb_log \
+    --logging_steps ${logging_steps} \
+    --save_steps ${save_steps} \
+    --save_total_limit 50 \
+    --dataloader_num_workers ${num_workers} \
+    --vae_name_or_path stabilityai/sd-vae-ft-mse \
+    --config_file ${config_file} \
+    --num_inference_steps 25 \
+    --use_ema True \
+    --max_grad_norm ${max_grad_norm} \
+    --overwrite_output_dir True \
+    --disable_tqdm True \
+    --fp16_opt_level ${FP16_OPT_LEVEL} \
+    --seed ${seed} \
+    --recompute ${recompute} \
+    --enable_xformers_memory_efficient_attention ${enable_xformers} \
+    --bf16 ${USE_AMP} \
+    --amp_master_grad 1 \
+    --dp_degree ${DP_DEGREE} \
+    --tensor_parallel_degree ${MP_DEGREE} \
+    --pipeline_parallel_degree ${PP_DEGREE} \
+    --sharding_parallel_degree ${SHARDING_DEGREE} \
+    --sharding "stage1" \
+    --sharding_parallel_config "enable_stage1_overlap enable_stage1_tensor_fusion" \
+    --hybrid_parallel_topo_order "sharding_first" \
+    --sep_parallel_degree 1 \
+    --enable_auto_parallel 1 \
+    --to_static $to_static \
+    # --fp16 ${USE_AMP} \

ppdiffusers/examples/class_conditional_image_generation/DiT/4_run_train_largedit_3b_trainer_auto.sh

@@ -0,0 +1,97 @@
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
+# 
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+# 
+#     http://www.apache.org/licenses/LICENSE-2.0
+# 
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+export FLAGS_use_cuda_managed_memory=true
+export FLAGS_enable_pir_api=true
+
+TRAINING_MODEL_RESUME="None"
+TRAINER_INSTANCES='127.0.0.1'
+MASTER='127.0.0.1:8080'
+TRAINERS_NUM=1 # nnodes, machine num
+TRAINING_GPUS_PER_NODE=8 # nproc_per_node
+DP_DEGREE=1 # data_parallel_degree
+MP_DEGREE=4 # tensor_parallel_degree
+PP_DEGREE=1 # pipeline_parallel_degree
+SHARDING_DEGREE=2 # sharding_parallel_degree
+
+# real dp_parallel_degree = nnodes * nproc_per_node / tensor_parallel_degree / sharding_parallel_degree
+# Please make sure: nnodes * nproc_per_node >= tensor_parallel_degree * sharding_parallel_degree
+
+config_file=config/LargeDiT_3B_patch2.json
+OUTPUT_DIR=./output_trainer/LargeDiT_3B_patch2_auto_trainer
+feature_path=./data/fastdit_imagenet256
+
+per_device_train_batch_size=32
+gradient_accumulation_steps=1
+
+resolution=256
+num_workers=8
+max_steps=7000000
+logging_steps=1
+save_steps=5000
+image_logging_steps=-1
+seed=0
+
+max_grad_norm=2.0
+
+USE_AMP=True
+FP16_OPT_LEVEL="O2"
+
+enable_tensorboard=True
+recompute=True
+enable_xformers=True
+to_static=0  # whether we use dynamic to static training
+
+TRAINING_PYTHON="python -m paddle.distributed.launch --master ${MASTER} --nnodes ${TRAINERS_NUM} --nproc_per_node ${TRAINING_GPUS_PER_NODE} --ips ${TRAINER_INSTANCES}"
+${TRAINING_PYTHON} train_image_generation_trainer_auto.py \
+    --do_train \
+    --feature_path ${feature_path} \
+    --output_dir ${OUTPUT_DIR} \
+    --per_device_train_batch_size ${per_device_train_batch_size} \
+    --gradient_accumulation_steps ${gradient_accumulation_steps} \
+    --learning_rate 1e-4 \
+    --resolution ${resolution} \
+    --weight_decay 0.0 \
+    --max_steps ${max_steps} \
+    --lr_scheduler_type "constant" \
+    --warmup_steps 0 \
+    --image_logging_steps ${image_logging_steps} \
+    --logging_dir ${OUTPUT_DIR}/tb_log \
+    --logging_steps ${logging_steps} \
+    --save_steps ${save_steps} \
+    --save_total_limit 50 \
+    --dataloader_num_workers ${num_workers} \
+    --vae_name_or_path stabilityai/sd-vae-ft-mse \
+    --config_file ${config_file} \
+    --num_inference_steps 25 \
+    --use_ema True \
+    --max_grad_norm ${max_grad_norm} \
+    --overwrite_output_dir True \
+    --disable_tqdm True \
+    --fp16_opt_level ${FP16_OPT_LEVEL} \
+    --seed ${seed} \
+    --recompute ${recompute} \
+    --enable_xformers_memory_efficient_attention ${enable_xformers} \
+    --bf16 ${USE_AMP} \
+    --amp_master_grad 1 \
+    --dp_degree ${DP_DEGREE} \
+    --tensor_parallel_degree ${MP_DEGREE} \
+    --pipeline_parallel_degree ${PP_DEGREE} \
+    --sharding_parallel_degree ${SHARDING_DEGREE} \
+    --sharding "stage1" \
+    --sharding_parallel_config "enable_stage1_overlap enable_stage1_tensor_fusion" \
+    --hybrid_parallel_topo_order "sharding_first" \
+    --sep_parallel_degree 1 \
+    --enable_auto_parallel 1 \
+    --to_static $to_static \
+    # --fp16 ${USE_AMP} \

ppdiffusers/examples/class_conditional_image_generation/DiT/README.md

@@ -161,6 +161,186 @@ python -u -m paddle.distributed.launch --gpus "0,1,2,3,4,5,6,7" \
     --ckpt_every ${save_steps} \
 ```
 
+### 1.5 使用分布式自动并行进行训练
+
+#### 1.5.1 硬件要求
+同1.3.1，与之前手动并行方式显存占用差别不大
+
+#### 1.5.2 单机多卡训练Dit
+可以直接运行`sh 0_run_train_dit_trainer_auto.sh`，或者
+
+```bash
+FLAGS_enable_pir_api=true
+
+TRAINING_MODEL_RESUME="None"
+TRAINER_INSTANCES='127.0.0.1'
+MASTER='127.0.0.1:8080'
+TRAINERS_NUM=1 # nnodes, machine num
+TRAINING_GPUS_PER_NODE=8 # nproc_per_node
+DP_DEGREE=1 # data_parallel_degree
+MP_DEGREE=1 # tensor_parallel_degree
+PP_DEGREE=1 # pipeline_parallel_degree
+SHARDING_DEGREE=8 # sharding_parallel_degree
+
+# real dp_parallel_degree = nnodes * nproc_per_node / tensor_parallel_degree / sharding_parallel_degree
+# Please make sure: nnodes * nproc_per_node >= tensor_parallel_degree * sharding_parallel_degree
+
+config_file=config/DiT_XL_patch2.json
+OUTPUT_DIR=./output_trainer/DiT_XL_patch2_auto_trainer
+feature_path=./data/fastdit_imagenet256
+
+per_device_train_batch_size=32
+gradient_accumulation_steps=1
+
+resolution=256
+num_workers=8
+max_steps=7000000
+logging_steps=1
+save_steps=5000
+image_logging_steps=-1
+seed=0
+
+max_grad_norm=-1
+
+USE_AMP=True
+FP16_OPT_LEVEL="O2"
+
+enable_tensorboard=True
+recompute=True
+enable_xformers=True
+to_static=0  # whether we use dynamic to static training
+
+TRAINING_PYTHON="python -m paddle.distributed.launch --master ${MASTER} --nnodes ${TRAINERS_NUM} --nproc_per_node ${TRAINING_GPUS_PER_NODE} --ips ${TRAINER_INSTANCES}"
+${TRAINING_PYTHON} train_image_generation_trainer_auto.py \
+    --do_train \
+    --feature_path ${feature_path} \
+    --output_dir ${OUTPUT_DIR} \
+    --per_device_train_batch_size ${per_device_train_batch_size} \
+    --gradient_accumulation_steps ${gradient_accumulation_steps} \
+    --learning_rate 1e-4 \
+    --weight_decay 0.0 \
+    --resolution ${resolution} \
+    --max_steps ${max_steps} \
+    --lr_scheduler_type "constant" \
+    --warmup_steps 0 \
+    --image_logging_steps ${image_logging_steps} \
+    --logging_dir ${OUTPUT_DIR}/tb_log \
+    --logging_steps ${logging_steps} \
+    --save_steps ${save_steps} \
+    --save_total_limit 50 \
+    --dataloader_num_workers ${num_workers} \
+    --vae_name_or_path stabilityai/sd-vae-ft-mse \
+    --config_file ${config_file} \
+    --num_inference_steps 25 \
+    --use_ema True \
+    --max_grad_norm ${max_grad_norm} \
+    --overwrite_output_dir True \
+    --disable_tqdm True \
+    --fp16_opt_level ${FP16_OPT_LEVEL} \
+    --seed ${seed} \
+    --recompute ${recompute} \
+    --enable_xformers_memory_efficient_attention ${enable_xformers} \
+    --bf16 ${USE_AMP} \
+    --amp_master_grad 1 \
+    --dp_degree ${DP_DEGREE} \
+    --tensor_parallel_degree ${MP_DEGREE} \
+    --pipeline_parallel_degree ${PP_DEGREE} \
+    --sharding_parallel_degree ${SHARDING_DEGREE} \
+    --sharding "stage1" \
+    --sharding_parallel_config "enable_stage1_overlap enable_stage1_tensor_fusion" \
+    --hybrid_parallel_topo_order "sharding_first" \
+    --sep_parallel_degree 1 \
+    --enable_auto_parallel 1 \
+    --to_static $to_static
+```
+
+#### 1.5.3 单机多卡训练LargeDit
+可以直接运行`sh 4_run_train_largedit_3b_trainer_auto.sh`，或者
+
+```bash
+FLAGS_enable_pir_api=true
+
+TRAINING_MODEL_RESUME="None"
+TRAINER_INSTANCES='127.0.0.1'
+MASTER='127.0.0.1:8080'
+TRAINERS_NUM=1 # nnodes, machine num
+TRAINING_GPUS_PER_NODE=8 # nproc_per_node
+DP_DEGREE=1 # data_parallel_degree
+MP_DEGREE=4 # tensor_parallel_degree
+PP_DEGREE=1 # pipeline_parallel_degree
+SHARDING_DEGREE=2 # sharding_parallel_degree
+
+# real dp_parallel_degree = nnodes * nproc_per_node / tensor_parallel_degree / sharding_parallel_degree
+# Please make sure: nnodes * nproc_per_node >= tensor_parallel_degree * sharding_parallel_degree
+
+config_file=config/LargeDiT_3B_patch2.json
+OUTPUT_DIR=./output_trainer/LargeDiT_3B_patch2_auto_trainer
+feature_path=./data/fastdit_imagenet256
+
+per_device_train_batch_size=32
+gradient_accumulation_steps=1
+
+resolution=256
+num_workers=8
+max_steps=7000000
+logging_steps=1
+save_steps=5000
+image_logging_steps=-1
+seed=0
+
+max_grad_norm=2.0
+
+USE_AMP=True
+FP16_OPT_LEVEL="O2"
+
+enable_tensorboard=True
+recompute=True
+enable_xformers=True
+to_static=0  # whether we use dynamic to static training
+
+TRAINING_PYTHON="python -m paddle.distributed.launch --master ${MASTER} --nnodes ${TRAINERS_NUM} --nproc_per_node ${TRAINING_GPUS_PER_NODE} --ips ${TRAINER_INSTANCES}"
+${TRAINING_PYTHON} train_image_generation_trainer_auto.py \
+    --do_train \
+    --feature_path ${feature_path} \
+    --output_dir ${OUTPUT_DIR} \
+    --per_device_train_batch_size ${per_device_train_batch_size} \
+    --gradient_accumulation_steps ${gradient_accumulation_steps} \
+    --learning_rate 1e-4 \
+    --resolution ${resolution} \
+    --weight_decay 0.0 \
+    --max_steps ${max_steps} \
+    --lr_scheduler_type "constant" \
+    --warmup_steps 0 \
+    --image_logging_steps ${image_logging_steps} \
+    --logging_dir ${OUTPUT_DIR}/tb_log \
+    --logging_steps ${logging_steps} \
+    --save_steps ${save_steps} \
+    --save_total_limit 50 \
+    --dataloader_num_workers ${num_workers} \
+    --vae_name_or_path stabilityai/sd-vae-ft-mse \
+    --config_file ${config_file} \
+    --num_inference_steps 25 \
+    --use_ema True \
+    --max_grad_norm ${max_grad_norm} \
+    --overwrite_output_dir True \
+    --disable_tqdm True \
+    --fp16_opt_level ${FP16_OPT_LEVEL} \
+    --seed ${seed} \
+    --recompute ${recompute} \
+    --enable_xformers_memory_efficient_attention ${enable_xformers} \
+    --bf16 ${USE_AMP} \
+    --amp_master_grad 1 \
+    --dp_degree ${DP_DEGREE} \
+    --tensor_parallel_degree ${MP_DEGREE} \
+    --pipeline_parallel_degree ${PP_DEGREE} \
+    --sharding_parallel_degree ${SHARDING_DEGREE} \
+    --sharding "stage1" \
+    --sharding_parallel_config "enable_stage1_overlap enable_stage1_tensor_fusion" \
+    --hybrid_parallel_topo_order "sharding_first" \
+    --sep_parallel_degree 1 \
+    --enable_auto_parallel 1 \
+    --to_static $to_static
+```
 
 ## 2 模型推理
 

ppdiffusers/examples/class_conditional_image_generation/DiT/config/LargeDiT_3B_patch2.json

@@ -4,5 +4,6 @@
   "in_channels": 4,
   "num_layers": 32,
   "num_attention_heads": 32,
-  "attention_head_dim": 96
+  "attention_head_dim": 96,
+  "qk_norm": false
 }