v2 on laionhr 1024

shreyazomato · Jul 14, 2022 · 1754106 · 1754106
1 parent 55bf957
commit 1754106
Show file tree

Hide file tree

Showing 3 changed files with 207 additions and 0 deletions.
diff --git a/configs/stable-diffusion/v2_laionhr1024.yaml b/configs/stable-diffusion/v2_laionhr1024.yaml
@@ -0,0 +1,132 @@
+model:
+  base_learning_rate: 1.0e-04
+  target: ldm.models.diffusion.ddpm.LatentDiffusion
+  params:
+    linear_start: 0.001
+    linear_end: 0.015
+    num_timesteps_cond: 1
+    log_every_t: 200
+    timesteps: 1000
+    first_stage_key: "jpg"
+    cond_stage_key: "txt"
+    image_size: 64
+    channels: 16
+    cond_stage_trainable: false   # Note: different from the one we trained before
+    conditioning_key: crossattn
+    monitor: val/loss_simple_ema
+    scale_factor: 0.22765929   # magic number
+
+    # NOTE disabled for resuming
+    #scheduler_config: # 10000 warmup steps
+    #  target: ldm.lr_scheduler.LambdaLinearScheduler
+    #  params:
+    #    warm_up_steps: [ 10000 ]
+    #    cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases
+    #    f_start: [ 1.e-6 ]
+    #    f_max: [ 1. ]
+    #    f_min: [ 1. ]
+
+    unet_config:
+      target: ldm.modules.diffusionmodules.openaimodel.UNetModel
+      params:
+        image_size: 64    # not really needed
+        in_channels: 16
+        out_channels: 16
+        model_channels: 320
+        attention_resolutions: [ 4, 2, 1 ]
+        num_res_blocks: 2
+        channel_mult: [ 1, 2, 4, 4 ]
+        num_heads: 8
+        use_spatial_transformer: True
+        transformer_depth: 1
+        context_dim: 768
+        use_checkpoint: True
+        legacy: False
+
+    first_stage_config:
+      target: ldm.models.autoencoder.AutoencoderKL
+      params:
+        embed_dim: 16
+        monitor: val/rec_loss
+        ddconfig:
+          double_z: True
+          z_channels: 16
+          resolution: 256
+          in_channels: 3
+          out_ch: 3
+          ch: 128
+          ch_mult: [ 1,1,2,2,4 ]  # num_down = len(ch_mult)-1
+          num_res_blocks: 2
+          attn_resolutions: [ 16 ]
+          dropout: 0.0
+        lossconfig:
+          target: torch.nn.Identity
+
+    cond_stage_config:
+      target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
+
+
+data:
+  target: ldm.data.laion.WebDataModuleFromConfig
+  params:
+    tar_base: "pipe:aws s3 cp s3://s-datasets/laion-high-resolution/"
+    batch_size: 3
+    num_workers: 4
+    multinode: True
+    train:
+      shards: '{00000..17279}.tar -'
+      shuffle: 10000
+      image_key: jpg
+      image_transforms:
+      - target: torchvision.transforms.Resize
+        params:
+          size: 1024
+          interpolation: 3
+      - target: torchvision.transforms.RandomCrop
+        params:
+          size: 1024
+
+    # NOTE use enough shards to avoid empty validation loops in workers
+    validation:
+      shards: '{17280..17535}.tar -'
+      shuffle: 0
+      image_key: jpg
+      image_transforms:
+      - target: torchvision.transforms.Resize
+        params:
+          size: 1024
+          interpolation: 3
+      - target: torchvision.transforms.CenterCrop
+        params:
+          size: 1024
+
+
+lightning:
+  find_unused_parameters: False
+
+  modelcheckpoint:
+    params:
+      every_n_train_steps: 2000
+
+  callbacks:
+    image_logger:
+      target: main.ImageLogger
+      params:
+        batch_frequency: 2000
+        max_images: 2
+        increase_log_steps: False
+        log_first_step: False
+        log_images_kwargs:
+          use_ema_scope: False
+          inpaint: False
+          plot_progressive_rows: False
+          plot_diffusion_rows: False
+          N: 2
+          unconditional_guidance_scale: 5.0
+          unconditional_guidance_label: [""]
+
+  trainer:
+    benchmark: True
+    val_check_interval: 5000000
+    num_sanity_val_steps: 0
+    accumulate_grad_batches: 4
diff --git a/scripts/slurm/v2_laionhr1024/launcher.sh b/scripts/slurm/v2_laionhr1024/launcher.sh
@@ -0,0 +1,33 @@
+#!/bin/bash
+
+# mpi version for node rank
+H=`hostname`
+THEID=`echo -e $HOSTNAMES  | python3 -c "import sys;[sys.stdout.write(str(i)) for i,line in enumerate(next(sys.stdin).split(' ')) if line.strip() == '$H'.strip()]"`
+export NODE_RANK=${THEID}
+echo THEID=$THEID
+
+echo "##########################################"
+echo MASTER_ADDR=${MASTER_ADDR}
+echo MASTER_PORT=${MASTER_PORT}
+echo NODE_RANK=${NODE_RANK}
+echo WORLD_SIZE=${WORLD_SIZE}
+echo "##########################################"
+# debug environment worked great so we stick with it
+# no magic there, just a miniconda python=3.9, pytorch=1.12, cudatoolkit=11.3
+# env with pip dependencies from stable diffusion's requirements.txt
+eval "$(/fsx/stable-diffusion/debug/miniconda3/bin/conda shell.bash hook)"
+conda activate stable
+cd /fsx/stable-diffusion/stable-diffusion
+
+CONFIG="/fsx/stable-diffusion/stable-diffusion/configs/stable-diffusion/v2_laionhr1024.yaml"
+
+# resume and set new seed to reshuffle data
+EXTRA="--seed 714 model.params.ckpt_path=/fsx/stable-diffusion/stable-diffusion/logs/2022-07-12T00-50-44_txt2img-multinode-clip-encoder-f16-1024-laion-hr/checkpoints/last.ckpt"
+
+# custom logdir
+#EXTRA="${EXTRA} --logdir rlogs"
+
+# debugging
+#EXTRA="${EXTRA} -d True lightning.callbacks.image_logger.params.batch_frequency=50"
+
+python main.py --base $CONFIG --gpus 0,1,2,3,4,5,6,7 -t --num_nodes ${WORLD_SIZE} --scale_lr False $EXTRA
diff --git a/scripts/slurm/v2_laionhr1024/sbatch.sh b/scripts/slurm/v2_laionhr1024/sbatch.sh
@@ -0,0 +1,42 @@
+#!/bin/bash
+#SBATCH --partition=compute-od-gpu
+#SBATCH --job-name=stable-diffusion-v2-laionhr1024
+#SBATCH --nodes 20
+#SBATCH --ntasks-per-node 1
+#SBATCH --cpus-per-gpu=4
+#SBATCH --gres=gpu:8
+#SBATCH --exclusive
+#SBATCH --output=%x_%j.out
+#SBATCH --comment "Key=Monitoring,Value=ON"
+
+module load intelmpi
+source /opt/intel/mpi/latest/env/vars.sh
+export LD_LIBRARY_PATH=/opt/aws-ofi-nccl/lib:/opt/amazon/efa/lib64:/usr/local/cuda-11.0/efa/lib:/usr/local/cuda-11.0/lib:/usr/local/cuda-11.0/lib64:/usr/local/cuda-11.0:/opt/nccl/build/lib:/opt/aws-ofi-nccl-inst
+all/lib:/opt/aws-ofi-nccl/lib:$LD_LIBRARY_PATH
+export NCCL_PROTO=simple
+export PATH=/opt/amazon/efa/bin:$PATH
+export LD_PRELOAD="/opt/nccl/build/lib/libnccl.so"
+export FI_EFA_FORK_SAFE=1
+export FI_LOG_LEVEL=1
+export FI_EFA_USE_DEVICE_RDMA=1 # use for p4dn
+export NCCL_DEBUG=info
+export PYTHONFAULTHANDLER=1
+export CUDA_LAUNCH_BLOCKING=0
+export OMPI_MCA_mtl_base_verbose=1
+export FI_EFA_ENABLE_SHM_TRANSFER=0
+export FI_PROVIDER=efa
+export FI_EFA_TX_MIN_CREDITS=64
+export NCCL_TREE_THRESHOLD=0
+
+# sent to sub script
+export HOSTNAMES=`scontrol show hostnames "$SLURM_JOB_NODELIST"`
+export MASTER_ADDR=$(scontrol show hostnames "$SLURM_JOB_NODELIST" | head -n 1)
+export MASTER_PORT=12802
+export COUNT_NODE=`scontrol show hostnames "$SLURM_JOB_NODELIST" | wc -l`
+export WORLD_SIZE=$COUNT_NODE
+
+echo go $COUNT_NODE
+echo $HOSTNAMES
+echo $WORLD_SIZE
+
+mpirun -n $COUNT_NODE -perhost 1 /fsx/stable-diffusion/stable-diffusion/scripts/slurm/v2_laionhr1024/launcher.sh