---
# This configuration requires 48GB of VRAM or more to operate
job: extension
config:
  # this name will be the folder and filename name
  name: "my_first_flex_finetune_v1"
  process:
    - type: 'sd_trainer'
      # root folder to save training sessions/samples/weights
      training_folder: "output"
      # uncomment to see performance stats in the terminal every N steps
      # performance_log_every: 1000
      device: cuda:0
      # if a trigger word is specified, it will be added to captions of training data if it does not already exist
      # alternatively, in your captions you can add [trigger] and it will be replaced with the trigger word
      # trigger_word: "p3r5on"
      save:
        dtype: bf16 # precision to save
        save_every: 250 # save every this many steps
        max_step_saves_to_keep: 2 # how many intermittent saves to keep
        save_format: 'diffusers' # 'diffusers'
      datasets:
        # datasets are a folder of images. captions need to be txt files with the same name as the image
        # for instance image2.jpg and image2.txt. Only jpg, jpeg, and png are supported currently
        # images will automatically be resized and bucketed into the resolution specified
        # on windows, escape back slashes with another backslash so
        # "C:\\path\\to\\images\\folder"
        - folder_path: "/path/to/images/folder"
          caption_ext: "txt"
          caption_dropout_rate: 0.05  # will drop out the caption 5% of time
          shuffle_tokens: false  # shuffle caption order, split by commas
          # cache_latents_to_disk: true  # leave this true unless you know what you're doing
          resolution: [ 512, 768, 1024 ]  # flex enjoys multiple resolutions
      train:
        batch_size: 1
        # IMPORTANT! For Flex, you must bypass the guidance embedder during training
        bypass_guidance_embedding: true

        # can be 'sigmoid', 'linear', or 'lognorm_blend'
        timestep_type: 'sigmoid' 
        
        steps: 2000  # total number of steps to train 500 - 4000 is a good range
        gradient_accumulation: 1
        train_unet: true
        train_text_encoder: false  # probably won't work with flex
        gradient_checkpointing: true  # need the on unless you have a ton of vram
        noise_scheduler: "flowmatch" # for training only
        optimizer: "adafactor"
        lr: 3e-5

        # Paramiter swapping can reduce vram requirements. Set factor from 1.0 to 0.0. 
        # 0.1 is 10% of paramiters active at easc step. Only works with adafactor

        # do_paramiter_swapping: true
        # paramiter_swapping_factor: 0.9

        # uncomment this to skip the pre training sample
        # skip_first_sample: true
        # uncomment to completely disable sampling
        # disable_sampling: true

        # ema will smooth out learning, but could slow it down. Recommended to leave on if you have the vram
        ema_config:
          use_ema: true
          ema_decay: 0.99

        # will probably need this if gpu supports it for flex, other dtypes may not work correctly
        dtype: bf16
      model:
        # huggingface model name or path
        name_or_path: "ostris/Flex.1-alpha"
        is_flux: true # flex is flux architecture
        # full finetuning quantized models is a crapshoot and results in subpar outputs
        # quantize: true
        # you can quantize just the T5 text encoder here to save vram
        quantize_te: true
        # only train the transformer blocks
        only_if_contains: 
          - "transformer.transformer_blocks."
          - "transformer.single_transformer_blocks."
      sample:
        sampler: "flowmatch" # must match train.noise_scheduler
        sample_every: 250 # sample every this many steps
        width: 1024
        height: 1024
        prompts:
          # you can add [trigger] to the prompts here and it will be replaced with the trigger word
          # - "[trigger] holding a sign that says 'I LOVE PROMPTS!'"\
          - "woman with red hair, playing chess at the park, bomb going off in the background"
          - "a woman holding a coffee cup, in a beanie, sitting at a cafe"
          - "a horse is a DJ at a night club, fish eye lens, smoke machine, lazer lights, holding a martini"
          - "a man showing off his cool new t shirt at the beach, a shark is jumping out of the water in the background"
          - "a bear building a log cabin in the snow covered mountains"
          - "woman playing the guitar, on stage, singing a song, laser lights, punk rocker"
          - "hipster man with a beard, building a chair, in a wood shop"
          - "photo of a man, white background, medium shot, modeling clothing, studio lighting, white backdrop"
          - "a man holding a sign that says, 'this is a sign'"
          - "a bulldog, in a post apocalyptic world, with a shotgun, in a leather jacket, in a desert, with a motorcycle"
        neg: ""  # not used on flex
        seed: 42
        walk_seed: true
        guidance_scale: 4
        sample_steps: 25
# you can add any additional meta info here. [name] is replaced with config name at top
meta:
  name: "[name]"
  version: '1.0'