Light ConceptsTraining data is a collection of various light concepts I enjoy using that are not overly represented in large datasets, trained as a single lora.<span style="background-color:rgb(26, 27, 30);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">ℹ️ LoRA work best when applied to the base models on which they are trained. Please read the About This Version <span style="background-color:rgb(26, 27, 30);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">on the appropriate base models and workflow/training information.I've trained many of these concepts before, generally they are nice to use to enhance the lighting of generations or give interesting effects.<span style="background-color:rgb(26, 27, 30);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">Trained on a large mixed NL and tags dataset, at mixed [1024, 1536] resolutions. Previews are mostly generated at 1024x1536 with a combination of tags and NL prompts.Concept tagsNot limited to, but collected by works containing:<pre><code>
dispersion
hue shifting
refraction
subsurface scattering
translucent

bioluminescence
caustics
dappled moonlight
glowing hot
ultraviolet light</code></pre>Works best in combination with NL if you name a character, describe their basic appearance, and finish with descriptions of light sources and their effects on the scene:<pre><code>
A vibrant and dynamic illustration of Hoshimachi Suisei from Hololive, featuring her squatting in front of a glowing triangular prism.
A beam of white light enters the prism from the left and refracts into a vibrant rainbow on the right. The background is a solid dark grey to emphasize the lighting effects.</code></pre>

Light Concepts

Trained on Anima Base 1Dataset with a mix of natural language and tag captions, unchanged from v3Partitioned dataset and trained at multi-res 512, 768, 1024, 1280, 1536Training config:<pre><code># trained using diffusion-pipe commit b0aa4f1e03169f3280c8518d37570a448420f8be
# NCCL_P2P_DISABLE="1" NCCL_IB_DISABLE="1" NCCL_CUMEM_ENABLE="0" deepspeed --num_gpus=1 train.py --deepspeed --config anima-lora.toml --i_know_what_i_am_doing

output_dir = '/mnt/d/anima/training_output/anima-base-1-light-v31'

dataset = 'dataset-anima-light.toml'

# training settings
epochs = 2
# Per-resolution batch sizes
micro_batch_size_per_gpu = [[512, 64], [768, 64], [1024, 32], [1280, 24], [1536, 16]]
pipeline_stages = 1
gradient_accumulation_steps = 1
gradient_clipping = 1
warmup_steps = 30
lr_scheduler = 'cosine'

# misc settings
save_every_n_epochs = 1
activation_checkpointing = true

partition_method = 'parameters'

save_dtype = 'bfloat16'
caching_batch_size = 1
map_num_proc = 8
steps_per_print = 1
compile = true

[model]
type = 'anima'
transformer_path = '/mnt/c/workspace/models/diffusion_models/anima-base-v1.0.safetensors'
vae_path = '/mnt/c/workspace/models/vae/qwen_image_vae.safetensors'
llm_path = '/mnt/c/workspace/models/text_encoders/qwen_3_06b_base.safetensors'
dtype = 'bfloat16'
#cache_text_embeddings = false
llm_adapter_lr = 8e-7
#timestep_sample_method = 'uniform'
flux_shift = true
multiscale_loss_weight = 0.5
sigmoid_scale = 1.3

[adapter]
type = 'lora'
rank = 32
dtype = 'bfloat16'

[optimizer]
type = 'adamw_optimi'
lr = 4e-5
betas = [0.9, 0.99]
weight_decay = 0.01
eps = 1e-8</code></pre><pre><code>resolutions = [512, 768, 1024, 1280, 1536]

enable_ar_bucket = true
min_ar = 0.5
max_ar = 2.0
num_ar_buckets = 9

# 16 repeats from captions.json
# 1,321 steps 2.02 hours per epoch
# 866 total images

# images_light\1536x1536\captions.json with 270 entries.
[[directory]]
path = '/mnt/d/training_data/images_light/1536x1536'
resolutions = [512, 1024, 1280, 1536]

# images_light\1280x1280\captions.json with 59 entries.
[[directory]]
path = '/mnt/d/training_data/images_light/1280x1280'
resolutions = [512, 1024, 1280]

# images_light\1024x1024\captions.json with 368 entries.
[[directory]]
path = '/mnt/d/training_data/images_light/1024x1024'
resolutions = [512, 768, 1024]

# images_light\768x768\captions.json with 162 entries.
[[directory]]
path = '/mnt/d/training_data/images_light/768x768'
resolutions = [512, 768]

# images_light\512x512\captions.json with 7 entries.
[[directory]]
path = '/mnt/d/training_data/images_light/512x512'
resolutions = [512]
</code></pre>

Files with SHA256: 98afa7627640814006b5b1cc1c2ee024deba411e2810bbb9620f076d6282db41