Upload folder using huggingface_hub

README.md

@@ -0,0 +1,46 @@
+---
+license: other
+license_name: flux-1-dev-non-commercial-license
+license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md
+language:
+- en
+tags:
+- flux
+- diffusers
+- lora
+- replicate
+base_model: "black-forest-labs/FLUX.1-dev"
+pipeline_tag: text-to-image
+# widget:
+#   - text: >-
+#       prompt
+#     output:
+#       url: https://...
+instance_prompt: hsr_lora
+---
+
+# Ruan Mei Topaz
+
+<!-- <Gallery /> -->
+
+Trained on Replicate using:
+
+https://replicate.com/ostris/flux-dev-lora-trainer/train
+
+
+## Trigger words
+You should use `hsr_lora` to trigger the image generation.
+
+
+## Use it with the [🧨 diffusers library](https://github.com/huggingface/diffusers)
+
+```py
+from diffusers import AutoPipelineForText2Image
+import torch
+
+pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.float16).to('cuda')
+pipeline.load_lora_weights('hanungaddi/ruan-mei-topaz', weight_name='lora.safetensors')
+image = pipeline('your prompt').images[0]
+```
+
+For more details, including weighting, merging and fusing LoRAs, check the [documentation on loading LoRAs in diffusers](https://huggingface.co/docs/diffusers/main/en/using-diffusers/loading_adapters)

config.yaml

@@ -0,0 +1,82 @@
+job: custom_job
+config:
+  name: flux_train_replicate
+  process:
+  - type: custom_sd_trainer
+    training_folder: output
+    device: cuda:0
+    trigger_word: hsr_lora
+    network:
+      type: lora
+      linear: 32
+      linear_alpha: 32
+    save:
+      dtype: float16
+      save_every: 200
+      max_step_saves_to_keep: 1
+    datasets:
+    - folder_path: input_images
+      caption_ext: txt
+      caption_dropout_rate: 0.05
+      shuffle_tokens: false
+      cache_latents_to_disk: false
+      cache_latents: true
+      resolution:
+      - 512
+      - 768
+      - 1024
+    train:
+      batch_size: 4
+      steps: 2000
+      gradient_accumulation_steps: 1
+      train_unet: true
+      train_text_encoder: false
+      content_or_style: balanced
+      gradient_checkpointing: true
+      noise_scheduler: flowmatch
+      optimizer: adamw8bit
+      lr: 0.0004
+      ema_config:
+        use_ema: true
+        ema_decay: 0.99
+      dtype: bf16
+    model:
+      name_or_path: FLUX.1-dev
+      is_flux: true
+      quantize: true
+    sample:
+      sampler: flowmatch
+      sample_every: 100
+      width: 1024
+      height: 1024
+      prompts:
+      - This is a digital anime-style drawing featuring the character topaz. Topaz
+        has a slender build, fair skin, and striking blue eyes. She has short, white
+        hair with a bold red streak on the left side, and a small, silver hairpin
+        on the right side. Her outfit is a mix of futuristic and fantasy elements,
+        featuring a red and white military-inspired jacket with gold accents, black
+        gloves, and a white, high-collared shirt underneath. The jacket is open at
+        the chest, revealing a hint of cleavage. Topaz is posing with her right hand
+        on her chin, and her left hand resting on her head, giving a confident and
+        slightly playful expression. The background is plain white, emphasizing the
+        character. There are color swatches on the left side of the image, indicating
+        the character's color palette.
+      - This is a digital anime-style drawing featuring Ruan Mei, a young woman with
+        long, flowing black hair and striking green eyes. She has a fair skin tone
+        and a petite, curvy figure with large breasts. Ruan Mei is dressed in an elaborate
+        fantasy outfit, primarily teal and brown with intricate gold accents, featuring
+        a high collar, puffed sleeves, and a flowing cape. Her attire includes a decorative
+        white necklace and a small, pink flower on her right side. The background
+        is plain white, keeping the focus on Ruan Mei's detailed and elegant appearance.
+      - digital anime-style illustration draped in a (shawl),dressed in blue dress,short
+        sleeves,shiny texture,watch,ring,standing upright,hand in hair,slight twist,close-up
+        shot,eye-level angle,natural lighting,fashion photography,violet smirking
+        honey brown  Watercolor art
+      neg: ''
+      seed: 42
+      walk_seed: true
+      guidance_scale: 3.5
+      sample_steps: 28
+meta:
+  name: flux_train_replicate
+  version: '1.0'

lora.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7c92c3a974641a52244d0019ab93ab65933bed26f94d7469efe728c0d431f1c
+size 343805464