Upload 12 files

captions/m1.txt

@@ -0,0 +1 @@
+A digital illustration of a cartoon character, a panda with a beard, wearing a red and black robe, standing in front of a mountain. The panda is holding out its hand, and the background is a mix of clouds and mountains. The character has a playful and whimsical appearance.

captions/m2.txt

@@ -0,0 +1 @@
+A digital illustration of a cartoon character, a red and white fox, wearing a red robe and holding a staff. The fox has a beard and mustache, and is smiling. The character is standing in front of a tree, and there is a chair in the background. The illustration is in a whimsical style, with a playful and lighthearted tone.

captions/m3.txt

@@ -0,0 +1 @@
+A cartoon illustration of a red and white panda bear wearing a robe and a long beard. The panda is looking angry and has its mouth open. The scene is set in a forest with a tree in the background. The panda is the main focus of the image, with its face taking up a significant portion of the frame. The style of the image is cartoonish and whimsical.

captions/m4.txt

@@ -0,0 +1 @@
+A digital illustration of a cartoon character, a red panda, sitting in a lotus position with his eyes closed, wearing a red robe. The panda is smiling and appears to be meditating. The background features a waterfall and a pagoda. The scene is rendered in a whimsical, playful style.

captions/m5.txt

@@ -0,0 +1 @@
+A cartoon illustration of a panda with a long white beard and mustache, wearing a red robe, sitting in a lotus position and looking angry. The panda is surrounded by a forest and a pagoda in the background. The scene is drawn in a cartoon style with a sense of humor and playfulness.

captions/m6.txt

@@ -0,0 +1 @@
+A digital illustration of a cartoon character, a panda, wearing a red robe and a white beard. The panda is holding a card in its hand and has a serious expression on its face. The character is standing in front of a pagoda, which serves as the background for the scene. The illustration is in a whimsical style, with the panda appearing to be a wise and powerful figure.

captions/m7.txt

@@ -0,0 +1 @@
+A cartoon illustration of a red panda with a white beard and mustache, wearing a red robe, sitting in a lotus position and looking angry. The panda is surrounded by a forest setting with a tree in the background and a lantern hanging from the tree. The illustration has a whimsical and playful style.

captions/m8.txt

@@ -0,0 +1 @@
+A digital illustration of a cartoon character, a red panda, wearing a red robe with gold designs. The panda has a white moustache and is looking angry. The character is standing in front of a statue, with mountains in the background. The panda is wearing a necklace and has a beard. The image is in a comic book style.

captions/m9.txt

@@ -0,0 +1 @@
+A digital illustration of a cartoon character, a panda, wearing a red robe and standing in front of a red building. The panda has a serious expression on its face and is wearing a gold necklace. The background features a mountain and a tree. The panda is the main focus of the image, with its large size and detailed facial expression.

config.yaml

@@ -0,0 +1,60 @@
+job: custom_job
+config:
+  name: flux_train_replicate
+  process:
+  - type: custom_sd_trainer
+    training_folder: output
+    device: cuda:0
+    trigger_word: TOK_MASTER_P
+    network:
+      type: lora
+      linear: 16
+      linear_alpha: 16
+    save:
+      dtype: float16
+      save_every: 1001
+      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: 1
+      steps: 1000
+      gradient_accumulation_steps: 1
+      train_unet: true
+      train_text_encoder: false
+      content_or_style: balanced
+      gradient_checkpointing: false
+      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: false
+    sample:
+      sampler: flowmatch
+      sample_every: 1001
+      width: 1024
+      height: 1024
+      prompts: []
+      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:e00da3d5248bf96b8e721d14bd47784f3a0191ed74557c9db10944c444e297c5
+size 171969432

trained_model.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2134284c79ba516986db8bf267a9e69aba61456dd926222032d834b75ea3383a
+size 171991040