add a new model

examples/flux2/model_training/full/Template-KleinBase4B-ContentRef.sh

@@ -0,0 +1,19 @@
+modelscope download --dataset DiffSynth-Studio/diffsynth_example_dataset --include "flux2/Template-KleinBase4B-ContentRef/*" --local_dir ./data/diffsynth_example_dataset
+
+accelerate launch examples/flux2/model_training/train.py \
+  --dataset_base_path data/diffsynth_example_dataset/flux2/Template-KleinBase4B-ContentRef \
+  --dataset_metadata_path data/diffsynth_example_dataset/flux2/Template-KleinBase4B-ContentRef/metadata.jsonl \
+  --extra_inputs "template_inputs" \
+  --max_pixels 1048576 \
+  --dataset_repeat 50 \
+  --model_id_with_origin_paths "black-forest-labs/FLUX.2-klein-4B:text_encoder/*.safetensors,black-forest-labs/FLUX.2-klein-base-4B:transformer/*.safetensors,black-forest-labs/FLUX.2-klein-4B:vae/diffusion_pytorch_model.safetensors" \
+  --template_model_id_or_path "DiffSynth-Studio/Template-KleinBase4B-ContentRef:" \
+  --tokenizer_path "black-forest-labs/FLUX.2-klein-4B:tokenizer/" \
+  --learning_rate 1e-4 \
+  --num_epochs 2 \
+  --remove_prefix_in_ckpt "pipe.template_model." \
+  --output_path "./models/train/Template-KleinBase4B-ContentRef_full" \
+  --trainable_models "template_model" \
+  --use_gradient_checkpointing \
+  --find_unused_parameters \
+  --enable_lora_hot_loading

examples/flux2/model_training/train.py

@@ -29,7 +29,7 @@ class Flux2ImageTrainingModule(DiffusionTrainingModule):
         tokenizer_config = self.parse_path_or_model_id(tokenizer_path, default_value=ModelConfig(model_id="black-forest-labs/FLUX.2-dev", origin_file_pattern="tokenizer/"))
         self.pipe = Flux2ImagePipeline.from_pretrained(torch_dtype=torch.bfloat16, device=device, model_configs=model_configs, tokenizer_config=tokenizer_config)
         self.pipe = self.load_training_template_model(self.pipe, template_model_id_or_path, args.use_gradient_checkpointing, args.use_gradient_checkpointing_offload)
-        self.pipe = self.split_pipeline_units(task, self.pipe, trainable_models, lora_base_model)
+        self.pipe = self.split_pipeline_units(task, self.pipe, trainable_models, lora_base_model, remove_unnecessary_params=True)
         if enable_lora_hot_loading: self.pipe.dit = self.pipe.enable_lora_hot_loading(self.pipe.dit)
 
         # Training mode

examples/flux2/model_training/validate_full/Template-KleinBase4B-ContentRef.py

@@ -0,0 +1,55 @@
+from diffsynth.diffusion.template import TemplatePipeline
+from diffsynth.pipelines.flux2_image import Flux2ImagePipeline, ModelConfig
+from diffsynth.core import load_state_dict
+import torch
+from modelscope import dataset_snapshot_download
+from PIL import Image
+import numpy as np
+
+pipe = Flux2ImagePipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="black-forest-labs/FLUX.2-klein-base-4B", origin_file_pattern="transformer/*.safetensors"),
+        ModelConfig(model_id="black-forest-labs/FLUX.2-klein-4B", origin_file_pattern="text_encoder/*.safetensors"),
+        ModelConfig(model_id="black-forest-labs/FLUX.2-klein-4B", origin_file_pattern="vae/diffusion_pytorch_model.safetensors"),
+    ],
+    tokenizer_config=ModelConfig(model_id="black-forest-labs/FLUX.2-klein-4B", origin_file_pattern="tokenizer/"),
+)
+pipe.dit = pipe.enable_lora_hot_loading(pipe.dit) # Important!
+template = TemplatePipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[ModelConfig(model_id="DiffSynth-Studio/Template-KleinBase4B-ContentRef")],
+)
+state_dict = load_state_dict("./models/train/Template-KleinBase4B-ContentRef_full/epoch-1.safetensors", torch_dtype=torch.bfloat16)
+template.models[0].load_state_dict(state_dict)
+dataset_snapshot_download(
+    "DiffSynth-Studio/examples_in_diffsynth",
+    allow_file_pattern=["templates/*"],
+    local_dir="data/examples",
+)
+image = template(
+    pipe,
+    prompt="A cat is sitting on a stone.",
+    seed=0, cfg_scale=4, num_inference_steps=50,
+    template_inputs = [{
+        "image": Image.open("data/examples/templates/image_style_1.jpg"),
+    }],
+    negative_template_inputs = [{
+        "image": Image.fromarray(np.zeros((1024, 1024, 3), dtype=np.uint8) + 128),
+    }],
+)
+image.save("image_ContentRef_1.jpg")
+image = template(
+    pipe,
+    prompt="A cat is sitting on a stone.",
+    seed=0, cfg_scale=4, num_inference_steps=50,
+    template_inputs = [{
+        "image": Image.open("data/examples/templates/image_style_2.jpg"),
+    }],
+    negative_template_inputs = [{
+        "image": Image.fromarray(np.zeros((1024, 1024, 3), dtype=np.uint8) + 128),
+    }],
+)
+image.save("image_ContentRef_2.jpg")