Add files using upload-large-folder tool

README.md

@@ -12,24 +12,10 @@ pipeline_tag: image-to-image
 
 # BiliSakura/HSIGene
 
-**Hyperspectral image generation** — HSIGene converted to diffusers format. Conditional generation with local controls (HED, MLSD, sketch, segmentation), global controls (content, text), and metadata embeddings. Outputs 48-band hyperspectral images (256×256 pixels).
+**Hyperspectral image generation** — HSIGene converted to diffusers format. Supports task-specific conditioning with local controls (HED, MLSD, sketch, segmentation), global controls (content or text), or metadata embeddings. Outputs 48-band hyperspectral images (256×256 pixels).
 
 > Source: [HSIGene](https://arxiv.org/abs/2409.12470). Converted to diffusers format; model dir is self-contained (no external project for inference).
 
-## Conversion
-
-The main diffusion checkpoint (`last.ckpt`) must be downloaded from [GoogleDrive](https://drive.google.com/file/d/1euJAbsxCgG1wIu_Eh5nPfmiSP9suWsR4/view?usp=drive_link) and placed in `projects/HSIGene-Diffusers/checkpoints/`.
-
-**Note:** `models/raw/HSIGene` contains annotator/auxiliary models (body pose, depth, SAM, etc.) only — not the main diffusion checkpoint.
-
-```bash
-cd projects/HSIGene-Diffusers
-python convert_to_diffusers.py \
-    --config_path configs/inference.yaml \
-    --ckpt_path checkpoints/last.ckpt \
-    --output_dir /root/worksapce/models/BiliSakura/HSIGene
-```
-
 ## Repository Structure (after conversion)
 
 | Component               | Path                     |
@@ -47,65 +33,59 @@ python convert_to_diffusers.py \
 
 ## Usage
 
-**Option 1 – No `sys.path.insert` (AeroGen-style):** Load the pipeline from the model path via `importlib`; the model dir is added to the path automatically.
+**Inference Demo (`DiffusionPipeline.from_pretrained`)**
 
 ```python
-import importlib.util
-import sys
-
-model_path = "/path/to/HSIGene"  # or "BiliSakura/HSIGene" for Hub
-spec = importlib.util.spec_from_file_location("pipeline_hsigene", f"{model_path}/pipeline_hsigene.py")
-mod = importlib.util.module_from_spec(spec)
-sys.modules["pipeline_hsigene"] = mod
-spec.loader.exec_module(mod)
-
-pipe = mod.HSIGenePipeline.from_pretrained(model_path)
+from diffusers import DiffusionPipeline
+pipe = DiffusionPipeline.from_pretrained(
+  "/path/to/BiliSakura/HSIGene",
+  trust_remote_code=True,
+  custom_pipeline="path/to/pipeline_hsigene.py",
+  model_path="path/to/BiliSakura/HSIGene"
+)
 pipe = pipe.to("cuda")
 ```
 
-**Option 2 – With `sys.path.insert`:** Simpler if you are fine adding the model dir to the path once.
+**Dependencies:** `pip install diffusers transformers torch einops safetensors`
 
-```python
-import sys
-sys.path.insert(0, "/path/to/HSIGene")
-from pipeline_hsigene import HSIGenePipeline
+### Per-Condition Inference Demos (Not Combined)
 
-pipe = HSIGenePipeline.from_pretrained("/path/to/HSIGene")
-pipe = pipe.to("cuda")
+`local_conditions` shape: `(B, 18, H, W)`; `global_conditions` shape: `(B, 768)`; `metadata` shape: `(7,)` or `(B, 7)`.
+
+```python
+# HED condition
+output = pipe(prompt="", local_conditions=hed_local, global_conditions=None, metadata=None)
 ```
 
-**Option 3 – `DiffusionPipeline.from_pretrained`:** May work with `trust_remote_code=True`. If you see "raw config (list)" errors (e.g. when loading from cache), use Option 1 or 2 instead.
+```python
+# MLSD condition
+output = pipe(prompt="", local_conditions=mlsd_local, global_conditions=None, metadata=None)
+```
 
 ```python
-from diffusers import DiffusionPipeline
-pipe = DiffusionPipeline.from_pretrained("/path/to/HSIGene", trust_remote_code=True)
-pipe = pipe.to("cuda")
+# Sketch condition
+output = pipe(prompt="", local_conditions=sketch_local, global_conditions=None, metadata=None)
 ```
 
-**Dependencies:** `pip install diffusers transformers torch einops safetensors`
+```python
+# Segmentation condition
+output = pipe(prompt="", local_conditions=seg_local, global_conditions=None, metadata=None)
+```
 
 ```python
-# Conditional generation
-output = pipe(
-    prompt="Wasteland",
-    num_samples=1,
-    height=256,
-    width=256,
-    num_inference_steps=50,
-    local_conditions=local_tensor,   # (B, 18, H, W) or None
-    global_conditions=global_tensor, # (B, 768) or None
-    metadata=metadata_tensor,        # (7,) or (B, 7) or None
-    guidance_scale=1.0,
-)
-images = output.images  # (B, H, W, 48) in [0, 1]
+# Content condition (global)
+output = pipe(prompt="", local_conditions=None, global_conditions=content_global, metadata=None)
 ```
 
-### Conditioning
+```python
+# Text condition
+output = pipe(prompt="Wasteland", local_conditions=None, global_conditions=None, metadata=None)
+```
 
-- **Local**: 18-channel maps (HED, MLSD, sketch, segmentation, etc.) at 512×512 default.
-- **Global**: 768-dim CLIP features from reference images.
-- **Metadata**: 7-dim vector.
-- **Text**: Via `prompt`; use `text_strength` to scale.
+```python
+# Metadata condition
+output = pipe(prompt="", local_conditions=None, global_conditions=None, metadata=metadata_vec)
+```
 
 ## Model Sources
 
@@ -116,12 +96,20 @@ images = output.images  # (B, H, W, 48) in [0, 1]
 ## Citation
 
 ```bibtex
-@misc{pang2024hsigenefoundationmodelhyperspectral,
-  title={HSIGene: A Foundation Model For Hyperspectral Image Generation},
-  author={Li Pang and Datao Tang and Shuang Xu and Deyu Meng and Xiangyong Cao},
-  year={2024},
-  eprint={2409.12470},
-  archivePrefix={arXiv},
-  primaryClass={cs.CV},
+@article{pangHSIGeneFoundationModel2026,
+  title = {{{HSIGene}}: {{A Foundation Model}} for {{Hyperspectral Image Generation}}},
+  shorttitle = {{{HSIGene}}},
+  author = {Pang, Li and Cao, Xiangyong and Tang, Datao and Xu, Shuang and Bai, Xueru and Zhou, Feng and Meng, Deyu},
+  year = 2026,
+  month = jan,
+  journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
+  volume = {48},
+  number = {1},
+  pages = {730--746},
+  issn = {1939-3539},
+  doi = {10.1109/TPAMI.2025.3610927},
+  urldate = {2026-01-02},
+  keywords = {Adaptation models,Computational modeling,Controllable generation,deep learning,diffusion model,Diffusion models,Foundation models,hyperspectral image synthesis,Hyperspectral imaging,Image synthesis,Noise reduction,Reliability,Superresolution,Training}
 }
+
 ```

model_index.json

@@ -1,5 +1,5 @@
 {
-  "_class_name": ["pipeline_hsigene", "HSIGenePipeline"],
+  "_class_name": "HSIGenePipeline",
   "_diffusers_version": "0.25.0",
   "scheduler": ["diffusers", "DDIMScheduler"],
   "unet": ["pipeline_hsigene", "HSIGenePipeline"],

pipeline_hsigene.py

@@ -212,6 +212,31 @@ def _is_component_list(v):
     return isinstance(v, (list, tuple)) and len(v) == 2 and isinstance(v[0], str) and isinstance(v[1], str)
 
 
+def _resolve_model_root(candidate: Optional[Union[str, Path]]) -> Optional[Path]:
+    """Resolve candidate path/repo to model root containing model_index.json."""
+    if not candidate:
+        return None
+    try:
+        path = Path(candidate)
+        if not path.exists():
+            from huggingface_hub import snapshot_download
+            path = Path(snapshot_download(str(candidate)))
+        path = path.resolve()
+        if (path / "model_index.json").exists():
+            return path
+        cur = path
+        for _ in range(5):
+            parent = cur.parent
+            if parent == cur:
+                break
+            if (parent / "model_index.json").exists():
+                return parent
+            cur = parent
+    except Exception:
+        return None
+    return None
+
+
 class HSIGenePipeline(DiffusionPipeline):
     """Pipeline for HSIGene hyperspectral image generation.
 
@@ -245,18 +270,51 @@ class HSIGenePipeline(DiffusionPipeline):
         scheduler=None,
         crs_model=None,
         scale_factor=0.18215,
+        model_path: Optional[Union[str, Path]] = None,
+        _name_or_path: Optional[Union[str, Path]] = None,
     ):
         super().__init__()
         if crs_model is not None:
             self.register_modules(crs_model=crs_model, scheduler=scheduler)
         else:
-            if any(_is_component_list(x) for x in (unet, vae, text_encoder, local_adapter,
-                    global_content_adapter, global_text_adapter, metadata_encoder) if x is not None):
-                raise ValueError(
-                    "HSIGene received raw config (list) instead of loaded components. "
-                    "Use HSIGenePipeline.from_pretrained(path) directly, or ensure the model "
-                    "directory (with hsigene package) is on the path when loading."
+            components_are_lists = any(
+                _is_component_list(x)
+                for x in (
+                    unet,
+                    vae,
+                    text_encoder,
+                    local_adapter,
+                    global_content_adapter,
+                    global_text_adapter,
+                    metadata_encoder,
+                )
+                if x is not None
+            )
+            if components_are_lists:
+                # Diffusers custom_pipeline may pass raw [library, class] placeholders to __init__.
+                # Resolve model root and materialize real components here.
+                model_root = (
+                    _resolve_model_root(model_path)
+                    or _resolve_model_root(_name_or_path)
+                    or _resolve_model_root(getattr(getattr(self, "config", None), "_name_or_path", None))
                 )
+                if model_root is None:
+                    raise ValueError(
+                        "HSIGene received raw config placeholders but could not resolve model path. "
+                        "Pass `model_path` to HSIGenePipeline or load via "
+                        "`DiffusionPipeline.from_pretrained(<path>, custom_pipeline=<pipeline_file>)` "
+                        "with a valid local model directory."
+                    )
+                loaded = load_components(model_root)
+                unet = loaded["unet"]
+                vae = loaded["vae"]
+                text_encoder = loaded["text_encoder"]
+                local_adapter = loaded["local_adapter"]
+                global_content_adapter = loaded["global_content_adapter"]
+                global_text_adapter = loaded["global_text_adapter"]
+                metadata_encoder = loaded["metadata_encoder"]
+                scheduler = loaded["scheduler"] if scheduler is None else scheduler
+                scale_factor = loaded["scale_factor"]
             crs_model = _CRSModelWrapper(
                 unet=unet,
                 vae=vae,