VictorMylle/Kohya-ss-sd-scripts
1
0
Fork 1
mirror of https://github.com/kohya-ss/sd-scripts.git synced 2026-04-18 09:30:28 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix: Replace CDC integer index lookup with image_key strings

Browse Source 
Fixes shape mismatch bug in multi-subset training where CDC preprocessing
and training used different index calculations, causing wrong CDC data to
be loaded for samples.

Changes:
- CDC cache now stores/loads data using image_key strings instead of integer indices
- Training passes image_key list instead of computed integer indices
- All CDC lookups use stable image_key identifiers
- Improved device compatibility check (handles "cuda" vs "cuda:0")
- Updated all 30 CDC tests to use image_key-based access

Root cause: Preprocessing used cumulative dataset indices while training
used sorted keys, resulting in mismatched lookups during shuffled multi-subset
training.
This commit is contained in:
rockerBOO
2025-10-09 17:15:07 -04:00
parent 4bea582601
commit 1d4c4d4cb2
9 changed files with 129 additions and 115 deletions
Download Patch File Download Diff File Expand all files Collapse all files

32
tests/library/test_cdc_eigenvalue_scaling.py
View File

@@ -30,7 +30,9 @@ class TestEigenvalueScaling:
latent[:, h, w] = (h * 8 + w) / 32.0 # Range [0, 2.0]
# Add per-sample variation
latent = latent + i * 0.1
preprocessor.add_latent(latent=latent, global_idx=i, shape=latent.shape)
metadata = {'image_key': f'test_image_{i}'}
preprocessor.add_latent(latent=latent, global_idx=i, shape=latent.shape, metadata=metadata)
output_path = tmp_path / "test_gamma_b.safetensors"
result_path = preprocessor.compute_all(save_path=output_path)
@@ -39,7 +41,7 @@ class TestEigenvalueScaling:
with safe_open(str(result_path), framework="pt", device="cpu") as f:
all_eigvals = []
for i in range(10):
eigvals = f.get_tensor(f"eigenvalues/{i}").numpy()
eigvals = f.get_tensor(f"eigenvalues/test_image_{i}").numpy()
all_eigvals.extend(eigvals)
all_eigvals = np.array(all_eigvals)
@@ -74,7 +76,9 @@ class TestEigenvalueScaling:
for h in range(4):
for w in range(4):
latent[c, h, w] = (c + h * 4 + w) / 32.0 + i * 0.2
preprocessor.add_latent(latent=latent, global_idx=i, shape=latent.shape)
metadata = {'image_key': f'test_image_{i}'}
preprocessor.add_latent(latent=latent, global_idx=i, shape=latent.shape, metadata=metadata)
output_path = tmp_path / "test_gamma_b.safetensors"
result_path = preprocessor.compute_all(save_path=output_path)
@@ -82,7 +86,7 @@ class TestEigenvalueScaling:
with safe_open(str(result_path), framework="pt", device="cpu") as f:
all_eigvals = []
for i in range(10):
eigvals = f.get_tensor(f"eigenvalues/{i}").numpy()
eigvals = f.get_tensor(f"eigenvalues/test_image_{i}").numpy()
all_eigvals.extend(eigvals)
all_eigvals = np.array(all_eigvals)
@@ -113,15 +117,17 @@ class TestEigenvalueScaling:
for w in range(8):
latent[:, h, w] = (h * 8 + w) / 16.0 # Range [0, 4.0]
latent = latent + i * 0.3
preprocessor.add_latent(latent=latent, global_idx=i, shape=latent.shape)
metadata = {'image_key': f'test_image_{i}'}
preprocessor.add_latent(latent=latent, global_idx=i, shape=latent.shape, metadata=metadata)
output_path = tmp_path / "test_gamma_b.safetensors"
result_path = preprocessor.compute_all(save_path=output_path)
with safe_open(str(result_path), framework="pt", device="cpu") as f:
# Check dtype is fp16
eigvecs = f.get_tensor("eigenvectors/0")
eigvals = f.get_tensor("eigenvalues/0")
eigvecs = f.get_tensor("eigenvectors/test_image_0")
eigvals = f.get_tensor("eigenvalues/test_image_0")
assert eigvecs.dtype == torch.float16, f"Expected fp16, got {eigvecs.dtype}"
assert eigvals.dtype == torch.float16, f"Expected fp16, got {eigvals.dtype}"
@@ -154,7 +160,9 @@ class TestEigenvalueScaling:
for w in range(4):
latent[c, h, w] = (c * 0.1 + h * 0.2 + w * 0.3) + i * 0.5
original_latents.append(latent.clone())
preprocessor.add_latent(latent=latent, global_idx=i, shape=latent.shape)
metadata = {'image_key': f'test_image_{i}'}
preprocessor.add_latent(latent=latent, global_idx=i, shape=latent.shape, metadata=metadata)
# Compute original latent statistics
orig_std = torch.stack(original_latents).std().item()
@@ -194,7 +202,9 @@ class TestTrainingLossScale:
for h in range(4):
for w in range(4):
latent[c, h, w] = (c + h + w) / 20.0 + i * 0.1
preprocessor.add_latent(latent=latent, global_idx=i, shape=latent.shape)
metadata = {'image_key': f'test_image_{i}'}
preprocessor.add_latent(latent=latent, global_idx=i, shape=latent.shape, metadata=metadata)
output_path = tmp_path / "test_gamma_b.safetensors"
cdc_path = preprocessor.compute_all(save_path=output_path)
@@ -211,9 +221,9 @@ class TestTrainingLossScale:
for w in range(4):
latents[b, c, h, w] = (b + c + h + w) / 24.0
t = torch.tensor([0.5, 0.7, 0.9]) # Different timesteps
indices = [0, 5, 9]
image_keys = ['test_image_0', 'test_image_5', 'test_image_9']
eigvecs, eigvals = gamma_b.get_gamma_b_sqrt(indices)
eigvecs, eigvals = gamma_b.get_gamma_b_sqrt(image_keys)
noise = gamma_b.compute_sigma_t_x(eigvecs, eigvals, latents, t)
# Check noise magnitude