Scaleflow - name change

pyproject.toml

@@ -3,7 +3,7 @@ build-backend = "hatchling.build"
 requires = [ "hatch-vcs", "hatchling" ]
 
 [project]
-name = "cellflow-tools"
+name = "scaleflow-tools"
 description = "Modeling complex perturbations with flow matching at single-cell resolution"
 readme = "README.md"
 license = "PolyForm-Noncommercial-1.0.0"
@@ -103,7 +103,7 @@ urls.Home-page = "https://github.com/theislab/cellflow"
 urls.Source = "https://github.com/theislab/cellflow"
 
 [tool.hatch.build.targets.wheel]
-packages = [ 'src/cellflow' ]
+packages = [ 'src/scaleflow' ]
 
 [tool.hatch.version]
 source = "vcs"
@@ -201,7 +201,7 @@ extras = test,pp,external,embedding
 pass_env = PYTEST_*,CI
 commands =
     coverage run -m pytest {tty:--color=yes} {posargs: \
-        --cov={env_site_packages_dir}{/}cellflow --cov-config={tox_root}{/}pyproject.toml \
+        --cov={env_site_packages_dir}{/}scaleflow --cov-config={tox_root}{/}pyproject.toml \
         --no-cov-on-fail --cov-report=xml --cov-report=term-missing:skip-covered}
 
 [testenv:lint-code]

scripts/create_tahoe.py

@@ -0,0 +1,32 @@
+from sc_exp_design.models import CellFlow
+import anndata as ad
+import h5py
+
+from anndata.experimental import read_lazy
+
+print("loading data")
+with h5py.File("/lustre/groups/ml01/workspace/100mil/100m_int_indices.h5ad", "r") as f:
+    adata_all = ad.AnnData(
+        obs=ad.io.read_elem(f["obs"]),
+        var=read_lazy(f["var"]),
+        uns = read_lazy(f["uns"]),
+        obsm = read_lazy(f["obsm"]),
+    )
+cf = CellFlow()
+
+print(" preparing train data ")
+cf.prepare_train_data(adata_all,  
+                      sample_rep="X_pca",
+        control_key="control",
+        perturbation_covariates={"drugs": ("drug",), "dosage": ("dosage",)},
+        perturbation_covariate_reps={"drugs": "drug_embeddings"},
+        sample_covariates=["cell_line"],
+        sample_covariate_reps={"cell_line": "cell_line_embeddings"},
+        split_covariates=["cell_line"])
+
+
+
+
+print("writing zarr")
+cf.train_data.write_zarr(f"/lustre/groups/ml01/workspace/100mil/tahoe_train_data.zarr")
+print("zarr written")