Merge pull request #931 from ComputationalCryoEM/workaround_patches

Workaround patches

Author: garrettwrong

.github/workflows/workflow.yml

@@ -64,14 +64,16 @@ jobs:
         shell: bash -el {0}
     strategy:
       matrix:
-        os: [ubuntu-latest, ubuntu-20.04, macOS-latest, macOS-11, windows-2019]
+        os: [ubuntu-latest, ubuntu-20.04, macOS-latest, macOS-11]
         backend: [default, openblas]
         python-version: ['3.8']
         include:
           - os: ubuntu-latest
             backend: intel
           - os: macOS-latest
             backend: accelerate
+          - os: windows-2019
+            backend: default
 
     steps:
     - uses: actions/checkout@v3

gallery/experiments/experimental_abinitio_pipeline_10028.py

@@ -98,6 +98,9 @@
 # logger.info("Invert the global density contrast")
 # src = src.invert_contrast()
 
+# Caching is used for speeding up large datasets on high memory machines.
+src = src.cache()
+
 # %%
 # Optional: CWF Denoising
 # -----------------------
@@ -119,6 +122,8 @@
     cwf_denoiser = DenoiserCov2D(src)
     # Use denoised src for classification
     classification_src = cwf_denoiser.denoise()
+    # Cache for speedup.  Avoids recomputing.
+    classification_src = classification_src.cache()
     # Peek, what do the denoised images look like...
     if interactive:
         classification_src.images[:10].show()

gallery/experiments/experimental_abinitio_pipeline_10081.py

@@ -75,6 +75,9 @@
 aiso_noise_estimator = AnisotropicNoiseEstimator(src)
 src.whiten(aiso_noise_estimator.filter)
 
+# Caching is used for speeding up large datasets on high memory machines.
+src = src.cache()
+
 # %%
 # Class Averaging
 # ----------------------

src/aspire/source/coordinates.py

@@ -366,7 +366,7 @@ def _extract_ctf(self, data_block):
 
         # get unique ctfs from the data block
         # i'th entry of `indices` contains the index of `filter_params` with corresponding CTF params
-        ctf_data = np.stack(data_block[c] for c in CTF_params).astype(self.dtype).T
+        ctf_data = np.stack([data_block[c] for c in CTF_params]).astype(self.dtype).T
         filter_params, indices = np.unique(
             ctf_data,
             return_inverse=True,

src/aspire/source/relion.py

@@ -107,7 +107,7 @@ def __init__(
         # If these all exist in the STAR file, we may create CTF filters for the source
         if set(CTF_params).issubset(metadata.keys()):
             # partition particles according to unique CTF parameters
-            ctf_data = np.stack(metadata[k] for k in CTF_params).T
+            ctf_data = np.stack([metadata[k] for k in CTF_params]).T
             filter_params, filter_indices = np.unique(
                 ctf_data,
                 return_inverse=True,