Merge pull request #1358 from flatironinstitute/dev

dev->main for release 1.11.1

.github/workflows/run_tests.yml

@@ -19,15 +19,16 @@ jobs:
       fail-fast: false
 
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
 
       - name: Set up Conda
-        uses: conda-incubator/setup-miniconda@v2
+        uses: conda-incubator/setup-miniconda@v3
         with:
           auto-update-conda: true
           python-version: ${{ matrix.python-version }}
           environment-file: environment-minimal.yml
           activate-environment: caiman
+          conda-solver: libmamba
 
       - name: Install OS Dependencies
         shell: bash -l {0}

caiman/caimanmanager.py

@@ -3,6 +3,7 @@
 import argparse
 import filecmp
 import glob
+import json
 import os
 import platform
 import psutil
@@ -52,6 +53,20 @@
 
 def do_install_to(targdir: str, inplace: bool = False, force: bool = False) -> None:
     global sourcedir_base
+    cwd = None # Assigning so it exists to avoid UnboundLocalError
+
+    try:
+        import importlib
+        import importlib_metadata
+        # A lot can change upstream with this code; I hope the APIs are stable, but just in case, make this best-effort
+        if json.loads(importlib_metadata.Distribution.from_name('caiman').read_text('direct_url.json'))['dir_info']['editable']:
+            inplace = True
+            cwd = os.getcwd()
+            os.chdir(str(importlib.resources.files('caiman').joinpath('..')))
+            print(f"Used editable fallback, entered {os.getcwd()} directory")
+    except:
+        print("Did not use editable fallback")
+
     ignore_pycache=shutil.ignore_patterns('__pycache__')
     if os.path.isdir(targdir) and not force:
         raise Exception(targdir + " already exists. You may move it out of the way, remove it, or use --force")
@@ -77,6 +92,8 @@ def do_install_to(targdir: str, inplace: bool = False, force: bool = False) -> N
         with open(os.path.join(targdir, 'RELEASE'), 'w') as verfile_fh:
             print(f"Version:{caiman.__version__}", file=verfile_fh)
     print("Installed " + targdir)
+    if cwd is not None:
+        os.chdir(cwd)
 
 
 def do_check_install(targdir: str, inplace: bool = False) -> None:

caiman/paths.py

@@ -38,7 +38,7 @@ def get_tempdir() -> str:
         if os.path.isdir(os.environ['CAIMAN_TEMP']):
             return os.environ['CAIMAN_TEMP']
         else:
-            logging.warning(f"CAIMAN_TEMP is set to nonexistent directory {os.environment['CAIMAN_TEMP']}. Ignoring")
+            logging.warning(f"CAIMAN_TEMP is set to nonexistent directory {os.environ['CAIMAN_TEMP']}. Ignoring")
     temp_under_data = os.path.join(caiman_datadir(), "temp")
     if not os.path.isdir(temp_under_data):
         logging.warning(f"Default temporary dir {temp_under_data} does not exist, creating")

caiman/source_extraction/cnmf/spatial.py

@@ -504,7 +504,7 @@ def threshold_components(A, dims, medw=None, thr_method='max', maxthr=0.1, nrgth
             res = dview.map_async(
                 threshold_components_parallel, pars).get(4294967)
         else:
-            res = dview.map_async(threshold_components_parallel, pars)
+            res = dview.map_sync(threshold_components_parallel, pars)
     else:
         res = list(map(threshold_components_parallel, pars))
 

caiman/utils/utils.py

@@ -561,25 +561,16 @@ def recursively_load_dict_contents_from_group(h5file:h5py.File, path:str) -> dic
 
     for key, item in h5file[path].items():
         if isinstance(item, h5py._hl.dataset.Dataset):
-            val_set = np.nan
-            if isinstance(item[()], str):
-                if item[()] == 'NoneType':
-                    ans[key] = None
-                else:
-                    ans[key] = item[()]
-
-            elif key in ['dims', 'medw', 'sigma_smooth_snmf', 'dxy', 'max_shifts', 'strides', 'overlaps']:
-                if isinstance(item[()], np.ndarray):
-                    ans[key] = tuple(item[()])
-                else:
-                    ans[key] = item[()]
+            val = item[()]
+            if isinstance(val, str) and val == 'NoneType' or isinstance(val, bytes) and val == b'NoneType':
+                ans[key] = None
+            elif key in ['dims', 'medw', 'sigma_smooth_snmf',
+                         'dxy', 'max_shifts', 'strides', 'overlaps'] and isinstance(val, np.ndarray):
+                    ans[key] = tuple(val)
+            elif isinstance(val, np.bool_): # sigh
+                ans[key] = bool(val)
             else:
-                if isinstance(item[()], np.bool_): # sigh
-                    ans[key] = bool(item[()])
-                else:
-                    ans[key] = item[()]
-                    if isinstance(ans[key], bytes) and ans[key] == b'NoneType':
-                        ans[key] = None
+                ans[key] = item[()]
 
         elif isinstance(item, h5py._hl.group.Group):
             if key in ('A', 'W', 'Ab', 'downscale_matrix', 'upscale_matrix'):

environment-minimal.yml

@@ -23,7 +23,7 @@ dependencies:
 - scikit-image >=0.19.0
 - scikit-learn >=1.2
 - scipy >= 1.10.1
-- tensorflow >=2.4.0
+- tensorflow >=2.4.0,<2.16
 - tifffile
 - tqdm
 - zarr

environment.yml

@@ -30,7 +30,7 @@ dependencies:
 - scikit-image >=0.19.0
 - scikit-learn >=1.2
 - scipy >= 1.10.1
-- tensorflow >=2.4.0
+- tensorflow >=2.4.0,<2.16
 - tifffile
 - tk
 - tqdm