few mods; bring in Izaskun's eb changes

Author: markrobinsonuzh

Clustering_conda.yml

@@ -94,13 +94,20 @@ stages:
         parameters:
           - method: "genie"
             gini_threshold: 0.5
+      - id: genieclust2
+        name: "genieclust"
+        software_environment: "clustbench"
+        repository:
+          url: https://github.com/imallona/clustbench_genieclust
+          commit: 7d9e799
+        parameters:
           - method: ["gic", "ica"]
       - id: fcps
         name: "fcps"
         software_environment: "fcps"
         repository:
-          url: https://github.com/imallona/clustbench_fcps
-          commit: e780fed
+          url: https://github.com/markrobinsonuzh/clustbench_fcps.git
+          commit: c19094f
         parameters:
           - method: ["FCPS_Minimax", "FCPS_MinEnergy", "FCPS_HDBSCAN_2", "FCPS_HDBSCAN_4", "FCPS_HDBSCAN_8", "FCPS_Diana", "FCPS_Fanny", "FCPS_Hardcl", "FCPS_Softcl", "FCPS_Clara", "FCPS_PAM"]
             seed: 2

Clustering_envmodules.yml

@@ -94,6 +94,13 @@ stages:
         parameters:
           - method: "genie"
             gini_threshold: 0.5
+      - id: genieclust2
+        name: "genieclust"
+        software_environment: "clustbench"
+        repository:
+          url: https://github.com/imallona/clustbench_genieclust
+          commit: 7d9e799
+        parameters:
           - method: ["gic", "ica"]
       - id: fcps
         name: "fcps"

Clustering_oras.yml

@@ -94,6 +94,13 @@ stages:
         parameters:
           - method: "genie"
             gini_threshold: 0.5
+      - id: genieclust2
+        name: "genieclust"
+        software_environment: "clustbench"
+        repository:
+          url: https://github.com/imallona/clustbench_genieclust
+          commit: 7d9e799
+        parameters:
           - method: ["gic", "ica"]
       - id: fcps
         name: "fcps"

Makefile

@@ -1,25 +1,25 @@
 MAX_CORES ?= 50
 
-# without continue on error (-k)
-OB_CMD=ob run benchmark --local --cores ${MAX_CORES} --yes
+# vanilla ob command
+OB_CMD=ob run --cores ${MAX_CORES}
 
 # prepare_apptainer_env:
 # 	cd envs && bash build_singularity.sh
 
-all: run_with_default_conda run_with_unpinned_oras run_with_default_envs # knit_report
+all: run_with_default_conda run_with_unpinned_oras run_with_default_envs
 
 run_with_default_conda:
-	${OB_CMD} -b Clustering_conda.yml
+	${OB_CMD} Clustering_conda.yml
 	cp Clustering_conda.yml out
 	mv out out_conda_$(shell date +'%Y%m%d_%H%M')
 
 run_with_unpinned_oras:
-	${OB_CMD} -b Clustering_oras.yml
+	${OB_CMD} Clustering_oras.yml
 	cp Clustering_oras.yml out
 	mv out out_singularity_$(shell date +'%Y%m%d_%H%M')
 
 run_with_default_envs:
-	${OB_CMD} -b Clustering_envmodules.yml
+	${OB_CMD} Clustering_envmodules.yml
 	cp Clustering_envmodules.yml out
 	mv out out_envmodules_$(shell date +'%Y%m%d_%H%M')
 

envs/README.md

@@ -87,13 +87,20 @@ eb --robot fcps.eb --job-cores=10
 eb --robot clustbench.eb --job-cores=10
 ```
 
-Running a benchmark:
+Running a benchmark assuming the modules were installed within the `MODULE_BASEPATH`/software/modules/all path (perhaps `MODULE_BASEPATH` equals `$EASYBUILD_PREFIX`):
 
 ```bash
+# during a debugging session, this was `/home/mark/envmodules_debug/.local/easybuild/`
+MODULE_BASEPATH='updateme'
+
 source /cvmfs/software.eessi.io/versions/2025.06/init/lmod/bash ## if not loaded
 module load EESSI-extend/2025.06-easybuild                      ## if not loaded
-export MODULEPATH="$EASYBUILD_PREFIX"/software/modules/all:"$MODULEPATH"
+export MODULEPATH="$MODULE_BASEPATH"/software/modules/all:"$MODULEPATH"
 module use $MODULEPATH
+echo $MODULEPATH
+
+# during a debugging session, the resulting $MODULEPATH contained both the MODULE_BASEPATH and EESSI's injections
+# /home/mark/envmodules_debug/.local/easybuild/software/modules/all:/home/mark/eessi/versions/2025.06/software/linux/x86_64/amd/zen2/modules/all:/cvmfs/software.eessi.io/host_injections/2025.06/software/linux/x86_64/amd/zen2/modules/all:/cvmfs/software.eessi.io/versions/2025.06/software/linux/x86_64/amd/zen2/modules/all:/cvmfs/software.eessi.io/versions/2025.06/init/modules
 
 ob run benchmark -b Clustering_envmodules.yml  --local-storage --cores 30
 ```

envs/clustbench.eb

@@ -23,8 +23,7 @@ builddependencies = [
 dependencies = [
     ('Python', '3.12.3'),
     ('SciPy-bundle', '2024.05'),
-    ('matplotlib', '3.9.2'),
-    ('scikit-learn', '1.6.1')
+    ('matplotlib', '3.9.2')
 ]
 
 sanity_pip_check = True
@@ -46,10 +45,10 @@ exts_list = [
     ('fastcluster', '1.2.6', {
         'checksums': ['aab886efa7b6bba7ac124f4498153d053e5a08b822d2254926b7206cdf5a8aa6'],
     }),
-    ## ('scikit-learn', '1.6.1', {
-    ##    'modulename': 'sklearn',
-    ##     'source_urls' : [PYPI_SOURCE]
-    ## }),
+    ('scikit-learn', '1.4.2', {
+        'modulename': 'sklearn',
+         'source_urls' : [PYPI_SOURCE]
+    }),
     ('genieclust', '1.1.6', {
         'source_urls': ['https://github.com/gagolews/genieclust/archive/refs/tags/'],
         'sources': ['v1.1.6.tar.gz'],