Updates mapping and covariance for speed

.github/workflows/ci.yml

@@ -152,9 +152,20 @@ jobs:
       run: |
         mkdir -p cache/workspaces
         source ./conda/bin/activate
+        # Just for the metadetect source pipeline
+        # we activate the debug mode with an env var
+        if [ "${{ matrix.pipeline }}" == "metadetect_source_only" ]; then
+          export TX_DASK_DEBUG=1
+        fi
         ceci examples/${{ matrix.pipeline }}/pipeline.yml
         test -f data/example/outputs_${{ matrix.pipeline }}/shear_xi_plus.png
 
+    - name: Show logs
+      if: always()
+      run: |
+        # always run
+        cat data/example/logs_${{ matrix.pipeline }}/*
+
     - name: Create output report
       if: matrix.os == 'macos-14'
       run: |

examples/cosmodc2/config.yml

@@ -143,9 +143,6 @@ TXRealGaussianCovariance:
     max_sep: 250.
     nbins: 20
     use_true_shear: false
-    nprocess: 4
-    threads_per_process: 2
-    nodes: 4
     galaxy_bias: [1.404, 1.458, 1.693, 1.922, 2.133] # Tinker bias values
 
 TXTwoPointFourier:

examples/metadetect_source_only/pipeline.yml

@@ -126,6 +126,6 @@ inputs:
 # if interrupted
 resume: true
 # where to put output logs for individual stages
-log_dir: data/example/metadetect_source_only
+log_dir: data/example/logs_metadetect_source_only
 # where to put an overall parsl pipeline log
 pipeline_log: data/example/log.txt

examples/redmagic/config.yml

@@ -180,8 +180,6 @@ TXSourceDiagnosticPlots:
     s2n_min: 1.25
     s2n_max: 300
 
-TXLensDiagnosticPlots: {}
-
 TXDiagnosticMaps:
     sparse: true  # Generate sparse maps - faster if using small areas
     snr_threshold: 10.0

examples/redmagic/pipeline.yml

@@ -42,7 +42,6 @@ stages:
         nz_plot: nz_lens
   - name: TXDiagnosticQuantiles
   - name: TXSourceDiagnosticPlots      # Make a suite of diagnostic plots
-  - name: TXLensDiagnosticPlots      # Make a suite of diagnostic plots
   - name: TXGammaTFieldCenters   # Compute and plot gamma_t around center points
     threads_per_process: 2
   - name: TXGammaTRandoms        # Compute and plot gamma_t around randoms