feat(neuron/netpyne): use %r instead of %e or %s

[sanjayankur31]

when writing recorded data to file.

%r or the repr function gives the "official" string representation of an object, so of a float too, while the %s or str method returns the "nicely printable string representation of an object".

Also, irrespective of which one we use, we must use the same for both NEURON and NetPyNE so that the outputs to files are of the same level of precision and same format etc.

With this PR, we get similar outputs for NEURON and NetPyNE:

instead of what we got before:

I haven't checked other exports, but if they're also being exported to Python scripts, we should update them too.

[pgleeson]

Ok, sounds good. Have you tested some omv tolerance=0 tests when this change is incorporated? May well change the actual value detected in the tests...

[sanjayankur31]

Ok, sounds good. Have you tested some omv tolerance=0 tests when this change is incorporated? May well change the actual value detected in the tests...

Testing them out now.

[sanjayankur31]

Did quick tests on the Allen institute cells. There's some change in tolerances, but not very dramatic:

diff --git a/CellTypesDatabase/models/NeuroML2/.test.483108201.jnmlnrn.omt b/CellTypesDatabase/models/NeuroML2/.test.483108201.jnmlnrn.omt
index 5f770ec..d889c20 100644
--- a/CellTypesDatabase/models/NeuroML2/.test.483108201.jnmlnrn.omt
+++ b/CellTypesDatabase/models/NeuroML2/.test.483108201.jnmlnrn.omt
@@ -14,4 +14,4 @@ experiments:
         spike detection: 
           method: threshold
           threshold: -20
-        tolerance: 0.0141
+        tolerance: 0.014139835945066405
diff --git a/CellTypesDatabase/models/NeuroML2/.test.486556811.jnmlnrn.omt b/CellTypesDatabase/models/NeuroML2/.test.486556811.jnmlnrn.omt
index c1954b8..2a5a36f 100644
--- a/CellTypesDatabase/models/NeuroML2/.test.486556811.jnmlnrn.omt
+++ b/CellTypesDatabase/models/NeuroML2/.test.486556811.jnmlnrn.omt
@@ -14,4 +14,4 @@ experiments:
         spike detection: 
           method: threshold
           threshold: -20
-        tolerance: 0
+        tolerance: 4.79184638209975e-05
diff --git a/CellTypesDatabase/models/NeuroML2/.test.jnmlnrn.omt b/CellTypesDatabase/models/NeuroML2/.test.jnmlnrn.omt
index e0361c8..4a661bd 100644
--- a/CellTypesDatabase/models/NeuroML2/.test.jnmlnrn.omt
+++ b/CellTypesDatabase/models/NeuroML2/.test.jnmlnrn.omt
@@ -14,7 +14,7 @@ experiments:
         spike detection: 
           method: threshold
           threshold: 0
-        tolerance: 2.2e-5
+        tolerance: 2.179620615235586e-05
   Ca:
     observables:
       spike times:
@@ -25,4 +25,4 @@ experiments:
         spike detection: 
           method: threshold
           threshold: 0.0014
-        tolerance: 0.00021
+        tolerance: 0.00020718032716604502
diff --git a/CellTypesDatabase/models/NeuroML2/.test.kv2like.jnmlnrn.omt b/CellTypesDatabase/models/NeuroML2/.test.kv2like.jnmlnrn.omt
index 797aa53..439f91d 100644
--- a/CellTypesDatabase/models/NeuroML2/.test.kv2like.jnmlnrn.omt
+++ b/CellTypesDatabase/models/NeuroML2/.test.kv2like.jnmlnrn.omt
@@ -14,7 +14,7 @@ experiments:
         spike detection: 
           method: threshold
           threshold: 0.885
-        tolerance: 1e-5
+        tolerance: 3.1032996127945875e-06
   h1:
     observables:
       spike times:
@@ -25,7 +25,7 @@ experiments:
         spike detection: 
           method: threshold
           threshold: 0.885
-        tolerance: 1e-5
+        tolerance: 8.696556449203658e-06
   h2:
     observables:
       spike times:
@@ -36,4 +36,4 @@ experiments:
         spike detection: 
           method: threshold
           threshold: 0.885
-        tolerance: 4e-5
+        tolerance: 4.1525039775804706e-05
diff --git a/CellTypesDatabase/models/NeuroML2/.test.nav.jnmlnrn.omt b/CellTypesDatabase/models/NeuroML2/.test.nav.jnmlnrn.omt
index ed785ae..2ab22de 100644
--- a/CellTypesDatabase/models/NeuroML2/.test.nav.jnmlnrn.omt
+++ b/CellTypesDatabase/models/NeuroML2/.test.nav.jnmlnrn.omt
@@ -14,7 +14,7 @@ experiments:
         spike detection: 
           method: threshold
           threshold: 0.2
-        tolerance: 2e-5
+        tolerance: 4.90571147973973e-06
   I4:
     observables:
       spike times:
@@ -25,4 +25,4 @@ experiments:
         spike detection: 
           method: threshold
           threshold: 0.03
-        tolerance: 2e-5
+        tolerance: 1.984186110237916e-05

There may be other things in the neuron export I guess? Also worth looking into next?