cleanup

codex/blueprints/app.py

@@ -75,13 +75,7 @@ def stats():
     whole_word = request.args.get("whole_word", 0, type=int)
 
     logger.info(f"Generating stats {activity_suffix(filter_string, data_version)}")
-    (
-        filtered_root_id_list,
-        num_items,
-        hint,
-        data_stats,
-        data_charts,
-    ) = stats_cached(
+    (filtered_root_id_list, num_items, hint, data_stats, data_charts,) = stats_cached(
         filter_string=filter_string,
         data_version=data_version,
         case_sensitive=case_sensitive,

codex/data/auto_naming.py

@@ -1,92 +1,16 @@
 from collections import defaultdict
 
-from codex.data.neurotransmitters import NEURO_TRANSMITTER_NAMES
-
-
-def extract_label_parts(lbl, with_subparts):
-    parts = []
-    for super_part in lbl.split(";"):
-        if with_subparts:
-            for sub_part in super_part.split(","):
-                parts.append(sub_part.strip())
-        else:
-            parts.append(super_part.strip())
-    return parts
-
-
-def make_canonic(s):
-    return s.upper().replace(" ", "_").replace("-", "_")
-
-
-def is_valid_token(t, canonic_coarse_annos):
-    if any([c in t for c in [" ", ".", ",", "?", "(", ")", "[", "]"]]):
-        return False
-    if any(
-        [
-            c in t.lower()
-            for c in [
-                "ascending",
-                "descending",
-                "unclassified",
-                "clone",
-                "test",
-                "odd",
-                "putative",
-                "fbbt",
-                "eye_",
-                "murthy",
-                "seung",
-            ]
-        ]
-    ):
-        return False
-    if any(
-        [
-            t.lower().startswith(c)
-            for c in [
-                "put_",
-            ]
-        ]
-    ):
-        return False
-    if t.isnumeric():
-        return False
-    if t.isalpha() and (t == t.lower() or t == t.capitalize()):
-        return False
-    if make_canonic(t) in canonic_coarse_annos:
-        return False
-    return True
-
 
 def assign_names_from_annotations(neuron_data):
     token_to_cell_counts = defaultdict(int)
     cell_to_potential_names = defaultdict(list)
 
-    canonic_coarse_annos = set(
-        [make_canonic(nd["super_class"]) for nd in neuron_data.values()]
-    )
-    canonic_coarse_annos |= set(
-        [make_canonic(nd["group"]) for nd in neuron_data.values()]
-    )
-    canonic_coarse_annos |= set(
-        [make_canonic(nt_key) for nt_key in NEURO_TRANSMITTER_NAMES.keys()]
-    )
-    canonic_coarse_annos |= set(
-        [make_canonic(nt_desc) for nt_desc in NEURO_TRANSMITTER_NAMES.values()]
-    )
-
     for rid, nd in neuron_data.items():
         token_to_cell_counts[nd["group"]] += 1
         cell_to_potential_names[rid].append(nd["group"])
 
     assert len(cell_to_potential_names) == len(neuron_data)
 
-    for t1 in list(token_to_cell_counts.keys()):
-        for t2 in list(token_to_cell_counts.keys()):
-            if t1 != t2 and make_canonic(t1) == make_canonic(t2):
-                print(f"{t1} --> {t2}")
-                assert False
-
     assigned_name_counts = defaultdict(int)
 
     # if there are multiple options, chose one deterministically

codex/data/local_data_loader.py

@@ -115,7 +115,6 @@ def _read_data(filename, with_timestamp=False):
         connectivity_cluster_rows=connectivity_cluster_rows,
         svd_rows=svd_rows,
         lr_matching_rows=lr_matching_rows,
-        olr_prediction_rows=[],
     )
     # free mem
     del neuron_rows

codex/data/neuron_data.py

@@ -53,7 +53,6 @@ def __init__(
         grouped_connection_counts,
         grouped_reciprocal_connection_counts,
         svd_rows,
-        olr_prediction_rows,
     ):
         self.neuron_data = neuron_attributes
         self.connections_ = Connections(neuron_connection_rows)

codex/data/neuron_data_initializer.py

@@ -116,7 +116,6 @@ def initialize_neuron_data(
     connectivity_cluster_rows,
     svd_rows,
     lr_matching_rows,
-    olr_prediction_rows,
 ):
     neuron_attributes = {}
     neuron_connection_rows = []
@@ -473,5 +472,4 @@ def label_row_to_dict(row):
         grouped_connection_counts=grouped_connection_counts,
         grouped_reciprocal_connection_counts=grouped_reciprocal_connection_counts,
         svd_rows=svd_rows,
-        olr_prediction_rows=olr_prediction_rows,
     )