update stats; looks great!

danvk · Nov 27, 2024 · 59d4940 · 59d4940
1 parent 20eb1d0
commit 59d4940
Show file tree

Hide file tree

Showing 3 changed files with 45 additions and 45 deletions.
diff --git a/data/lat-lon-to-ids.json b/data/lat-lon-to-ids.json
diff --git a/test/geocode-performance.txt b/test/geocode-performance.txt
@@ -1,6 +1,6 @@
 
 Results:
   Geocodes
-    229 / 269 = 85.13% of locatable images correctly located.
-     12 / 241 = 4.98% incorrectly located.
+    232 / 269 = 86.25% of locatable images correctly located.
+     10 / 242 = 4.13% incorrectly located.
 
diff --git a/test/geocoding-stats.txt b/test/geocoding-stats.txt
@@ -3,69 +3,69 @@
 alt titles matched: 0
      total matches: 31270
           counters: [('boro-int', 30503), ('title', 30096), ('alt_title', 1174), ('at-int', 288), ('num-prefix', 271), ('between', 208)]
-            grid: 24513 (31113 attempts)
+            grid: 25380 (31113 attempts)
 Grid statistics:
-           Counts: [('exact', 13054), ('dir strip', 12520), ('exact: str', 1014), ('interpolated', 280), ('extrapolated', 179), ('unclaimed', 163), ('cursed', 7), ('exact_grid', 5)]
-  Unknown avenues: Counter({'13': 65, 'Broadway': 9, '36': 7, '19': 6, '85': 5, '62': 3, '53': 2, '88': 2, '93': 2, '95': 2, '22': 2, '98': 2, '57': 1, '49': 1, '26': 1, '46': 1, '44': 1, '96': 1})
+           Counts: [('exact', 13920), ('dir strip', 12957), ('exact: str', 759), ('interpolated', 176), ('unclaimed', 136), ('extrapolated', 105), ('cursed', 7), ('exact_grid', 3)]
+  Unknown avenues: Counter({'13': 38, 'Broadway': 9, '36': 7, '19': 6, '85': 5, '62': 3, '53': 2, '88': 2, '93': 2, '95': 2, '22': 2, '98': 2, '57': 1, '49': 1, '26': 1, '46': 1, '44': 1, '96': 1})
   Unknown streets: Counter({'212': 11, '193': 10, '145': 6, '129': 3, '213': 3, '157': 2, '139': 2, '143': 2, '192': 2, '142': 2, '181': 2, '130': 2, '159': 1, '174': 1, '208': 1})
-          google: 1876
-   boro mismatch: 480
-        failures: 4238
+          google: 1700
+   boro mismatch: 434
+        failures: 3593
 Google geocoder stats:
      Cache misses: 0
-  Cache files hit: 7927
-[('google: intersection - fail', 9791), ('google: intersection - success', 2440), ('google: address - success', 1851), ('google: intersection - boro mismatch', 1014), ('google: address - fail', 159), ('google: address - boro mismatch', 131), ('cursed', 10)]
+  Cache files hit: 7313
+[('google: intersection - fail', 8242), ('google: intersection - success', 2217), ('google: address - success', 1660), ('google: intersection - boro mismatch', 902), ('google: address - fail', 138), ('google: address - boro mismatch', 79), ('cursed', 10)]
 -- Finalizing title-address --
- address matches: 626
-        patterns: [('street_pound', 426), ('num_street', 200)]
-          google: 601
+ address matches: 625
+        patterns: [('street_pound', 426), ('num_street', 199)]
+          google: 600
    boro mismatch: 9
         failures: 16
 Google geocoder stats:
      Cache misses: 0
-  Cache files hit: 7927
-[('google: intersection - fail', 9791), ('google: intersection - success', 2440), ('google: address - success', 1851), ('google: intersection - boro mismatch', 1014), ('google: address - fail', 159), ('google: address - boro mismatch', 131), ('cursed', 10)]
+  Cache files hit: 7313
+[('google: intersection - fail', 8242), ('google: intersection - success', 2217), ('google: address - success', 1660), ('google: intersection - boro mismatch', 902), ('google: address - fail', 138), ('google: address - boro mismatch', 79), ('cursed', 10)]
 -- Finalizing gpt --
-GPT POI:          14958
-GPT address:      3355
-GPT intersection: 11078
-            grid: 2529 (9296 attempts)
+GPT POI:          14556
+GPT address:      2795
+GPT intersection: 10062
+            grid: 2540 (8290 attempts)
 Grid statistics:
-           Counts: [('exact', 13054), ('dir strip', 12520), ('exact: str', 1014), ('interpolated', 280), ('extrapolated', 179), ('unclaimed', 163), ('cursed', 7), ('exact_grid', 5)]
-  Unknown avenues: Counter({'13': 65, 'Broadway': 9, '36': 7, '19': 6, '85': 5, '62': 3, '53': 2, '88': 2, '93': 2, '95': 2, '22': 2, '98': 2, '57': 1, '49': 1, '26': 1, '46': 1, '44': 1, '96': 1})
+           Counts: [('exact', 13920), ('dir strip', 12957), ('exact: str', 759), ('interpolated', 176), ('unclaimed', 136), ('extrapolated', 105), ('cursed', 7), ('exact_grid', 3)]
+  Unknown avenues: Counter({'13': 38, 'Broadway': 9, '36': 7, '19': 6, '85': 5, '62': 3, '53': 2, '88': 2, '93': 2, '95': 2, '22': 2, '98': 2, '57': 1, '49': 1, '26': 1, '46': 1, '44': 1, '96': 1})
   Unknown streets: Counter({'212': 11, '193': 10, '145': 6, '129': 3, '213': 3, '157': 2, '139': 2, '143': 2, '192': 2, '142': 2, '181': 2, '130': 2, '159': 1, '174': 1, '208': 1})
-          google: 1814
-   boro mismatch: 656
-        failures: 5696
+          google: 1577
+   boro mismatch: 538
+        failures: 4771
 Google geocoder stats:
      Cache misses: 0
-  Cache files hit: 7927
-[('google: intersection - fail', 9791), ('google: intersection - success', 2440), ('google: address - success', 1851), ('google: intersection - boro mismatch', 1014), ('google: address - fail', 159), ('google: address - boro mismatch', 131), ('cursed', 10)]
+  Cache files hit: 7313
+[('google: intersection - fail', 8242), ('google: intersection - success', 2217), ('google: address - success', 1660), ('google: intersection - boro mismatch', 902), ('google: address - fail', 138), ('google: address - boro mismatch', 79), ('cursed', 10)]
 -- Finalizing special --
-Special cases: [('Columbus Circle', 27), ('China Daily News', 23), ('Squatters: Camp Thomas Paine', 6), ('Mt. Sinai', 3), ('St. John the Divine', 1)]
+Special cases: [('Columbus Circle', 25), ('China Daily News', 23), ('Squatters: Camp Thomas Paine', 6), ('Mt. Sinai', 3), ('St. John the Divine', 1)]
 -- Finalizing subjects --
 POI/subject geocoding:
- 911	n_both
-1913	n_geo
- 109	n_geo_multi
-1804	n_geo_unambig
- 769	n_out_both_close
+ 909	n_both
+1847	n_geo
+ 104	n_geo_multi
+1743	n_geo_unambig
+ 767	n_out_both_close
   28	n_out_both_fallback_title
   26	n_out_both_subject
   88	n_out_both_title
- 919	n_out_subject
-1074	n_out_title
-1100	n_title
- 349	n_title_bridge
+ 860	n_out_subject
+1072	n_out_title
+1098	n_title
+ 347	n_title_bridge
  250	n_title_island
  501	n_title_park
 -- Final stats --
-26389 title-cross
-  601 title-address
- 4343 gpt
-   60 special
- 1993 subjects
-33386 (total)
+27080 title-cross
+  600 title-address
+ 4117 gpt
+   58 special
+ 1932 subjects
+33787 (total)
 Dropped w/ no date: 0
-Unique lat/longs: 10471
-Total photographs: 33386
+Unique lat/longs: 10597
+Total photographs: 33787