Fraser-Greenlee
commited on
Commit
·
0fc8b1d
1
Parent(s):
1126e74
add toodo
Browse files- make_variations.py +20 -3
make_variations.py
CHANGED
|
@@ -9,6 +9,7 @@ import ast
|
|
| 9 |
|
| 10 |
|
| 11 |
def write_rows(rows):
|
|
|
|
| 12 |
with open('data.10_simple_variations.jsonl', 'a') as f:
|
| 13 |
f.writelines(rows)
|
| 14 |
|
|
@@ -129,15 +130,26 @@ def make_alternative_rows(start, code):
|
|
| 129 |
variations[alt_start] = alt_start_code_end
|
| 130 |
n_tries += 1
|
| 131 |
|
|
|
|
| 132 |
'''
|
| 133 |
-
|
| 134 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 135 |
'''
|
| 136 |
return [
|
| 137 |
json.dumps({'start': st, 'code': cd, 'end': en}) + '\n' for st, cd, en in variations.values()
|
| 138 |
]
|
| 139 |
|
| 140 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 141 |
with open('new_all_states.txt', 'r') as f:
|
| 142 |
rows = []
|
| 143 |
prev_lines = []
|
|
@@ -159,7 +171,12 @@ with open('new_all_states.txt', 'r') as f:
|
|
| 159 |
rows.append(json.dumps({
|
| 160 |
'start': start, 'code': code, 'end': end
|
| 161 |
}) + '\n')
|
| 162 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 163 |
if len(rows) > 100_000:
|
| 164 |
breakpoint()
|
| 165 |
write_rows(rows)
|
|
|
|
| 9 |
|
| 10 |
|
| 11 |
def write_rows(rows):
|
| 12 |
+
# TODO write this as a gzip for much better compression
|
| 13 |
with open('data.10_simple_variations.jsonl', 'a') as f:
|
| 14 |
f.writelines(rows)
|
| 15 |
|
|
|
|
| 130 |
variations[alt_start] = alt_start_code_end
|
| 131 |
n_tries += 1
|
| 132 |
|
| 133 |
+
# TODO change the names (keep alphabetical order)
|
| 134 |
'''
|
| 135 |
+
get number of vals in start states (N)
|
| 136 |
+
get N random lowercase letters in alphabetical order
|
| 137 |
+
parse start state and shuffle the expr.body
|
| 138 |
+
make name map old->new using new characters
|
| 139 |
+
use visitor with name map to swap variable names using map
|
| 140 |
+
do this for start, code, end seperately
|
| 141 |
'''
|
| 142 |
return [
|
| 143 |
json.dumps({'start': st, 'code': cd, 'end': en}) + '\n' for st, cd, en in variations.values()
|
| 144 |
]
|
| 145 |
|
| 146 |
|
| 147 |
+
STATS = {
|
| 148 |
+
'alt_count': 0,
|
| 149 |
+
'num_rows': 0
|
| 150 |
+
}
|
| 151 |
+
|
| 152 |
+
|
| 153 |
with open('new_all_states.txt', 'r') as f:
|
| 154 |
rows = []
|
| 155 |
prev_lines = []
|
|
|
|
| 171 |
rows.append(json.dumps({
|
| 172 |
'start': start, 'code': code, 'end': end
|
| 173 |
}) + '\n')
|
| 174 |
+
alt_rows = make_alternative_rows(start, code)
|
| 175 |
+
rows += alt_rows
|
| 176 |
+
|
| 177 |
+
STATS['alt_count'] += len(alt_rows)
|
| 178 |
+
STATS['num_rows'] += 1
|
| 179 |
+
|
| 180 |
if len(rows) > 100_000:
|
| 181 |
breakpoint()
|
| 182 |
write_rows(rows)
|