Update process.py
Browse files- process.py +4 -38
process.py
CHANGED
@@ -7,8 +7,8 @@ import pandas as pd
|
|
7 |
from datasets import load_dataset
|
8 |
|
9 |
|
10 |
-
def process(
|
11 |
-
data = load_dataset("relbert/t_rex",
|
12 |
df = data.to_pandas()
|
13 |
df.pop('text')
|
14 |
df.pop('title')
|
@@ -22,43 +22,9 @@ def process(name, split, output):
|
|
22 |
f.write('\n'.join([json.dumps(i) for i in rel_sim_data]))
|
23 |
|
24 |
|
25 |
-
parameters_min_e_freq = [1, 2, 3, 4]
|
26 |
-
parameters_max_p_freq = [100, 50, 25, 10]
|
27 |
os.makedirs("data", exist_ok=True)
|
28 |
|
29 |
-
for
|
30 |
-
|
31 |
-
process(
|
32 |
-
name=f"filter_unified.min_entity_{min_e_freq}_max_predicate_{max_p_freq}",
|
33 |
-
split=s,
|
34 |
-
output=f"data/filter_unified.min_entity_{min_e_freq}_max_predicate_{max_p_freq}.{s}.jsonl")
|
35 |
|
36 |
-
process(
|
37 |
-
name=f"filter_unified.min_entity_{min_e_freq}_max_predicate_{max_p_freq}",
|
38 |
-
split='test',
|
39 |
-
output=f"data/filter_unified.test.jsonl")
|
40 |
-
|
41 |
-
|
42 |
-
stats = []
|
43 |
-
for min_e_freq, max_p_freq in product(parameters_min_e_freq, parameters_max_p_freq):
|
44 |
-
stats_tmp = {"data": f"filter_unified.min_entity_{min_e_freq}_max_predicate_{max_p_freq}"}
|
45 |
-
for s in ['train', 'validation']:
|
46 |
-
with open(f"data/filter_unified.min_entity_{min_e_freq}_max_predicate_{max_p_freq}.{s}.jsonl") as f:
|
47 |
-
tmp = [json.loads(i) for i in f.read().split('\n') if len(i) > 0]
|
48 |
-
stats_tmp[f'num of relation types ({s})'] = len(tmp)
|
49 |
-
stats_tmp[f'average num of positive pairs ({s})'] = round(mean([len(i['positives']) for i in tmp]))
|
50 |
-
stats_tmp[f'average num of negative pairs ({s})'] = round(mean([len(i['negatives']) for i in tmp]))
|
51 |
-
stats.append(stats_tmp)
|
52 |
-
df_stats = pd.DataFrame(stats)
|
53 |
-
df_stats.index = df_stats.pop('data')
|
54 |
-
print(df_stats.to_markdown())
|
55 |
-
|
56 |
-
stats_tmp = {}
|
57 |
-
with open("data/filter_unified.test.jsonl") as f:
|
58 |
-
tmp = [json.loads(i) for i in f.read().split('\n') if len(i) > 0]
|
59 |
-
stats_tmp[f'num of relation types (test)'] = len(tmp)
|
60 |
-
stats_tmp[f'average num of positive pairs (test)'] = round(mean([len(i['positives']) for i in tmp]))
|
61 |
-
stats_tmp[f'average num of negative pairs (test)'] = round(mean([len(i['negatives']) for i in tmp]))
|
62 |
-
df_stats_test = pd.DataFrame([stats_tmp])
|
63 |
-
print(df_stats_test.to_markdown(index=False))
|
64 |
|
|
|
7 |
from datasets import load_dataset
|
8 |
|
9 |
|
10 |
+
def process(split, output):
|
11 |
+
data = load_dataset("relbert/t_rex", split=split)
|
12 |
df = data.to_pandas()
|
13 |
df.pop('text')
|
14 |
df.pop('title')
|
|
|
22 |
f.write('\n'.join([json.dumps(i) for i in rel_sim_data]))
|
23 |
|
24 |
|
|
|
|
|
25 |
os.makedirs("data", exist_ok=True)
|
26 |
|
27 |
+
for s in ['train', 'validation', 'test']:
|
28 |
+
process(split=s, output=f"data/filter_unified.{s}.jsonl")
|
|
|
|
|
|
|
|
|
29 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
30 |
|