Spaces:
Runtime error
Runtime error
Commit ·
7a18ba5
1
Parent(s): 9caffdf
Update app.py
Browse files
app.py
CHANGED
|
@@ -4,6 +4,7 @@ import numpy as np
|
|
| 4 |
|
| 5 |
import streamlit as st
|
| 6 |
from pathlib import Path
|
|
|
|
| 7 |
|
| 8 |
import sys
|
| 9 |
path_root = Path("./")
|
|
@@ -83,6 +84,22 @@ def preferences_from_hits(list_of_hits):
|
|
| 83 |
return np.array(preferences), id2doc
|
| 84 |
|
| 85 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 86 |
def aggregate(list_of_hits):
|
| 87 |
import numpy as np
|
| 88 |
from permsc import KemenyOptimalAggregator, sum_kendall_tau, ranks_from_preferences
|
|
@@ -92,15 +109,16 @@ def aggregate(list_of_hits):
|
|
| 92 |
y_optimal = KemenyOptimalAggregator().aggregate(preferences)
|
| 93 |
# y_optimal = BordaRankAggregator().aggregate(preferences)
|
| 94 |
|
| 95 |
-
print("-------------------------------------")
|
| 96 |
-
print("preference:")
|
| 97 |
-
print(preferences)
|
| 98 |
-
print("preferences shape: ", preferences.shape)
|
| 99 |
-
print("y_optimal: ", y_optimal)
|
| 100 |
|
| 101 |
return [id2doc[id] for id in y_optimal]
|
| 102 |
|
| 103 |
aggregated_ranking = aggregate(query2outputs[search_query])
|
|
|
|
| 104 |
|
| 105 |
if search_query or button_clicked:
|
| 106 |
|
|
@@ -112,17 +130,22 @@ if search_query or button_clicked:
|
|
| 112 |
st.write(
|
| 113 |
f'<p align=\"right\" style=\"color:grey;\"> Before aggregation for query [{search_query}] ms</p>', unsafe_allow_html=True)
|
| 114 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 115 |
for i, result in enumerate(search_results):
|
| 116 |
result_id = result["docid"]
|
| 117 |
contents = result["content"]
|
| 118 |
|
|
|
|
| 119 |
# output = f'<div class="row"> <b>Rank</b>: {i+1} | <b>Document ID</b>: {result_id} | <b>Score</b>:{result_score:.2f}</div>'
|
| 120 |
-
output = f'<div class="row"> <b>Rank</b>: {i+1} | <b>Document ID</b>: {result_id}'
|
| 121 |
|
| 122 |
try:
|
| 123 |
st.write(output, unsafe_allow_html=True)
|
| 124 |
st.write(
|
| 125 |
-
f'<div class="row">{contents}</div>', unsafe_allow_html=True)
|
| 126 |
|
| 127 |
except:
|
| 128 |
pass
|
|
|
|
| 4 |
|
| 5 |
import streamlit as st
|
| 6 |
from pathlib import Path
|
| 7 |
+
from collections import defaultdict
|
| 8 |
|
| 9 |
import sys
|
| 10 |
path_root = Path("./")
|
|
|
|
| 84 |
return np.array(preferences), id2doc
|
| 85 |
|
| 86 |
|
| 87 |
+
def load_qrels(name):
|
| 88 |
+
import ir_datasets
|
| 89 |
+
if name == "dl19":
|
| 90 |
+
ds_name = "msmarco-passage/trec-dl-2019/judged"
|
| 91 |
+
elif name == "dl20":
|
| 92 |
+
ds_name = "msmarco-passage/trec-dl-2020/judged"
|
| 93 |
+
else:
|
| 94 |
+
raise ValueError(name)
|
| 95 |
+
|
| 96 |
+
dataset = ir_datasets.load(ds_name)
|
| 97 |
+
qrels = defaultdict(dict)
|
| 98 |
+
for qrel in dataset.qrels_iter():
|
| 99 |
+
qrels[qrel.query_id][qrel.doc_id] = qrel.relevance
|
| 100 |
+
return qrels
|
| 101 |
+
|
| 102 |
+
|
| 103 |
def aggregate(list_of_hits):
|
| 104 |
import numpy as np
|
| 105 |
from permsc import KemenyOptimalAggregator, sum_kendall_tau, ranks_from_preferences
|
|
|
|
| 109 |
y_optimal = KemenyOptimalAggregator().aggregate(preferences)
|
| 110 |
# y_optimal = BordaRankAggregator().aggregate(preferences)
|
| 111 |
|
| 112 |
+
# print("-------------------------------------")
|
| 113 |
+
# print("preference:")
|
| 114 |
+
# print(preferences)
|
| 115 |
+
# print("preferences shape: ", preferences.shape)
|
| 116 |
+
# print("y_optimal: ", y_optimal)
|
| 117 |
|
| 118 |
return [id2doc[id] for id in y_optimal]
|
| 119 |
|
| 120 |
aggregated_ranking = aggregate(query2outputs[search_query])
|
| 121 |
+
qrels = load_qrels("dl19")
|
| 122 |
|
| 123 |
if search_query or button_clicked:
|
| 124 |
|
|
|
|
| 130 |
st.write(
|
| 131 |
f'<p align=\"right\" style=\"color:grey;\"> Before aggregation for query [{search_query}] ms</p>', unsafe_allow_html=True)
|
| 132 |
|
| 133 |
+
qid = {result["qid"] for result in search_results}
|
| 134 |
+
assert len(qid) == 1
|
| 135 |
+
qid = list(qid)[0]
|
| 136 |
+
|
| 137 |
for i, result in enumerate(search_results):
|
| 138 |
result_id = result["docid"]
|
| 139 |
contents = result["content"]
|
| 140 |
|
| 141 |
+
style = "style=\"color:grey;\"" if qrels[qid].get(result_id, 0) else "style=\"color:red;\""
|
| 142 |
# output = f'<div class="row"> <b>Rank</b>: {i+1} | <b>Document ID</b>: {result_id} | <b>Score</b>:{result_score:.2f}</div>'
|
| 143 |
+
output = f'<div class="row" {style}> <b>Rank</b>: {i+1} | <b>Document ID</b>: {result_id}'
|
| 144 |
|
| 145 |
try:
|
| 146 |
st.write(output, unsafe_allow_html=True)
|
| 147 |
st.write(
|
| 148 |
+
f'<div class="row" {style}>{contents}</div>', unsafe_allow_html=True)
|
| 149 |
|
| 150 |
except:
|
| 151 |
pass
|