Update README.md

README.md

@@ -27,3 +27,78 @@ configs:
   - split: train
     path: data/train-*
 ---
+
+This is reranking dataset built from 179 queris from [MSMARCO-V2 passages set](https://ir-datasets.com/msmarco-passage-v2.html).
+
+Below are the data creation process:
+
+```
+import sys
+from pyserini.search import get_topics, get_qrels
+from run_evaluation import THE_TOPICS, THE_INDEX
+from trec_eval import EvalFunction
+import json
+qrels = {}
+for data in ['dl21', 'dl22', 'dl23']:
+    qrels_file = get_qrels_file(THE_TOPICS[data])
+    with open(qrels_file, 'r') as f:
+        for line in f:
+            qid, _, docid, rel = line.strip().split()
+            if qid not in qrels:
+                qrels[qid] = {}
+            qrels[qid][docid] = int(rel)
+           
+# Save the combined qrels to a file
+output_file = './combined_qrels.txt'
+with open(output_file, 'w') as f:
+    for qid in sorted(qrels.keys()):
+        for docid, rel in qrels[qid].items():
+            f.write(f"{qid} 0 {docid} {rel}\n")
+
+for data in ['dl21', 'dl22', 'dl23']:
+  topics = get_topics(THE_TOPICS[data] if data not in dl else data)
+  print(f"\nEvaluating {data}:, len(topics): {len(topics)}")
+  EvalFunction.main(THE_TOPICS[data], f'./20{data[2:]}_passage_top100.txt')
+
+# Concate the files
+combined_rank_results = []
+for data in ['dl21', 'dl22', 'dl23']:
+    with open(f'{data}_bm25_rank_results.json', 'r') as f:
+        rank_results = json.load(f)
+        print(f"len(rank_results): {len(rank_results)}")
+        combined_rank_results.extend(rank_results)
+
+# Multiple Random Sampling
+
+import copy
+import random
+replicate_rank_results = []
+replicate_times = 100
+for item in combined_rank_results:
+    for _ in range(replicate_times):
+        new_item = copy.deepcopy(item)
+        # Randomly select 20 hits from original hits list
+        random_select_index = random.sample(range(len(item['hits'])), min(20, len(item['hits'])))
+        random_select_index.sort()
+        new_item['hits'] = [new_item['hits'][i] for i in random_select_index]
+        replicate_rank_results.append(new_item)
+print(len(replicate_rank_results))
+
+# Push to huggingface
+
+from datasets import Dataset
+import pandas as pd
+from huggingface_hub import login
+df = pd.DataFrame(replicate_rank_results)
+dataset = Dataset.from_pandas(df)
+
+# Print dataset info
+print(f"Dataset size: {len(dataset)}")
+print(dataset)
+
+# Upload the dataset to Huggingface Hub
+dataset.push_to_hub(
+    "le723z/DeepRerank",  # Replace with your desired repository name
+    private=False,  # Set to True if you want a private dataset
+)
+```