akhooli commited on
Commit
ebb0647
·
verified ·
1 Parent(s): 1a1b3d6

Upload 2 files

Browse files
colbert/indexes/akhooli/akhooli/Arabic-ColBERT-100knew_index/centroids.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84262883be997e310519aa31e6f706168363055b180925039ef56854553d2dc1
3
+ size 1049766
colbert/indexes/akhooli/akhooli/Arabic-ColBERT-100knew_index/plan.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "config": {
3
+ "query_token_id": "[unused0]",
4
+ "doc_token_id": "[unused1]",
5
+ "query_token": "[Q]",
6
+ "doc_token": "[D]",
7
+ "ncells": null,
8
+ "centroid_score_threshold": null,
9
+ "ndocs": null,
10
+ "load_index_with_mmap": false,
11
+ "index_path": null,
12
+ "index_bsize": 32,
13
+ "nbits": 8,
14
+ "kmeans_niters": 20,
15
+ "resume": false,
16
+ "similarity": "cosine",
17
+ "bsize": 64,
18
+ "accumsteps": 1,
19
+ "lr": 5e-6,
20
+ "maxsteps": 100000,
21
+ "save_every": 3315,
22
+ "warmup": 3315,
23
+ "warmup_bert": null,
24
+ "relu": false,
25
+ "nway": 2,
26
+ "use_ib_negatives": true,
27
+ "reranker": false,
28
+ "distillation_alpha": 1.0,
29
+ "ignore_scores": false,
30
+ "model_name": "Arabic-ColBERT-100k",
31
+ "query_maxlen": 32,
32
+ "attend_to_mask_tokens": false,
33
+ "interaction": "colbert",
34
+ "dim": 128,
35
+ "doc_maxlen": 256,
36
+ "mask_punctuation": true,
37
+ "checkpoint": "akhooli\/Arabic-ColBERT-100k",
38
+ "triples": "data\/triples.train.colbert.jsonl",
39
+ "collection": [
40
+ "list with 1663 elements starting with...",
41
+ [
42
+ "\u0628\u0633\u0645 \u0627\u0644\u0644\u0647 \u0627\u0644\u0631\u062d\u0645\u0646 \u0627\u0644\u0631\u062d\u064a\u0645 0.\n\u0633\u0628\u062d \u0644\u0644\u0647 \u0645\u0627 \u0641\u064a \u0627\u0644\u0633\u0645\u0627\u0648\u0627\u062a \u0648\u0627\u0644\u0623\u0631\u0636 \u0648\u0647\u0648 \u0627\u0644\u0639\u0632\u064a\u0632 \u0627\u0644\u062d\u0643\u064a\u0645 1.\n\u0644\u0647 \u0645\u0644\u0643 \u0627\u0644\u0633\u0645\u0627\u0648\u0627\u062a \u0648\u0627\u0644\u0623\u0631\u0636 \u064a\u062d\u064a\u064a \u0648\u064a\u0645\u064a\u062a \u0648\u0647\u0648 \u0639\u0644\u0649 \u0643\u0644 \u0634\u064a\u0621 \u0642\u062f\u064a\u0631 2.\n\u0647\u0648 \u0627\u0644\u0623\u0648\u0644 \u0648\u0627\u0644\u0622\u062e\u0631 \u0648\u0627\u0644\u0638\u0627\u0647\u0631 \u0648\u0627\u0644\u0628\u0627\u0637\u0646 \u0648\u0647\u0648 \u0628\u0643\u0644 \u0634\u064a\u0621 \u0639\u0644\u064a\u0645 3.\n\u0647\u0648 \u0627\u0644\u0630\u064a \u062e\u0644\u0642 \u0627\u0644\u0633\u0645\u0627\u0648\u0627\u062a \u0648\u0627\u0644\u0623\u0631\u0636 \u0641\u064a \u0633\u062a\u0629 \u0623\u064a\u0627\u0645 \u062b\u0645 \u0627\u0633\u062a\u0648\u0649 \u0639\u0644\u0649 \u0627\u0644\u0639\u0631\u0634 \u064a\u0639\u0644\u0645 \u0645\u0627 \u064a\u0644\u062c \u0641\u064a \u0627\u0644\u0623\u0631\u0636 \u0648\u0645\u0627 \u064a\u062e\u0631\u062c \u0645\u0646\u0647\u0627 \u0648\u0645\u0627 \u064a\u0646\u0632\u0644 \u0645\u0646 \u0627\u0644\u0633\u0645\u0627\u0621 \u0648\u0645\u0627 \u064a\u0639\u0631\u062c \u0641\u064a\u0647\u0627 \u0648\u0647\u0648 \u0645\u0639\u0643\u0645 \u0623\u064a\u0646 \u0645\u0627 \u0643\u0646\u062a\u0645 \u0648\u0627\u0644\u0644\u0647 \u0628\u0645\u0627 \u062a\u0639\u0645\u0644\u0648\u0646 \u0628\u0635\u064a\u0631 4.",
43
+ "\u0644\u0647 \u0645\u0644\u0643 \u0627\u0644\u0633\u0645\u0627\u0648\u0627\u062a \u0648\u0627\u0644\u0623\u0631\u0636 \u0648\u0625\u0644\u0649 \u0627\u0644\u0644\u0647 \u062a\u0631\u062c\u0639 \u0627\u0644\u0623\u0645\u0648\u0631 5.\n\u064a\u0648\u0644\u062c \u0627\u0644\u0644\u064a\u0644 \u0641\u064a \u0627\u0644\u0646\u0647\u0627\u0631 \u0648\u064a\u0648\u0644\u062c \u0627\u0644\u0646\u0647\u0627\u0631 \u0641\u064a \u0627\u0644\u0644\u064a\u0644 \u0648\u0647\u0648 \u0639\u0644\u064a\u0645 \u0628\u0630\u0627\u062a \u0627\u0644\u0635\u062f\u0648\u0631 6.\n\u0622\u0645\u0646\u0648\u0627 \u0628\u0627\u0644\u0644\u0647 \u0648\u0631\u0633\u0648\u0644\u0647 \u0648\u0623\u0646\u0641\u0642\u0648\u0627 \u0645\u0645\u0627 \u062c\u0639\u0644\u0643\u0645 \u0645\u0633\u062a\u062e\u0644\u0641\u064a\u0646 \u0641\u064a\u0647 \u0641\u0627\u0644\u0630\u064a\u0646 \u0622\u0645\u0646\u0648\u0627 \u0645\u0646\u0643\u0645 \u0648\u0623\u0646\u0641\u0642\u0648\u0627 \u0644\u0647\u0645 \u0623\u062c\u0631 \u0643\u0628\u064a\u0631 7.\n\u0648\u0645\u0627 \u0644\u0643\u0645 \u0644\u0627 \u062a\u0624\u0645\u0646\u0648\u0646 \u0628\u0627\u0644\u0644\u0647 \u0648\u0627\u0644\u0631\u0633\u0648\u0644 \u064a\u062f\u0639\u0648\u0643\u0645 \u0644\u062a\u0624\u0645\u0646\u0648\u0627 \u0628\u0631\u0628\u0643\u0645 \u0648\u0642\u062f \u0623\u062e\u0630 \u0645\u064a\u062b\u0627\u0642\u0643\u0645 \u0625\u0646 \u0643\u0646\u062a\u0645 \u0645\u0624\u0645\u0646\u064a\u0646 8.",
44
+ "\u0647\u0648 \u0627\u0644\u0630\u064a \u064a\u0646\u0632\u0644 \u0639\u0644\u0649 \u0639\u0628\u062f\u0647 \u0622\u064a\u0627\u062a \u0628\u064a\u0646\u0627\u062a \u0644\u064a\u062e\u0631\u062c\u0643\u0645 \u0645\u0646 \u0627\u0644\u0638\u0644\u0645\u0627\u062a \u0625\u0644\u0649 \u0627\u0644\u0646\u0648\u0631 \u0648\u0625\u0646 \u0627\u0644\u0644\u0647 \u0628\u0643\u0645 \u0644\u0631\u0621\u0648\u0641 \u0631\u062d\u064a\u0645 9.\n\u0648\u0645\u0627 \u0644\u0643\u0645 \u0623\u0644\u0627 \u062a\u0646\u0641\u0642\u0648\u0627 \u0641\u064a \u0633\u0628\u064a\u0644 \u0627\u0644\u0644\u0647 \u0648\u0644\u0644\u0647 \u0645\u064a\u0631\u0627\u062b \u0627\u0644\u0633\u0645\u0627\u0648\u0627\u062a \u0648\u0627\u0644\u0623\u0631\u0636 \u0644\u0627 \u064a\u0633\u062a\u0648\u064a \u0645\u0646\u0643\u0645 \u0645\u0646 \u0623\u0646\u0641\u0642 \u0645\u0646 \u0642\u0628\u0644 \u0627\u0644\u0641\u062a\u062d \u0648\u0642\u0627\u062a\u0644 \u0623\u0648\u0644\u0626\u0643 \u0623\u0639\u0638\u0645 \u062f\u0631\u062c\u0629 \u0645\u0646 \u0627\u0644\u0630\u064a\u0646 \u0623\u0646\u0641\u0642\u0648\u0627 \u0645\u0646 \u0628\u0639\u062f \u0648\u0642\u0627\u062a\u0644\u0648\u0627 \u0648\u0643\u0644\u0627 \u0648\u0639\u062f \u0627\u0644\u0644\u0647 \u0627\u0644\u062d\u0633\u0646\u0649 \u0648\u0627\u0644\u0644\u0647 \u0628\u0645\u0627 \u062a\u0639\u0645\u0644\u0648\u0646 \u062e\u0628\u064a\u0631 10.\n\u0645\u0646 \u0630\u0627 \u0627\u0644\u0630\u064a \u064a\u0642\u0631\u0636 \u0627\u0644\u0644\u0647 \u0642\u0631\u0636\u0627 \u062d\u0633\u0646\u0627 \u0641\u064a\u0636\u0627\u0639\u0641\u0647 \u0644\u0647 \u0648\u0644\u0647 \u0623\u062c\u0631 \u0643\u0631\u064a\u0645 11."
45
+ ]
46
+ ],
47
+ "queries": "data\/queries.train.colbert.tsv",
48
+ "index_name": "akhooli\/Arabic-ColBERT-100knew_index",
49
+ "overwrite": false,
50
+ "root": ".ragatouille\/",
51
+ "experiment": "colbert",
52
+ "index_root": null,
53
+ "name": "2024-07\/19\/11.19.26",
54
+ "rank": 0,
55
+ "nranks": 1,
56
+ "amp": true,
57
+ "gpus": 0,
58
+ "avoid_fork_if_possible": false
59
+ },
60
+ "num_chunks": 1,
61
+ "num_partitions": 4096,
62
+ "num_embeddings_est": 130844.00312042236,
63
+ "avg_doclen_est": 78.67949676513672
64
+ }