Hugging Face's logo

Medyassino
/
mon-workspace

ONNX
Model card Files
xet
 Community
Medyassino commited on
Commit
6f2d2e4
·
verified ·
1 Parent(s): 23b29aa

Add files using upload-large-folder tool

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. trained_models/RAGHUB_orchestratoor/MODEL_NAME.txt +1 -0
  2. trained_models/RAGHUB_orchestratoor/config_train.json +25 -0
  3. trained_models/RAGHUB_orchestratoor/history.json +92 -0
  4. trained_models/RAGHUB_orchestratoor/model_info.json +20 -0
  5. trained_models/RAGHUB_orchestratoor/tokenizer.json +0 -0
  6. trained_models/RAGHUB_orchestratoor/tokenizer_config.json +20 -0
  7. trained_models/RAGHUB_orchestratoor/training_info.json +7 -0
  8. trained_models/RAGHUB_orchestrator_full_v1/MODEL_NAME.txt +1 -0
  9. trained_models/RAGHUB_orchestrator_full_v1/config_train.json +33 -0
  10. trained_models/RAGHUB_orchestrator_full_v1/history.json +29 -0
  11. trained_models/RAGHUB_orchestrator_full_v1/model_info.json +20 -0
  12. trained_models/RAGHUB_orchestrator_full_v1/tokenizer.json +0 -0
  13. trained_models/RAGHUB_orchestrator_full_v1/tokenizer_config.json +20 -0
  14. trained_models/RAGHUB_orchestrator_full_v1/training_info.json +7 -0
  15. trained_models/RAGHUB_orchestrator_mixed_v1/MODEL_NAME.txt +1 -0
  16. trained_models/RAGHUB_orchestrator_mixed_v1/config_train.json +33 -0
  17. trained_models/RAGHUB_orchestrator_mixed_v1/history.json +47 -0
  18. trained_models/RAGHUB_orchestrator_mixed_v1/model_info.json +20 -0
  19. trained_models/RAGHUB_orchestrator_mixed_v1/tokenizer.json +0 -0
  20. trained_models/RAGHUB_orchestrator_mixed_v1/tokenizer_config.json +20 -0
  21. trained_models/RAGHUB_orchestrator_mixed_v1/training_info.json +7 -0
  22. trained_models/RAGHUB_orchestrator_v1_Yassino/MODEL_NAME.txt +1 -0
  23. trained_models/RAGHUB_orchestrator_v1_Yassino/config_train.json +33 -0
  24. trained_models/RAGHUB_orchestrator_v1_Yassino/history.json +56 -0
  25. trained_models/RAGHUB_orchestrator_v1_Yassino/model_info.json +20 -0
  26. trained_models/RAGHUB_orchestrator_v1_Yassino/tokenizer.json +0 -0
  27. trained_models/RAGHUB_orchestrator_v1_Yassino/tokenizer_config.json +20 -0
  28. trained_models/RAGHUB_orchestrator_v1_Yassino/training_info.json +7 -0
  29. trained_models/RAGHUB_orchestrator_v2_expert/MODEL_NAME.txt +1 -0
  30. trained_models/RAGHUB_orchestrator_v2_expert/training_info.json +7 -0
  31. trained_models/checkpoints_rag_fr/history.json +27 -0
  32. trained_models/checkpoints_rag_fr/model_epoch3.pt +3 -0
  33. trained_models/checkpoints_rag_fr/model_epoch4.pt +3 -0
  34. trained_models/checkpoints_rag_fr/tokenizer.json +0 -0
  35. trained_models/checkpoints_rag_fr/tokenizer_config.json +20 -0
  36. trained_models/checkpoints_rag_fr/training_info.json +7 -0
  37. trained_models/checkpoints_rag_fr_20260428_174120/history.json +27 -0
  38. trained_models/checkpoints_rag_fr_20260428_174120/tokenizer.json +0 -0
  39. trained_models/checkpoints_rag_fr_20260428_174120/tokenizer_config.json +20 -0
  40. trained_models/checkpoints_rag_fr_20260428_174120/training_info.json +7 -0
  41. trained_models/checkpoints_rag_fr_v2/history.json +82 -0
  42. trained_models/checkpoints_rag_fr_v2/tokenizer.json +0 -0
  43. trained_models/checkpoints_rag_fr_v2/tokenizer_config.json +20 -0
  44. trained_models/checkpoints_rag_fr_v2/training_info.json +7 -0
  45. trained_models/checkpoints_rag_fr_v2_20260428_174118/history.json +82 -0
  46. trained_models/checkpoints_rag_fr_v2_20260428_174118/tokenizer.json +0 -0
  47. trained_models/checkpoints_rag_fr_v2_20260428_174118/tokenizer_config.json +20 -0
  48. trained_models/checkpoints_rag_fr_v2_20260428_174118/training_info.json +7 -0
  49. trained_models/checkpoints_rag_fr_v2_continue/history_continue.json +42 -0
  50. trained_models/checkpoints_rag_fr_v2_continue/tokenizer.json +0 -0
trained_models/RAGHUB_orchestratoor/MODEL_NAME.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ RAGHUB orchestratoor
trained_models/RAGHUB_orchestratoor/config_train.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "trained_model_name": "RAGHUB orchestratoor",
3
+ "model_name": "camembert-base",
4
+ "output_root": "/workspace/trained_models",
5
+ "run_name": "RAGHUB_orchestratoor",
6
+ "max_seq_len": 192,
7
+ "batch_size": 64,
8
+ "grad_accum_steps": 2,
9
+ "epochs": 10,
10
+ "lr": 2e-05,
11
+ "weight_decay": 0.01,
12
+ "warmup_ratio": 0.05,
13
+ "temperature": 0.05,
14
+ "grad_clip": 1.0,
15
+ "seed": 42,
16
+ "num_workers": 4,
17
+ "max_mmarco": 120000,
18
+ "max_piaf": 10000,
19
+ "max_xnli": 50000,
20
+ "max_french_sts": 20000,
21
+ "eval_size": 2000,
22
+ "use_bf16": true,
23
+ "use_compile": false,
24
+ "save_every_epoch": true
25
+ }
trained_models/RAGHUB_orchestratoor/history.json ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "epoch": 1,
4
+ "train_loss": 0.49692287032412313,
5
+ "train_acc": 0.9029349035608308,
6
+ "R@1": 0.9005355776587605,
7
+ "R@5": 0.9693955623565417,
8
+ "R@10": 0.981637337413925,
9
+ "MRR": 0.9318775149989658
10
+ },
11
+ {
12
+ "epoch": 2,
13
+ "train_loss": 0.07072129767548392,
14
+ "train_acc": 0.9831695103857567,
15
+ "R@1": 0.9143075745983168,
16
+ "R@5": 0.9801071155317521,
17
+ "R@10": 0.9885233358837031,
18
+ "MRR": 0.9435282144423083
19
+ },
20
+ {
21
+ "epoch": 3,
22
+ "train_loss": 0.04076581896843737,
23
+ "train_acc": 0.9912833827893175,
24
+ "R@1": 0.9143075745983168,
25
+ "R@5": 0.9824024483550114,
26
+ "R@10": 0.9892884468247896,
27
+ "MRR": 0.9442865113082207
28
+ },
29
+ {
30
+ "epoch": 4,
31
+ "train_loss": 0.026430810675021302,
32
+ "train_acc": 0.9948843966369931,
33
+ "R@1": 0.9188982402448355,
34
+ "R@5": 0.9839326702371843,
35
+ "R@10": 0.9892884468247896,
36
+ "MRR": 0.9470647398103765
37
+ },
38
+ {
39
+ "epoch": 5,
40
+ "train_loss": 0.019742724098795777,
41
+ "train_acc": 0.9964298961424333,
42
+ "R@1": 0.9219586840091814,
43
+ "R@5": 0.9846977811782709,
44
+ "R@10": 0.990053557765876,
45
+ "MRR": 0.9497522837794463
46
+ },
47
+ {
48
+ "epoch": 6,
49
+ "train_loss": 0.01529984119275454,
50
+ "train_acc": 0.9975426557863502,
51
+ "R@1": 0.9234889058913542,
52
+ "R@5": 0.9854628921193573,
53
+ "R@10": 0.9885233358837031,
54
+ "MRR": 0.9510699586252107
55
+ },
56
+ {
57
+ "epoch": 7,
58
+ "train_loss": 0.013311921477476254,
59
+ "train_acc": 0.9977203882294757,
60
+ "R@1": 0.9273144605967866,
61
+ "R@5": 0.9846977811782709,
62
+ "R@10": 0.990053557765876,
63
+ "MRR": 0.9531188814479907
64
+ },
65
+ {
66
+ "epoch": 8,
67
+ "train_loss": 0.011790111897703036,
68
+ "train_acc": 0.9979290306627102,
69
+ "R@1": 0.9273144605967866,
70
+ "R@5": 0.9854628921193573,
71
+ "R@10": 0.9908186687069626,
72
+ "MRR": 0.9534072144031361
73
+ },
74
+ {
75
+ "epoch": 9,
76
+ "train_loss": 0.011065685890364423,
77
+ "train_acc": 0.9984622279920871,
78
+ "R@1": 0.9288446824789595,
79
+ "R@5": 0.9854628921193573,
80
+ "R@10": 0.9908186687069626,
81
+ "MRR": 0.9542217896295068
82
+ },
83
+ {
84
+ "epoch": 10,
85
+ "train_loss": 0.01090054404456629,
86
+ "train_acc": 0.9981299455984174,
87
+ "R@1": 0.9303749043611323,
88
+ "R@5": 0.9854628921193573,
89
+ "R@10": 0.9908186687069626,
90
+ "MRR": 0.9550046170280994
91
+ }
92
+ ]
trained_models/RAGHUB_orchestratoor/model_info.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "trained_model_name": "RAGHUB orchestratoor",
3
+ "run_name": "RAGHUB_orchestratoor",
4
+ "base_model_name": "camembert-base",
5
+ "epoch": 10,
6
+ "best": true,
7
+ "metrics": {
8
+ "R@1": 0.9303749043611323,
9
+ "R@5": 0.9854628921193573,
10
+ "R@10": 0.9908186687069626,
11
+ "MRR": 0.9550046170280994
12
+ },
13
+ "files": {
14
+ "best_checkpoint": "model_best.pt",
15
+ "epoch_checkpoint": "model_epoch_10.pt",
16
+ "tokenizer": "tokenizer.json",
17
+ "training_config": "config_train.json",
18
+ "history": "history.json"
19
+ }
20
+ }
trained_models/RAGHUB_orchestratoor/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
trained_models/RAGHUB_orchestratoor/tokenizer_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<s>",
5
+ "cls_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "extra_special_tokens": [
8
+ "<s>NOTUSED",
9
+ "</s>NOTUSED",
10
+ "<unk>NOTUSED"
11
+ ],
12
+ "is_local": false,
13
+ "local_files_only": false,
14
+ "mask_token": "<mask>",
15
+ "model_max_length": 512,
16
+ "pad_token": "<pad>",
17
+ "sep_token": "</s>",
18
+ "tokenizer_class": "CamembertTokenizer",
19
+ "unk_token": "<unk>"
20
+ }
trained_models/RAGHUB_orchestratoor/training_info.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "latest_checkpoint": "trained_models/RAGHUB_orchestratoor/model_best.pt",
3
+ "latest_mtime": 1777461702.8608189,
4
+ "latest_mtime_iso": "2026-04-29T11:21:42.860819+00:00",
5
+ "size_bytes": 444935514,
6
+ "epoch": null
7
+ }
trained_models/RAGHUB_orchestrator_full_v1/MODEL_NAME.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ RAGHUB orchestrator v4 full
trained_models/RAGHUB_orchestrator_full_v1/config_train.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "trained_model_name": "RAGHUB orchestrator v4 full",
3
+ "model_name": "camembert-base",
4
+ "output_root": "/workspace/trained_models",
5
+ "run_name": "RAGHUB_orchestrator_full_v1",
6
+ "max_seq_len": 256,
7
+ "batch_size": 32,
8
+ "grad_accum_steps": 2,
9
+ "epochs": 3,
10
+ "lr": 1e-05,
11
+ "weight_decay": 0.01,
12
+ "warmup_ratio": 0.05,
13
+ "temperature": 0.05,
14
+ "grad_clip": 1.0,
15
+ "seed": 42,
16
+ "num_workers": 4,
17
+ "max_mmarco": 0,
18
+ "max_piaf": 0,
19
+ "max_xnli": 0,
20
+ "max_french_sts": 0,
21
+ "eval_size": 2000,
22
+ "use_bf16": true,
23
+ "use_compile": false,
24
+ "save_every_epoch": true,
25
+ "hf_pair_dataset": "ibm-research/REAL-MM-RAG_FinTabTrainSet_rephrased",
26
+ "hf_pair_config": "",
27
+ "hf_pair_split": "train",
28
+ "hf_anchor_col": "query",
29
+ "hf_positive_col": "answer",
30
+ "hf_negative_col": "",
31
+ "expert_pairs_path": "",
32
+ "resume_checkpoint": ""
33
+ }
trained_models/RAGHUB_orchestrator_full_v1/history.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "epoch": 1,
4
+ "train_loss": 2.332637549658953,
5
+ "train_acc": 0.3234533898305085,
6
+ "R@1": 0.2222222222222222,
7
+ "R@5": 0.36268343815513626,
8
+ "R@10": 0.4360587002096436,
9
+ "MRR": 0.2964828550868599
10
+ },
11
+ {
12
+ "epoch": 2,
13
+ "train_loss": 1.7514312114553936,
14
+ "train_acc": 0.45199152542372883,
15
+ "R@1": 0.24109014675052412,
16
+ "R@5": 0.3689727463312369,
17
+ "R@10": 0.480083857442348,
18
+ "MRR": 0.3204429044688647
19
+ },
20
+ {
21
+ "epoch": 3,
22
+ "train_loss": 1.6255126011169563,
23
+ "train_acc": 0.48627118644067796,
24
+ "R@1": 0.24528301886792453,
25
+ "R@5": 0.39622641509433965,
26
+ "R@10": 0.4863731656184486,
27
+ "MRR": 0.325460651386674
28
+ }
29
+ ]
trained_models/RAGHUB_orchestrator_full_v1/model_info.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "trained_model_name": "RAGHUB orchestrator v4 full",
3
+ "run_name": "RAGHUB_orchestrator_full_v1",
4
+ "base_model_name": "camembert-base",
5
+ "epoch": 3,
6
+ "best": true,
7
+ "metrics": {
8
+ "R@1": 0.24528301886792453,
9
+ "R@5": 0.39622641509433965,
10
+ "R@10": 0.4863731656184486,
11
+ "MRR": 0.325460651386674
12
+ },
13
+ "files": {
14
+ "best_checkpoint": "model_best.pt",
15
+ "epoch_checkpoint": "model_epoch_3.pt",
16
+ "tokenizer": "tokenizer.json",
17
+ "training_config": "config_train.json",
18
+ "history": "history.json"
19
+ }
20
+ }
trained_models/RAGHUB_orchestrator_full_v1/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
trained_models/RAGHUB_orchestrator_full_v1/tokenizer_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<s>",
5
+ "cls_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "extra_special_tokens": [
8
+ "<s>NOTUSED",
9
+ "</s>NOTUSED",
10
+ "<unk>NOTUSED"
11
+ ],
12
+ "is_local": false,
13
+ "local_files_only": false,
14
+ "mask_token": "<mask>",
15
+ "model_max_length": 512,
16
+ "pad_token": "<pad>",
17
+ "sep_token": "</s>",
18
+ "tokenizer_class": "CamembertTokenizer",
19
+ "unk_token": "<unk>"
20
+ }
trained_models/RAGHUB_orchestrator_full_v1/training_info.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "latest_checkpoint": "trained_models/RAGHUB_orchestrator_full_v1/model_best.pt",
3
+ "latest_mtime": 1777502643.0225697,
4
+ "latest_mtime_iso": "2026-04-29T22:44:03.022570+00:00",
5
+ "size_bytes": 444935834,
6
+ "epoch": null
7
+ }
trained_models/RAGHUB_orchestrator_mixed_v1/MODEL_NAME.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ RAGHUB orchestrator v3 mixed
trained_models/RAGHUB_orchestrator_mixed_v1/config_train.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "trained_model_name": "RAGHUB orchestrator v3 mixed",
3
+ "model_name": "camembert-base",
4
+ "output_root": "/workspace/trained_models",
5
+ "run_name": "RAGHUB_orchestrator_mixed_v1",
6
+ "max_seq_len": 256,
7
+ "batch_size": 32,
8
+ "grad_accum_steps": 2,
9
+ "epochs": 5,
10
+ "lr": 1e-05,
11
+ "weight_decay": 0.01,
12
+ "warmup_ratio": 0.05,
13
+ "temperature": 0.05,
14
+ "grad_clip": 1.0,
15
+ "seed": 42,
16
+ "num_workers": 4,
17
+ "max_mmarco": 120000,
18
+ "max_piaf": 10000,
19
+ "max_xnli": 50000,
20
+ "max_french_sts": 20000,
21
+ "eval_size": 2000,
22
+ "use_bf16": true,
23
+ "use_compile": false,
24
+ "save_every_epoch": true,
25
+ "hf_pair_dataset": "ibm-research/REAL-MM-RAG_FinTabTrainSet_rephrased",
26
+ "hf_pair_config": "",
27
+ "hf_pair_split": "train",
28
+ "hf_anchor_col": "query",
29
+ "hf_positive_col": "answer",
30
+ "hf_negative_col": "",
31
+ "expert_pairs_path": "",
32
+ "resume_checkpoint": ""
33
+ }
trained_models/RAGHUB_orchestrator_mixed_v1/history.json ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "epoch": 1,
4
+ "train_loss": 0.5679403888953999,
5
+ "train_acc": 0.821861981525086,
6
+ "R@1": 0.7303811659192825,
7
+ "R@5": 0.8183856502242153,
8
+ "R@10": 0.8469730941704036,
9
+ "MRR": 0.7722298411034937
10
+ },
11
+ {
12
+ "epoch": 2,
13
+ "train_loss": 0.2541170532207378,
14
+ "train_acc": 0.9061198152508604,
15
+ "R@1": 0.7449551569506726,
16
+ "R@5": 0.8363228699551569,
17
+ "R@10": 0.8693946188340808,
18
+ "MRR": 0.7893742970627043
19
+ },
20
+ {
21
+ "epoch": 3,
22
+ "train_loss": 0.2047349282629279,
23
+ "train_acc": 0.9242381362072088,
24
+ "R@1": 0.7483183856502242,
25
+ "R@5": 0.8452914798206278,
26
+ "R@10": 0.8783632286995515,
27
+ "MRR": 0.7942090176031011
28
+ },
29
+ {
30
+ "epoch": 4,
31
+ "train_loss": 0.1836437827701198,
32
+ "train_acc": 0.9320605415685564,
33
+ "R@1": 0.75,
34
+ "R@5": 0.8469730941704036,
35
+ "R@10": 0.8789237668161435,
36
+ "MRR": 0.7960958088015041
37
+ },
38
+ {
39
+ "epoch": 5,
40
+ "train_loss": 0.17780673120449503,
41
+ "train_acc": 0.9336114381452636,
42
+ "R@1": 0.75,
43
+ "R@5": 0.8469730941704036,
44
+ "R@10": 0.8778026905829597,
45
+ "MRR": 0.7967083589364699
46
+ }
47
+ ]
trained_models/RAGHUB_orchestrator_mixed_v1/model_info.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "trained_model_name": "RAGHUB orchestrator v3 mixed",
3
+ "run_name": "RAGHUB_orchestrator_mixed_v1",
4
+ "base_model_name": "camembert-base",
5
+ "epoch": 5,
6
+ "best": true,
7
+ "metrics": {
8
+ "R@1": 0.75,
9
+ "R@5": 0.8469730941704036,
10
+ "R@10": 0.8778026905829597,
11
+ "MRR": 0.7967083589364699
12
+ },
13
+ "files": {
14
+ "best_checkpoint": "model_best.pt",
15
+ "epoch_checkpoint": "model_epoch_5.pt",
16
+ "tokenizer": "tokenizer.json",
17
+ "training_config": "config_train.json",
18
+ "history": "history.json"
19
+ }
20
+ }
trained_models/RAGHUB_orchestrator_mixed_v1/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
trained_models/RAGHUB_orchestrator_mixed_v1/tokenizer_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<s>",
5
+ "cls_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "extra_special_tokens": [
8
+ "<s>NOTUSED",
9
+ "</s>NOTUSED",
10
+ "<unk>NOTUSED"
11
+ ],
12
+ "is_local": false,
13
+ "local_files_only": false,
14
+ "mask_token": "<mask>",
15
+ "model_max_length": 512,
16
+ "pad_token": "<pad>",
17
+ "sep_token": "</s>",
18
+ "tokenizer_class": "CamembertTokenizer",
19
+ "unk_token": "<unk>"
20
+ }
trained_models/RAGHUB_orchestrator_mixed_v1/training_info.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "latest_checkpoint": "trained_models/RAGHUB_orchestrator_mixed_v1/model_best.pt",
3
+ "latest_mtime": 1777500568.5610309,
4
+ "latest_mtime_iso": "2026-04-29T22:09:28.561031+00:00",
5
+ "size_bytes": 444935834,
6
+ "epoch": null
7
+ }
trained_models/RAGHUB_orchestrator_v1_Yassino/MODEL_NAME.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ RAGHUB orchestrator v2 YASSINO
trained_models/RAGHUB_orchestrator_v1_Yassino/config_train.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "trained_model_name": "RAGHUB orchestrator v2 YASSINO",
3
+ "model_name": "camembert-base",
4
+ "output_root": "/workspace/trained_models",
5
+ "run_name": "RAGHUB_orchestrator_v1_Yassino",
6
+ "max_seq_len": 192,
7
+ "batch_size": 64,
8
+ "grad_accum_steps": 2,
9
+ "epochs": 6,
10
+ "lr": 5e-06,
11
+ "weight_decay": 0.01,
12
+ "warmup_ratio": 0.05,
13
+ "temperature": 0.05,
14
+ "grad_clip": 1.0,
15
+ "seed": 42,
16
+ "num_workers": 4,
17
+ "max_mmarco": 120000,
18
+ "max_piaf": 10000,
19
+ "max_xnli": 50000,
20
+ "max_french_sts": 20000,
21
+ "eval_size": 2000,
22
+ "use_bf16": true,
23
+ "use_compile": false,
24
+ "save_every_epoch": true,
25
+ "hf_pair_dataset": "ibm-research/REAL-MM-RAG_FinTabTrainSet_rephrased",
26
+ "hf_pair_config": "",
27
+ "hf_pair_split": "train",
28
+ "hf_anchor_col": "query",
29
+ "hf_positive_col": "answer",
30
+ "hf_negative_col": "",
31
+ "expert_pairs_path": "",
32
+ "resume_checkpoint": "/workspace/trained_models/RAGHUB_orchestratoor/model_best.pt"
33
+ }
trained_models/RAGHUB_orchestrator_v1_Yassino/history.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "epoch": 1,
4
+ "train_loss": 0.5027125855745829,
5
+ "train_acc": 0.8411911231884058,
6
+ "R@1": 0.7982062780269058,
7
+ "R@5": 0.8447309417040358,
8
+ "R@10": 0.8604260089686099,
9
+ "MRR": 0.8224298701955011
10
+ },
11
+ {
12
+ "epoch": 2,
13
+ "train_loss": 0.37192658019670544,
14
+ "train_acc": 0.8724411231884058,
15
+ "R@1": 0.8015695067264574,
16
+ "R@5": 0.8565022421524664,
17
+ "R@10": 0.8721973094170403,
18
+ "MRR": 0.8285277030475686
19
+ },
20
+ {
21
+ "epoch": 3,
22
+ "train_loss": 0.3293720392132367,
23
+ "train_acc": 0.8853430706521739,
24
+ "R@1": 0.8021300448430493,
25
+ "R@5": 0.8570627802690582,
26
+ "R@10": 0.8789237668161435,
27
+ "MRR": 0.8305888642021839
28
+ },
29
+ {
30
+ "epoch": 4,
31
+ "train_loss": 0.3057551719356274,
32
+ "train_acc": 0.8932461503623188,
33
+ "R@1": 0.8038116591928252,
34
+ "R@5": 0.859304932735426,
35
+ "R@10": 0.8839686098654709,
36
+ "MRR": 0.8329696671339848
37
+ },
38
+ {
39
+ "epoch": 5,
40
+ "train_loss": 0.294394488356994,
41
+ "train_acc": 0.8970052083333333,
42
+ "R@1": 0.8082959641255605,
43
+ "R@5": 0.8621076233183856,
44
+ "R@10": 0.8845291479820628,
45
+ "MRR": 0.8359274838702688
46
+ },
47
+ {
48
+ "epoch": 6,
49
+ "train_loss": 0.290229741419814,
50
+ "train_acc": 0.8980695199275363,
51
+ "R@1": 0.8066143497757847,
52
+ "R@5": 0.8632286995515696,
53
+ "R@10": 0.8822869955156951,
54
+ "MRR": 0.8348901612009362
55
+ }
56
+ ]
trained_models/RAGHUB_orchestrator_v1_Yassino/model_info.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "trained_model_name": "RAGHUB orchestrator v2 YASSINO",
3
+ "run_name": "RAGHUB_orchestrator_v1_Yassino",
4
+ "base_model_name": "camembert-base",
5
+ "epoch": 6,
6
+ "best": false,
7
+ "metrics": {
8
+ "R@1": 0.8066143497757847,
9
+ "R@5": 0.8632286995515696,
10
+ "R@10": 0.8822869955156951,
11
+ "MRR": 0.8348901612009362
12
+ },
13
+ "files": {
14
+ "best_checkpoint": "model_best.pt",
15
+ "epoch_checkpoint": "model_epoch_6.pt",
16
+ "tokenizer": "tokenizer.json",
17
+ "training_config": "config_train.json",
18
+ "history": "history.json"
19
+ }
20
+ }
trained_models/RAGHUB_orchestrator_v1_Yassino/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
trained_models/RAGHUB_orchestrator_v1_Yassino/tokenizer_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<s>",
5
+ "cls_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "extra_special_tokens": [
8
+ "<s>NOTUSED",
9
+ "</s>NOTUSED",
10
+ "<unk>NOTUSED"
11
+ ],
12
+ "is_local": false,
13
+ "local_files_only": false,
14
+ "mask_token": "<mask>",
15
+ "model_max_length": 512,
16
+ "pad_token": "<pad>",
17
+ "sep_token": "</s>",
18
+ "tokenizer_class": "CamembertTokenizer",
19
+ "unk_token": "<unk>"
20
+ }
trained_models/RAGHUB_orchestrator_v1_Yassino/training_info.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "latest_checkpoint": "trained_models/RAGHUB_orchestrator_v1_Yassino/model_epoch_6.pt",
3
+ "latest_mtime": 1777489920.16389,
4
+ "latest_mtime_iso": "2026-04-29T19:12:00.163890+00:00",
5
+ "size_bytes": 444936519,
6
+ "epoch": 6
7
+ }
trained_models/RAGHUB_orchestrator_v2_expert/MODEL_NAME.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ RAGHUB orchestrator v2 expert
trained_models/RAGHUB_orchestrator_v2_expert/training_info.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "latest_checkpoint": "trained_models/RAGHUB_orchestrator_v2_expert/model_best.pt",
3
+ "latest_mtime": 1777466921.7961936,
4
+ "latest_mtime_iso": "2026-04-29T12:48:41.796194+00:00",
5
+ "size_bytes": 444935514,
6
+ "epoch": null
7
+ }
trained_models/checkpoints_rag_fr/history.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "epoch": 1,
4
+ "eval_loss": 0.9426132928241383,
5
+ "eval_acc": 0.7903251268646934
6
+ },
7
+ {
8
+ "epoch": 2,
9
+ "eval_loss": 0.7285588058558378,
10
+ "eval_acc": 0.8373141126199202
11
+ },
12
+ {
13
+ "epoch": 3,
14
+ "eval_loss": 0.6514262557029724,
15
+ "eval_acc": 0.8508084308017384
16
+ },
17
+ {
18
+ "epoch": 4,
19
+ "eval_loss": 0.5993350066921928,
20
+ "eval_acc": 0.86045349186117
21
+ },
22
+ {
23
+ "epoch": 5,
24
+ "eval_loss": 0.5905819237232208,
25
+ "eval_acc": 0.8625841736793518
26
+ }
27
+ ]
trained_models/checkpoints_rag_fr/model_epoch3.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d3892bf3b605068b0d21d30cc330a09bec471045c0a6eb2aab2c2d17b28c0218
3
+ size 93358821
trained_models/checkpoints_rag_fr/model_epoch4.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e89267c048314a57377a9ddcdb34069745a9520a209ffa49c79f9629eb5d4a91
3
+ size 93358821
trained_models/checkpoints_rag_fr/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
trained_models/checkpoints_rag_fr/tokenizer_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<s>",
5
+ "cls_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "extra_special_tokens": [
8
+ "<s>NOTUSED",
9
+ "</s>NOTUSED",
10
+ "<unk>NOTUSED"
11
+ ],
12
+ "is_local": false,
13
+ "local_files_only": false,
14
+ "mask_token": "<mask>",
15
+ "model_max_length": 512,
16
+ "pad_token": "<pad>",
17
+ "sep_token": "</s>",
18
+ "tokenizer_class": "CamembertTokenizer",
19
+ "unk_token": "<unk>"
20
+ }
trained_models/checkpoints_rag_fr/training_info.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "latest_checkpoint": "trained_models/checkpoints_rag_fr/model_epoch5.pt",
3
+ "latest_mtime": 1777372635.702895,
4
+ "latest_mtime_iso": "2026-04-28T10:37:15.702895+00:00",
5
+ "size_bytes": 93358821,
6
+ "epoch": 5
7
+ }
trained_models/checkpoints_rag_fr_20260428_174120/history.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "epoch": 1,
4
+ "eval_loss": 0.9426132928241383,
5
+ "eval_acc": 0.7903251268646934
6
+ },
7
+ {
8
+ "epoch": 2,
9
+ "eval_loss": 0.7285588058558378,
10
+ "eval_acc": 0.8373141126199202
11
+ },
12
+ {
13
+ "epoch": 3,
14
+ "eval_loss": 0.6514262557029724,
15
+ "eval_acc": 0.8508084308017384
16
+ },
17
+ {
18
+ "epoch": 4,
19
+ "eval_loss": 0.5993350066921928,
20
+ "eval_acc": 0.86045349186117
21
+ },
22
+ {
23
+ "epoch": 5,
24
+ "eval_loss": 0.5905819237232208,
25
+ "eval_acc": 0.8625841736793518
26
+ }
27
+ ]
trained_models/checkpoints_rag_fr_20260428_174120/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
trained_models/checkpoints_rag_fr_20260428_174120/tokenizer_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<s>",
5
+ "cls_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "extra_special_tokens": [
8
+ "<s>NOTUSED",
9
+ "</s>NOTUSED",
10
+ "<unk>NOTUSED"
11
+ ],
12
+ "is_local": false,
13
+ "local_files_only": false,
14
+ "mask_token": "<mask>",
15
+ "model_max_length": 512,
16
+ "pad_token": "<pad>",
17
+ "sep_token": "</s>",
18
+ "tokenizer_class": "CamembertTokenizer",
19
+ "unk_token": "<unk>"
20
+ }
trained_models/checkpoints_rag_fr_20260428_174120/training_info.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "latest_checkpoint": "trained_models/checkpoints_rag_fr_20260428_174120/model_epoch5.pt",
3
+ "latest_mtime": 1777372635.702895,
4
+ "latest_mtime_iso": "2026-04-28T10:37:15.702895+00:00",
5
+ "size_bytes": 93358821,
6
+ "epoch": 5
7
+ }
trained_models/checkpoints_rag_fr_v2/history.json ADDED
@@ -0,0 +1,82 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "epoch": 1,
4
+ "R@1": 0.1504039764404297,
5
+ "R@5": 0.21006835997104645,
6
+ "R@10": 0.23492851853370667,
7
+ "MRR": 0.18460102379322052,
8
+ "train_loss": 3.6289227961731565
9
+ },
10
+ {
11
+ "epoch": 2,
12
+ "R@1": 0.39651957154273987,
13
+ "R@5": 0.5288999080657959,
14
+ "R@10": 0.5811062455177307,
15
+ "MRR": 0.4589481055736542,
16
+ "train_loss": 1.7728186886754633
17
+ },
18
+ {
19
+ "epoch": 3,
20
+ "R@1": 0.5376009941101074,
21
+ "R@5": 0.6743319034576416,
22
+ "R@10": 0.7165941596031189,
23
+ "MRR": 0.6019946932792664,
24
+ "train_loss": 1.2733176767016106
25
+ },
26
+ {
27
+ "epoch": 4,
28
+ "R@1": 0.6239900588989258,
29
+ "R@5": 0.7582349181175232,
30
+ "R@10": 0.8048477172851562,
31
+ "MRR": 0.6874814629554749,
32
+ "train_loss": 0.9967152672134143
33
+ },
34
+ {
35
+ "epoch": 5,
36
+ "R@1": 0.6693598628044128,
37
+ "R@5": 0.7986326813697815,
38
+ "R@10": 0.8427594900131226,
39
+ "MRR": 0.7317501306533813,
40
+ "train_loss": 0.8231126829493437
41
+ },
42
+ {
43
+ "epoch": 6,
44
+ "R@1": 0.7016780376434326,
45
+ "R@5": 0.8216283321380615,
46
+ "R@10": 0.8651336431503296,
47
+ "MRR": 0.7590409517288208,
48
+ "train_loss": 0.7098938351054843
49
+ },
50
+ {
51
+ "epoch": 7,
52
+ "R@1": 0.7153511643409729,
53
+ "R@5": 0.835301399230957,
54
+ "R@10": 0.8744561672210693,
55
+ "MRR": 0.7721759080886841,
56
+ "train_loss": 0.6391438583104159
57
+ },
58
+ {
59
+ "epoch": 8,
60
+ "R@1": 0.7203232049942017,
61
+ "R@5": 0.8464884757995605,
62
+ "R@10": 0.8868862390518188,
63
+ "MRR": 0.7790595293045044,
64
+ "train_loss": 0.5962411920181948
65
+ },
66
+ {
67
+ "epoch": 9,
68
+ "R@1": 0.7240521907806396,
69
+ "R@5": 0.8495960235595703,
70
+ "R@10": 0.8881292939186096,
71
+ "MRR": 0.7832603454589844,
72
+ "train_loss": 0.5732792998722893
73
+ },
74
+ {
75
+ "epoch": 10,
76
+ "R@1": 0.7259166836738586,
77
+ "R@5": 0.8533250093460083,
78
+ "R@10": 0.8912367820739746,
79
+ "MRR": 0.7857035398483276,
80
+ "train_loss": 0.5627455476233714
81
+ }
82
+ ]
trained_models/checkpoints_rag_fr_v2/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
trained_models/checkpoints_rag_fr_v2/tokenizer_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<s>",
5
+ "cls_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "extra_special_tokens": [
8
+ "<s>NOTUSED",
9
+ "</s>NOTUSED",
10
+ "<unk>NOTUSED"
11
+ ],
12
+ "is_local": false,
13
+ "local_files_only": false,
14
+ "mask_token": "<mask>",
15
+ "model_max_length": 512,
16
+ "pad_token": "<pad>",
17
+ "sep_token": "</s>",
18
+ "tokenizer_class": "CamembertTokenizer",
19
+ "unk_token": "<unk>"
20
+ }
trained_models/checkpoints_rag_fr_v2/training_info.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "latest_checkpoint": "trained_models/checkpoints_rag_fr_v2/model_best.pt",
3
+ "latest_mtime": 1777378177.9374852,
4
+ "latest_mtime_iso": "2026-04-28T12:09:37.937485+00:00",
5
+ "size_bytes": 93378357,
6
+ "epoch": null
7
+ }
trained_models/checkpoints_rag_fr_v2_20260428_174118/history.json ADDED
@@ -0,0 +1,82 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "epoch": 1,
4
+ "R@1": 0.1504039764404297,
5
+ "R@5": 0.21006835997104645,
6
+ "R@10": 0.23492851853370667,
7
+ "MRR": 0.18460102379322052,
8
+ "train_loss": 3.6289227961731565
9
+ },
10
+ {
11
+ "epoch": 2,
12
+ "R@1": 0.39651957154273987,
13
+ "R@5": 0.5288999080657959,
14
+ "R@10": 0.5811062455177307,
15
+ "MRR": 0.4589481055736542,
16
+ "train_loss": 1.7728186886754633
17
+ },
18
+ {
19
+ "epoch": 3,
20
+ "R@1": 0.5376009941101074,
21
+ "R@5": 0.6743319034576416,
22
+ "R@10": 0.7165941596031189,
23
+ "MRR": 0.6019946932792664,
24
+ "train_loss": 1.2733176767016106
25
+ },
26
+ {
27
+ "epoch": 4,
28
+ "R@1": 0.6239900588989258,
29
+ "R@5": 0.7582349181175232,
30
+ "R@10": 0.8048477172851562,
31
+ "MRR": 0.6874814629554749,
32
+ "train_loss": 0.9967152672134143
33
+ },
34
+ {
35
+ "epoch": 5,
36
+ "R@1": 0.6693598628044128,
37
+ "R@5": 0.7986326813697815,
38
+ "R@10": 0.8427594900131226,
39
+ "MRR": 0.7317501306533813,
40
+ "train_loss": 0.8231126829493437
41
+ },
42
+ {
43
+ "epoch": 6,
44
+ "R@1": 0.7016780376434326,
45
+ "R@5": 0.8216283321380615,
46
+ "R@10": 0.8651336431503296,
47
+ "MRR": 0.7590409517288208,
48
+ "train_loss": 0.7098938351054843
49
+ },
50
+ {
51
+ "epoch": 7,
52
+ "R@1": 0.7153511643409729,
53
+ "R@5": 0.835301399230957,
54
+ "R@10": 0.8744561672210693,
55
+ "MRR": 0.7721759080886841,
56
+ "train_loss": 0.6391438583104159
57
+ },
58
+ {
59
+ "epoch": 8,
60
+ "R@1": 0.7203232049942017,
61
+ "R@5": 0.8464884757995605,
62
+ "R@10": 0.8868862390518188,
63
+ "MRR": 0.7790595293045044,
64
+ "train_loss": 0.5962411920181948
65
+ },
66
+ {
67
+ "epoch": 9,
68
+ "R@1": 0.7240521907806396,
69
+ "R@5": 0.8495960235595703,
70
+ "R@10": 0.8881292939186096,
71
+ "MRR": 0.7832603454589844,
72
+ "train_loss": 0.5732792998722893
73
+ },
74
+ {
75
+ "epoch": 10,
76
+ "R@1": 0.7259166836738586,
77
+ "R@5": 0.8533250093460083,
78
+ "R@10": 0.8912367820739746,
79
+ "MRR": 0.7857035398483276,
80
+ "train_loss": 0.5627455476233714
81
+ }
82
+ ]
trained_models/checkpoints_rag_fr_v2_20260428_174118/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
trained_models/checkpoints_rag_fr_v2_20260428_174118/tokenizer_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<s>",
5
+ "cls_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "extra_special_tokens": [
8
+ "<s>NOTUSED",
9
+ "</s>NOTUSED",
10
+ "<unk>NOTUSED"
11
+ ],
12
+ "is_local": false,
13
+ "local_files_only": false,
14
+ "mask_token": "<mask>",
15
+ "model_max_length": 512,
16
+ "pad_token": "<pad>",
17
+ "sep_token": "</s>",
18
+ "tokenizer_class": "CamembertTokenizer",
19
+ "unk_token": "<unk>"
20
+ }
trained_models/checkpoints_rag_fr_v2_20260428_174118/training_info.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "latest_checkpoint": "trained_models/checkpoints_rag_fr_v2_20260428_174118/model_best.pt",
3
+ "latest_mtime": 1777378177.9374852,
4
+ "latest_mtime_iso": "2026-04-28T12:09:37.937485+00:00",
5
+ "size_bytes": 93378357,
6
+ "epoch": null
7
+ }
trained_models/checkpoints_rag_fr_v2_continue/history_continue.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "continue_epoch": 1,
4
+ "R@1": 0.7271597385406494,
5
+ "R@5": 0.8551895618438721,
6
+ "R@10": 0.8918582797050476,
7
+ "MRR": 0.7866597175598145,
8
+ "train_loss": 0.5913846994237267
9
+ },
10
+ {
11
+ "continue_epoch": 2,
12
+ "R@1": 0.7290242314338684,
13
+ "R@5": 0.8551895618438721,
14
+ "R@10": 0.8943443298339844,
15
+ "MRR": 0.7887516021728516,
16
+ "train_loss": 0.5500548621542829
17
+ },
18
+ {
19
+ "continue_epoch": 3,
20
+ "R@1": 0.7346177697181702,
21
+ "R@5": 0.8570540547370911,
22
+ "R@10": 0.8955873250961304,
23
+ "MRR": 0.7928776741027832,
24
+ "train_loss": 0.5177965851234113
25
+ },
26
+ {
27
+ "continue_epoch": 4,
28
+ "R@1": 0.7371037602424622,
29
+ "R@5": 0.8595401048660278,
30
+ "R@10": 0.8962088227272034,
31
+ "MRR": 0.7950453758239746,
32
+ "train_loss": 0.4953275489435493
33
+ },
34
+ {
35
+ "continue_epoch": 5,
36
+ "R@1": 0.7395898103713989,
37
+ "R@5": 0.8620260953903198,
38
+ "R@10": 0.8974518179893494,
39
+ "MRR": 0.7968562245368958,
40
+ "train_loss": 0.48535463792814626
41
+ }
42
+ ]
trained_models/checkpoints_rag_fr_v2_continue/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff