96abhishekarora commited on
Commit
ba6f1fe
1 Parent(s): e704e62

Modified validation and training for linktransformer model

Browse files
.gitattributes CHANGED
@@ -36,3 +36,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
36
  pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
37
  .git/lfs/objects/ac/e3/ace3c71b97a96f01b2feceec24958b690700ea37ad188af305b32ef472101546 filter=lfs diff=lfs merge=lfs -text
38
  .git/lfs/objects/a1/ca/a1caf77aeeb33f22a280e7dd98da99969ecbc0b510530aacd240d85af85b27f5 filter=lfs diff=lfs merge=lfs -text
 
 
 
36
  pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
37
  .git/lfs/objects/ac/e3/ace3c71b97a96f01b2feceec24958b690700ea37ad188af305b32ef472101546 filter=lfs diff=lfs merge=lfs -text
38
  .git/lfs/objects/a1/ca/a1caf77aeeb33f22a280e7dd98da99969ecbc0b510530aacd240d85af85b27f5 filter=lfs diff=lfs merge=lfs -text
39
+ model.safetensors filter=lfs diff=lfs merge=lfs -text
40
+ .git/lfs/objects/29/5e/295e70b58fbf265c9bdad1a24cf9f099be0ff2381e7ade96ccf9e457aafcc36e filter=lfs diff=lfs merge=lfs -text
1_Pooling/config.json CHANGED
@@ -3,5 +3,7 @@
3
  "pooling_mode_cls_token": true,
4
  "pooling_mode_mean_tokens": false,
5
  "pooling_mode_max_tokens": false,
6
- "pooling_mode_mean_sqrt_len_tokens": false
 
 
7
  }
 
3
  "pooling_mode_cls_token": true,
4
  "pooling_mode_mean_tokens": false,
5
  "pooling_mode_max_tokens": false,
6
+ "pooling_mode_mean_sqrt_len_tokens": false,
7
+ "pooling_mode_weightedmean_tokens": false,
8
+ "pooling_mode_lasttoken": false
9
  }
LT_training_config.json CHANGED
@@ -6,7 +6,8 @@
6
  "train_batch_size": 64,
7
  "num_epochs": 100,
8
  "warm_up_perc": 1,
9
- "learning_rate": 2e-06,
 
10
  "val_perc": 0.2,
11
  "wandb_names": {
12
  "project": "linkage",
@@ -16,10 +17,11 @@
16
  },
17
  "add_pooling_layer": false,
18
  "large_val": true,
19
- "eval_steps_perc": 0.1,
20
  "test_at_end": true,
21
  "save_val_test_pickles": true,
22
  "val_query_prop": 0.5,
 
23
  "eval_type": "retrieval",
24
  "training_dataset": "dataframe",
25
  "base_model_path": "multi-qa-mpnet-base-dot-v1",
 
6
  "train_batch_size": 64,
7
  "num_epochs": 100,
8
  "warm_up_perc": 1,
9
+ "learning_rate": 2e-05,
10
+ "loss_type": "supcon",
11
  "val_perc": 0.2,
12
  "wandb_names": {
13
  "project": "linkage",
 
17
  },
18
  "add_pooling_layer": false,
19
  "large_val": true,
20
+ "eval_steps_perc": 0.5,
21
  "test_at_end": true,
22
  "save_val_test_pickles": true,
23
  "val_query_prop": 0.5,
24
+ "loss_params": {},
25
  "eval_type": "retrieval",
26
  "training_dataset": "dataframe",
27
  "base_model_path": "multi-qa-mpnet-base-dot-v1",
README.md CHANGED
@@ -12,7 +12,7 @@ tags:
12
 
13
  # dell-research-harvard/lt-un-data-fine-fine-en
14
 
15
- This is a [LinkTransformer](https://github.com/dell-research-harvard/linktransformer) model. At its core this model this is a sentence transformer model [sentence-transformers](https://www.SBERT.net) model- it just wraps around the class.
16
  It is designed for quick and easy record linkage (entity-matching) through the LinkTransformer package. The tasks include clustering, deduplication, linking, aggregation and more.
17
  Notwithstanding that, it can be used for any sentence similarity task within the sentence-transformers framework as well.
18
  It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search.
@@ -110,12 +110,12 @@ Parameters of the fit()-Method:
110
  ```
111
  {
112
  "epochs": 100,
113
- "evaluation_steps": 1500,
114
  "evaluator": "sentence_transformers.evaluation.SequentialEvaluator.SequentialEvaluator",
115
  "max_grad_norm": 1,
116
  "optimizer_class": "<class 'torch.optim.adamw.AdamW'>",
117
  "optimizer_params": {
118
- "lr": 2e-06
119
  },
120
  "scheduler": "WarmupLinear",
121
  "steps_per_epoch": null,
@@ -129,10 +129,20 @@ Parameters of the fit()-Method:
129
 
130
  LinkTransformer(
131
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: MPNetModel
132
- (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
133
  )
134
  ```
135
 
136
  ## Citing & Authors
137
 
138
- <!--- Describe where people can find more information -->
 
 
 
 
 
 
 
 
 
 
 
12
 
13
  # dell-research-harvard/lt-un-data-fine-fine-en
14
 
15
+ This is a [LinkTransformer](https://linktransformer.github.io/) model. At its core this model this is a sentence transformer model [sentence-transformers](https://www.SBERT.net) model- it just wraps around the class.
16
  It is designed for quick and easy record linkage (entity-matching) through the LinkTransformer package. The tasks include clustering, deduplication, linking, aggregation and more.
17
  Notwithstanding that, it can be used for any sentence similarity task within the sentence-transformers framework as well.
18
  It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search.
 
110
  ```
111
  {
112
  "epochs": 100,
113
+ "evaluation_steps": 75,
114
  "evaluator": "sentence_transformers.evaluation.SequentialEvaluator.SequentialEvaluator",
115
  "max_grad_norm": 1,
116
  "optimizer_class": "<class 'torch.optim.adamw.AdamW'>",
117
  "optimizer_params": {
118
+ "lr": 2e-05
119
  },
120
  "scheduler": "WarmupLinear",
121
  "steps_per_epoch": null,
 
129
 
130
  LinkTransformer(
131
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: MPNetModel
132
+ (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False})
133
  )
134
  ```
135
 
136
  ## Citing & Authors
137
 
138
+ ```
139
+ @misc{arora2023linktransformer,
140
+ title={LinkTransformer: A Unified Package for Record Linkage with Transformer Language Models},
141
+ author={Abhishek Arora and Melissa Dell},
142
+ year={2023},
143
+ eprint={2309.00789},
144
+ archivePrefix={arXiv},
145
+ primaryClass={cs.CL}
146
+ }
147
+
148
+ ```
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "models/linkage_un_data_en_fine_fine/",
3
  "architectures": [
4
  "MPNetModel"
5
  ],
@@ -19,6 +19,6 @@
19
  "pad_token_id": 1,
20
  "relative_attention_num_buckets": 32,
21
  "torch_dtype": "float32",
22
- "transformers_version": "4.31.0",
23
  "vocab_size": 30527
24
  }
 
1
  {
2
+ "_name_or_path": "models/linkage_un_data_en_fine_fine",
3
  "architectures": [
4
  "MPNetModel"
5
  ],
 
19
  "pad_token_id": 1,
20
  "relative_attention_num_buckets": 32,
21
  "torch_dtype": "float32",
22
+ "transformers_version": "4.35.1",
23
  "vocab_size": 30527
24
  }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fbd246fa5fe8529451aee46397c8e583166d5588080b036919b0641d5b8751a5
3
+ size 437967672
special_tokens_map.json CHANGED
@@ -1,7 +1,25 @@
1
  {
2
- "bos_token": "<s>",
3
- "cls_token": "<s>",
4
- "eos_token": "</s>",
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5
  "mask_token": {
6
  "content": "<mask>",
7
  "lstrip": true,
@@ -9,7 +27,25 @@
9
  "rstrip": false,
10
  "single_word": false
11
  },
12
- "pad_token": "<pad>",
13
- "sep_token": "</s>",
14
- "unk_token": "[UNK]"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
15
  }
 
1
  {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
  "mask_token": {
24
  "content": "<mask>",
25
  "lstrip": true,
 
27
  "rstrip": false,
28
  "single_word": false
29
  },
30
+ "pad_token": {
31
+ "content": "<pad>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "</s>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "[UNK]",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
  }
tokenizer_config.json CHANGED
@@ -1,15 +1,72 @@
1
  {
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2
  "bos_token": "<s>",
3
  "clean_up_tokenization_spaces": true,
4
  "cls_token": "<s>",
5
  "do_lower_case": true,
6
  "eos_token": "</s>",
7
  "mask_token": "<mask>",
 
8
  "model_max_length": 512,
 
9
  "pad_token": "<pad>",
 
 
10
  "sep_token": "</s>",
 
11
  "strip_accents": null,
12
  "tokenize_chinese_chars": true,
13
  "tokenizer_class": "MPNetTokenizer",
 
 
14
  "unk_token": "[UNK]"
15
  }
 
1
  {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": true,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "104": {
36
+ "content": "[UNK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "30526": {
44
+ "content": "<mask>",
45
+ "lstrip": true,
46
+ "normalized": false,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": true
50
+ }
51
+ },
52
  "bos_token": "<s>",
53
  "clean_up_tokenization_spaces": true,
54
  "cls_token": "<s>",
55
  "do_lower_case": true,
56
  "eos_token": "</s>",
57
  "mask_token": "<mask>",
58
+ "max_length": 250,
59
  "model_max_length": 512,
60
+ "pad_to_multiple_of": null,
61
  "pad_token": "<pad>",
62
+ "pad_token_type_id": 0,
63
+ "padding_side": "right",
64
  "sep_token": "</s>",
65
+ "stride": 0,
66
  "strip_accents": null,
67
  "tokenize_chinese_chars": true,
68
  "tokenizer_class": "MPNetTokenizer",
69
+ "truncation_side": "right",
70
+ "truncation_strategy": "longest_first",
71
  "unk_token": "[UNK]"
72
  }