96abhishekarora commited on
Commit
05824ae
1 Parent(s): 4b3a896

Modified validation and training for linktransformer model

Browse files
.gitattributes CHANGED
@@ -35,3 +35,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
  pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
37
  .git/lfs/objects/90/ac/90ac7bf125986985f5f4235d75ac355323d510d32c59bc4ae08a39e6fcdd2b9b filter=lfs diff=lfs merge=lfs -text
 
 
 
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
  pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
37
  .git/lfs/objects/90/ac/90ac7bf125986985f5f4235d75ac355323d510d32c59bc4ae08a39e6fcdd2b9b filter=lfs diff=lfs merge=lfs -text
38
+ model.safetensors filter=lfs diff=lfs merge=lfs -text
39
+ .git/lfs/objects/7b/99/7b994472fa016892e4f8d14aa7bdb6581348e10438fd2a5838ec760305a91131 filter=lfs diff=lfs merge=lfs -text
1_Pooling/config.json CHANGED
@@ -3,5 +3,7 @@
3
  "pooling_mode_cls_token": false,
4
  "pooling_mode_mean_tokens": true,
5
  "pooling_mode_max_tokens": false,
6
- "pooling_mode_mean_sqrt_len_tokens": false
 
 
7
  }
 
3
  "pooling_mode_cls_token": false,
4
  "pooling_mode_mean_tokens": true,
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",
@@ -16,9 +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
  "eval_type": "retrieval",
23
  "training_dataset": "dataframe",
24
  "base_model_path": "hiiamsid/sentence_similarity_spanish_es",
 
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": "hiiamsid/sentence_similarity_spanish_es",
README.md CHANGED
@@ -12,7 +12,7 @@ tags:
12
 
13
  # dell-research-harvard/lt-wikidata-comp-es
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.
@@ -97,7 +97,7 @@ The model was trained with the parameters:
97
 
98
  **DataLoader**:
99
 
100
- `torch.utils.data.dataloader.DataLoader` of length 104 with parameters:
101
  ```
102
  {'batch_size': 64, 'sampler': 'torch.utils.data.dataloader._InfiniteConstantSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'}
103
  ```
@@ -110,16 +110,16 @@ Parameters of the fit()-Method:
110
  ```
111
  {
112
  "epochs": 100,
113
- "evaluation_steps": 1040,
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,
122
- "warmup_steps": 10400,
123
  "weight_decay": 0.01
124
  }
125
  ```
@@ -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: BertModel
132
- (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, '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-wikidata-comp-es
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.
 
97
 
98
  **DataLoader**:
99
 
100
+ `torch.utils.data.dataloader.DataLoader` of length 258 with parameters:
101
  ```
102
  {'batch_size': 64, 'sampler': 'torch.utils.data.dataloader._InfiniteConstantSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'}
103
  ```
 
110
  ```
111
  {
112
  "epochs": 100,
113
+ "evaluation_steps": 129,
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,
122
+ "warmup_steps": 25800,
123
  "weight_decay": 0.01
124
  }
125
  ```
 
129
 
130
  LinkTransformer(
131
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel
132
+ (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, '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_es_aliases/",
3
  "architectures": [
4
  "BertModel"
5
  ],
@@ -20,7 +20,7 @@
20
  "pad_token_id": 1,
21
  "position_embedding_type": "absolute",
22
  "torch_dtype": "float32",
23
- "transformers_version": "4.31.0",
24
  "type_vocab_size": 2,
25
  "use_cache": true,
26
  "vocab_size": 31002
 
1
  {
2
+ "_name_or_path": "models/linkage_es_aliases",
3
  "architectures": [
4
  "BertModel"
5
  ],
 
20
  "pad_token_id": 1,
21
  "position_embedding_type": "absolute",
22
  "torch_dtype": "float32",
23
+ "transformers_version": "4.35.1",
24
  "type_vocab_size": 2,
25
  "use_cache": true,
26
  "vocab_size": 31002
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:75e6e3f228dafb759c1c8552b6713799970f1df5ded9db033a7c670b977404de
3
+ size 439425888
special_tokens_map.json CHANGED
@@ -1,7 +1,37 @@
1
  {
2
- "cls_token": "[CLS]",
3
- "mask_token": "[MASK]",
4
- "pad_token": "[PAD]",
5
- "sep_token": "[SEP]",
6
- "unk_token": "[UNK]"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7
  }
 
1
  {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
  }
tokenizer_config.json CHANGED
@@ -1,15 +1,64 @@
1
  {
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2
  "clean_up_tokenization_spaces": true,
3
  "cls_token": "[CLS]",
4
  "do_basic_tokenize": true,
5
  "do_lower_case": false,
6
  "mask_token": "[MASK]",
 
7
  "model_max_length": 512,
8
  "never_split": null,
 
9
  "pad_token": "[PAD]",
 
 
10
  "sep_token": "[SEP]",
 
11
  "strip_accents": false,
12
  "tokenize_chinese_chars": true,
13
  "tokenizer_class": "BertTokenizer",
 
 
14
  "unk_token": "[UNK]"
15
  }
 
1
  {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[MASK]",
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
+ "3": {
20
+ "content": "[UNK]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "4": {
28
+ "content": "[CLS]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "5": {
36
+ "content": "[SEP]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
  "clean_up_tokenization_spaces": true,
45
  "cls_token": "[CLS]",
46
  "do_basic_tokenize": true,
47
  "do_lower_case": false,
48
  "mask_token": "[MASK]",
49
+ "max_length": 512,
50
  "model_max_length": 512,
51
  "never_split": null,
52
+ "pad_to_multiple_of": null,
53
  "pad_token": "[PAD]",
54
+ "pad_token_type_id": 0,
55
+ "padding_side": "right",
56
  "sep_token": "[SEP]",
57
+ "stride": 0,
58
  "strip_accents": false,
59
  "tokenize_chinese_chars": true,
60
  "tokenizer_class": "BertTokenizer",
61
+ "truncation_side": "right",
62
+ "truncation_strategy": "longest_first",
63
  "unk_token": "[UNK]"
64
  }