lfoppiano commited on
Commit
2327e8f
·
1 Parent(s): ea8085c

Migrate to HF

Browse files
dataseer-binary_bert/config.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_name": "dataseer-binary_bert",
3
+ "architecture": "bert",
4
+ "embeddings_name": null,
5
+ "char_embedding_size": 25,
6
+ "word_embedding_size": 0,
7
+ "dropout": 0.5,
8
+ "recurrent_dropout": 0.25,
9
+ "maxlen": 300,
10
+ "dense_size": 32,
11
+ "use_char_feature": false,
12
+ "list_classes": [
13
+ "dataset",
14
+ "no_dataset"
15
+ ],
16
+ "fold_number": 1,
17
+ "batch_size": 16,
18
+ "transformer_name": "allenai/scibert_scivocab_cased"
19
+ }
dataseer-binary_bert/model_weights.hdf5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:881e7efd5339f2264fe81dd2c5345d2870729ebcaa37d609b691554798090962
3
+ size 440033696
dataseer-binary_bert/transformer-config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "allenai/scibert_scivocab_cased",
3
+ "attention_probs_dropout_prob": 0.1,
4
+ "classifier_dropout": null,
5
+ "hidden_act": "gelu",
6
+ "hidden_dropout_prob": 0.1,
7
+ "hidden_size": 768,
8
+ "initializer_range": 0.02,
9
+ "intermediate_size": 3072,
10
+ "layer_norm_eps": 1e-12,
11
+ "max_position_embeddings": 512,
12
+ "model_type": "bert",
13
+ "num_attention_heads": 12,
14
+ "num_hidden_layers": 12,
15
+ "pad_token_id": 0,
16
+ "position_embedding_type": "absolute",
17
+ "transformers_version": "4.15.0",
18
+ "type_vocab_size": 2,
19
+ "use_cache": true,
20
+ "vocab_size": 31116
21
+ }
dataseer-binary_bert/transformer-tokenizer/special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
dataseer-binary_bert/transformer-tokenizer/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
dataseer-binary_bert/transformer-tokenizer/tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "add_special_tokens": true, "max_length": 300, "add_prefix_space": true, "special_tokens_map_file": null, "name_or_path": "allenai/scibert_scivocab_cased", "do_basic_tokenize": true, "never_split": null, "tokenizer_class": "BertTokenizer"}
dataseer-binary_bert/transformer-tokenizer/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
dataseer-first_bert/config.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_name": "dataseer-first_bert",
3
+ "architecture": "bert",
4
+ "embeddings_name": null,
5
+ "char_embedding_size": 25,
6
+ "word_embedding_size": 0,
7
+ "dropout": 0.5,
8
+ "recurrent_dropout": 0.25,
9
+ "maxlen": 300,
10
+ "dense_size": 32,
11
+ "use_char_feature": false,
12
+ "list_classes": [
13
+ "calorimetry",
14
+ "chromatography",
15
+ "coulombimetry data",
16
+ "densitometry",
17
+ "electrocardiograph",
18
+ "electroencephalogram",
19
+ "electromyography",
20
+ "electrooculography",
21
+ "electrophysiology",
22
+ "electroretinography",
23
+ "emission flame photometry",
24
+ "flow cytometry",
25
+ "genetic data",
26
+ "image",
27
+ "mass spectrometry",
28
+ "no_dataset",
29
+ "protein data",
30
+ "sound data",
31
+ "spectrometry",
32
+ "spectrum analysis",
33
+ "systematic review",
34
+ "tabular data",
35
+ "video recording",
36
+ "voltammetry data",
37
+ "x-ray diffraction data"
38
+ ],
39
+ "fold_number": 1,
40
+ "batch_size": 16,
41
+ "transformer_name": "allenai/scibert_scivocab_cased"
42
+ }
dataseer-first_bert/model_weights.hdf5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:428e552bf791ab71f7b699516ab9e403cc6dab1a77cf293dd7f2fc9214fd6541
3
+ size 440104352
dataseer-first_bert/transformer-config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "allenai/scibert_scivocab_cased",
3
+ "attention_probs_dropout_prob": 0.1,
4
+ "classifier_dropout": null,
5
+ "hidden_act": "gelu",
6
+ "hidden_dropout_prob": 0.1,
7
+ "hidden_size": 768,
8
+ "initializer_range": 0.02,
9
+ "intermediate_size": 3072,
10
+ "layer_norm_eps": 1e-12,
11
+ "max_position_embeddings": 512,
12
+ "model_type": "bert",
13
+ "num_attention_heads": 12,
14
+ "num_hidden_layers": 12,
15
+ "pad_token_id": 0,
16
+ "position_embedding_type": "absolute",
17
+ "transformers_version": "4.15.0",
18
+ "type_vocab_size": 2,
19
+ "use_cache": true,
20
+ "vocab_size": 31116
21
+ }
dataseer-first_bert/transformer-tokenizer/special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
dataseer-first_bert/transformer-tokenizer/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
dataseer-first_bert/transformer-tokenizer/tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "add_special_tokens": true, "max_length": 300, "add_prefix_space": true, "special_tokens_map_file": null, "name_or_path": "allenai/scibert_scivocab_cased", "do_basic_tokenize": true, "never_split": null, "tokenizer_class": "BertTokenizer"}
dataseer-first_bert/transformer-tokenizer/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
dataseer-reuse_bert/config.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_name": "dataseer-reuse_bert",
3
+ "architecture": "bert",
4
+ "embeddings_name": null,
5
+ "char_embedding_size": 25,
6
+ "word_embedding_size": 0,
7
+ "dropout": 0.5,
8
+ "recurrent_dropout": 0.25,
9
+ "maxlen": 300,
10
+ "dense_size": 32,
11
+ "use_char_feature": false,
12
+ "list_classes": [
13
+ "no_reuse",
14
+ "reuse"
15
+ ],
16
+ "fold_number": 1,
17
+ "batch_size": 16,
18
+ "transformer_name": "allenai/scibert_scivocab_cased"
19
+ }
dataseer-reuse_bert/model_weights.hdf5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b92337a7679870884c2cb7051199664b9691d19422f7b592e9deb4bed0ecdb01
3
+ size 440033696
dataseer-reuse_bert/transformer-config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "allenai/scibert_scivocab_cased",
3
+ "attention_probs_dropout_prob": 0.1,
4
+ "classifier_dropout": null,
5
+ "hidden_act": "gelu",
6
+ "hidden_dropout_prob": 0.1,
7
+ "hidden_size": 768,
8
+ "initializer_range": 0.02,
9
+ "intermediate_size": 3072,
10
+ "layer_norm_eps": 1e-12,
11
+ "max_position_embeddings": 512,
12
+ "model_type": "bert",
13
+ "num_attention_heads": 12,
14
+ "num_hidden_layers": 12,
15
+ "pad_token_id": 0,
16
+ "position_embedding_type": "absolute",
17
+ "transformers_version": "4.15.0",
18
+ "type_vocab_size": 2,
19
+ "use_cache": true,
20
+ "vocab_size": 31116
21
+ }
dataseer-reuse_bert/transformer-tokenizer/special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
dataseer-reuse_bert/transformer-tokenizer/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
dataseer-reuse_bert/transformer-tokenizer/tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "add_special_tokens": true, "max_length": 300, "add_prefix_space": true, "special_tokens_map_file": null, "name_or_path": "allenai/scibert_scivocab_cased", "do_basic_tokenize": true, "never_split": null, "tokenizer_class": "BertTokenizer"}
dataseer-reuse_bert/transformer-tokenizer/vocab.txt ADDED
The diff for this file is too large to render. See raw diff