ofirzaf commited on
Commit
8378a24
1 Parent(s): a75623d

Upload model files

Browse files
README.md ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: en
3
+ ---
4
+
5
+ # Sparse BERT base model fine tuned to MNLI without classifier layer (uncased)
6
+
7
+ Fine tuned sparse BERT base to MNLI (GLUE Benchmark) task from [bert-base-uncased-sparse-70-unstructured](https://huggingface.co/Intel/bert-base-uncased-sparse-70-unstructured).
8
+ <br>
9
+ This model doesn't have a classifier layer to enable easier loading of the model for training to other downstream tasks.
10
+ In all the other layers this model is similar to [bert-base-uncased-mnli-sparse-70-unstructured](https://huggingface.co/Intel/bert-base-uncased-mnli-sparse-70-unstructured).
11
+ <br><br>
12
+ Note: This model requires `transformers==2.10.0`
13
+
14
+ ## Evaluation Results
15
+ Matched: 82.5%
16
+ Mismatched: 83.3%
17
+
18
+ This model can be further fine-tuned to other tasks and achieve the following evaluation results:
19
+
20
+ | Task | QQP (Acc/F1) | QNLI (Acc) | SST-2 (Acc) | STS-B (Pears/Spear) | SQuADv1.1 (Acc/F1) |
21
+ |------|--------------|------------|-------------|---------------------|--------------------|
22
+ | | 90.2/86.7 | 90.3 | 91.5 | 88.9/88.6 | 80.5/88.2 |
config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "BertForSequenceClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "finetuning_task": "mnli",
7
+ "gradient_checkpointing": false,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "LABEL_0",
13
+ "1": "LABEL_1",
14
+ "2": "LABEL_2"
15
+ },
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 3072,
18
+ "label2id": {
19
+ "LABEL_0": 0,
20
+ "LABEL_1": 1,
21
+ "LABEL_2": 2
22
+ },
23
+ "layer_norm_eps": 1e-12,
24
+ "max_position_embeddings": 512,
25
+ "model_type": "bert",
26
+ "num_attention_heads": 12,
27
+ "num_hidden_layers": 12,
28
+ "pad_token_id": 0,
29
+ "type_vocab_size": 2,
30
+ "vocab_size": 30522
31
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:955735e9502bbbb96f358a3db7af654d4cd272cb0fa5390f39369ee060d8084d
3
+ size 438008068
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
1
+ {"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
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, "model_max_length": 512, "name_or_path": "bert-base-uncased"}
vocab.txt ADDED
The diff for this file is too large to render. See raw diff