yujiepan commited on
Commit
c6b8155
1 Parent(s): ab3b3a7

upload model

Browse files
compressed_graph.dot ADDED
The diff for this file is too large to render. See raw diff
 
config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google/mobilebert-uncased",
3
+ "architectures": [
4
+ "NNCFNetwork"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_activation": false,
8
+ "classifier_dropout": null,
9
+ "embedding_size": 128,
10
+ "hidden_act": "relu",
11
+ "hidden_dropout_prob": 0.0,
12
+ "hidden_size": 512,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 512,
15
+ "intra_bottleneck_size": 128,
16
+ "key_query_shared_bottleneck": true,
17
+ "layer_norm_eps": 1e-12,
18
+ "max_position_embeddings": 512,
19
+ "model_type": "mobilebert",
20
+ "normalization_type": "no_norm",
21
+ "num_attention_heads": 4,
22
+ "num_feedforward_networks": 4,
23
+ "num_hidden_layers": 12,
24
+ "pad_token_id": 0,
25
+ "torch_dtype": "float32",
26
+ "transformers_version": "4.26.0",
27
+ "trigram_input": true,
28
+ "true_hidden_size": 128,
29
+ "type_vocab_size": 2,
30
+ "use_bottleneck": true,
31
+ "use_bottleneck_attention": false,
32
+ "vocab_size": 30522
33
+ }
mobilebert-base-qat.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "algorithm": "quantization",
3
+ "preset": "mixed",
4
+ "overflow_fix": "disable",
5
+ "initializer": {
6
+ "range": {
7
+ "num_init_samples": 300,
8
+ "type": "mean_min_max"
9
+ },
10
+ "batchnorm_adaptation": {
11
+ "num_bn_adaptation_samples": 0
12
+ },
13
+ },
14
+ "scope_overrides": {
15
+ "activations": {
16
+ "{re}.*matmul_0": {
17
+ "mode": "symmetric"
18
+ }
19
+ }
20
+ },
21
+ "ignored_scopes": [
22
+ "{re}.*Embedding*",
23
+ "{re}.*__add___[0-1]",
24
+ // "{re}.*layer_norm_0",
25
+ "{re}.*matmul_1",
26
+ "{re}.*__truediv__*",
27
+ ],
28
+ }
openvino_config.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "compression": {
3
+ "algorithm": "quantization",
4
+ "export_to_onnx_standard_ops": false,
5
+ "ignored_scopes": [
6
+ "{re}.*Embedding*",
7
+ "{re}.*__add___[0-1]",
8
+ "{re}.*matmul_1",
9
+ "{re}.*__truediv__*"
10
+ ],
11
+ "initializer": {
12
+ "batchnorm_adaptation": {
13
+ "num_bn_adaptation_samples": 0
14
+ },
15
+ "range": {
16
+ "num_init_samples": 300,
17
+ "type": "mean_min_max"
18
+ }
19
+ },
20
+ "overflow_fix": "disable",
21
+ "preset": "mixed",
22
+ "scope_overrides": {
23
+ "activations": {
24
+ "{re}.*matmul_0": {
25
+ "mode": "symmetric"
26
+ }
27
+ }
28
+ }
29
+ },
30
+ "input_info": [
31
+ {
32
+ "keyword": "input_ids",
33
+ "sample_size": [
34
+ 32,
35
+ 384
36
+ ],
37
+ "type": "long"
38
+ },
39
+ {
40
+ "keyword": "token_type_ids",
41
+ "sample_size": [
42
+ 32,
43
+ 384
44
+ ],
45
+ "type": "long"
46
+ },
47
+ {
48
+ "keyword": "attention_mask",
49
+ "sample_size": [
50
+ 32,
51
+ 384
52
+ ],
53
+ "type": "long"
54
+ }
55
+ ],
56
+ "log_dir": "/nvme2/yujiepan/workspace/jpqd-test/LOGS/optimum-mobilebert-qa/0308_sbtv_mobile_12blks_qat_lr6e-05",
57
+ "optimum_version": "1.6.3",
58
+ "save_onnx_model": false,
59
+ "transformers_version": "4.26.0"
60
+ }
openvino_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da7263d64223ea490a2a16da6baebf5dde123e3375d73429d70f9c1f57e83d73
3
+ size 28010788
openvino_model.xml ADDED
The diff for this file is too large to render. See raw diff
 
original_graph.dot ADDED
The diff for this file is too large to render. See raw diff
 
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c7f2855d097f594412ce2481d2cadb93f909abd7699c27b9e000e37e356fb82
3
+ size 59127415
special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "do_lower_case": true,
4
+ "mask_token": "[MASK]",
5
+ "model_max_length": 1000000000000000019884624838656,
6
+ "name_or_path": "google/mobilebert-uncased",
7
+ "pad_token": "[PAD]",
8
+ "sep_token": "[SEP]",
9
+ "special_tokens_map_file": null,
10
+ "strip_accents": null,
11
+ "tokenize_chinese_chars": true,
12
+ "tokenizer_class": "MobileBertTokenizer",
13
+ "unk_token": "[UNK]"
14
+ }
trainer_states.json ADDED
The diff for this file is too large to render. See raw diff
 
vocab.txt ADDED
The diff for this file is too large to render. See raw diff