Upload folder using huggingface_hub
Browse files- config.json +53 -0
- generation_config.json +6 -0
- merges.txt +0 -0
- onnx/model.onnx +3 -0
- onnx/model_quantized.onnx +3 -0
- quantize_config.json +33 -0
- special_tokens_map.json +20 -0
- tokenizer_config.json +32 -0
- vocab.json +0 -0
config.json
ADDED
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "xlm-mlm-enfr-1024",
|
3 |
+
"architectures": [
|
4 |
+
"XLMWithLMHeadModel"
|
5 |
+
],
|
6 |
+
"asm": false,
|
7 |
+
"attention_dropout": 0.1,
|
8 |
+
"bos_index": 0,
|
9 |
+
"bos_token_id": 0,
|
10 |
+
"causal": false,
|
11 |
+
"dropout": 0.1,
|
12 |
+
"emb_dim": 1024,
|
13 |
+
"embed_init_std": 0.02209708691207961,
|
14 |
+
"end_n_top": 5,
|
15 |
+
"eos_index": 1,
|
16 |
+
"gelu_activation": true,
|
17 |
+
"id2lang": {
|
18 |
+
"0": "en",
|
19 |
+
"1": "fr"
|
20 |
+
},
|
21 |
+
"init_std": 0.02,
|
22 |
+
"is_encoder": true,
|
23 |
+
"lang2id": {
|
24 |
+
"en": 0,
|
25 |
+
"fr": 1
|
26 |
+
},
|
27 |
+
"lang_id": 0,
|
28 |
+
"layer_norm_eps": 1e-12,
|
29 |
+
"mask_index": 5,
|
30 |
+
"mask_token_id": 0,
|
31 |
+
"max_position_embeddings": 512,
|
32 |
+
"max_vocab": -1,
|
33 |
+
"min_count": 0,
|
34 |
+
"model_type": "xlm",
|
35 |
+
"n_heads": 8,
|
36 |
+
"n_langs": 2,
|
37 |
+
"n_layers": 6,
|
38 |
+
"pad_index": 2,
|
39 |
+
"pad_token_id": 2,
|
40 |
+
"same_enc_dec": true,
|
41 |
+
"share_inout_emb": true,
|
42 |
+
"sinusoidal_embeddings": false,
|
43 |
+
"start_n_top": 5,
|
44 |
+
"summary_activation": null,
|
45 |
+
"summary_first_dropout": 0.1,
|
46 |
+
"summary_proj_to_labels": true,
|
47 |
+
"summary_type": "first",
|
48 |
+
"summary_use_proj": true,
|
49 |
+
"transformers_version": "4.34.0.dev0",
|
50 |
+
"unk_index": 3,
|
51 |
+
"use_lang_emb": true,
|
52 |
+
"vocab_size": 64139
|
53 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 0,
|
4 |
+
"pad_token_id": 2,
|
5 |
+
"transformers_version": "4.34.0.dev0"
|
6 |
+
}
|
merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
onnx/model.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:88846b744b481e18c79890a9b992c54a96d02a1e181a14da82b6856a1b0a0847
|
3 |
+
size 567487125
|
onnx/model_quantized.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5388b959fda66f838d907f9f57de2cdf6db23ef8a303e90ed1f651e8b6e36891
|
3 |
+
size 142735420
|
quantize_config.json
ADDED
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"per_channel": true,
|
3 |
+
"reduce_range": true,
|
4 |
+
"per_model_config": {
|
5 |
+
"model": {
|
6 |
+
"op_types": [
|
7 |
+
"Erf",
|
8 |
+
"Equal",
|
9 |
+
"Cast",
|
10 |
+
"Concat",
|
11 |
+
"Softmax",
|
12 |
+
"Reshape",
|
13 |
+
"Sqrt",
|
14 |
+
"Mul",
|
15 |
+
"Unsqueeze",
|
16 |
+
"Pow",
|
17 |
+
"Transpose",
|
18 |
+
"Expand",
|
19 |
+
"Where",
|
20 |
+
"MatMul",
|
21 |
+
"ReduceMean",
|
22 |
+
"Slice",
|
23 |
+
"Gather",
|
24 |
+
"Div",
|
25 |
+
"Shape",
|
26 |
+
"Constant",
|
27 |
+
"Add",
|
28 |
+
"Sub"
|
29 |
+
],
|
30 |
+
"weight_type": "QInt8"
|
31 |
+
}
|
32 |
+
}
|
33 |
+
}
|
special_tokens_map.json
ADDED
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
"<special0>",
|
4 |
+
"<special1>",
|
5 |
+
"<special2>",
|
6 |
+
"<special3>",
|
7 |
+
"<special4>",
|
8 |
+
"<special5>",
|
9 |
+
"<special6>",
|
10 |
+
"<special7>",
|
11 |
+
"<special8>",
|
12 |
+
"<special9>"
|
13 |
+
],
|
14 |
+
"bos_token": "<s>",
|
15 |
+
"cls_token": "</s>",
|
16 |
+
"mask_token": "<special1>",
|
17 |
+
"pad_token": "<pad>",
|
18 |
+
"sep_token": "</s>",
|
19 |
+
"unk_token": "<unk>"
|
20 |
+
}
|
tokenizer_config.json
ADDED
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
"<special0>",
|
4 |
+
"<special1>",
|
5 |
+
"<special2>",
|
6 |
+
"<special3>",
|
7 |
+
"<special4>",
|
8 |
+
"<special5>",
|
9 |
+
"<special6>",
|
10 |
+
"<special7>",
|
11 |
+
"<special8>",
|
12 |
+
"<special9>"
|
13 |
+
],
|
14 |
+
"bos_token": "<s>",
|
15 |
+
"clean_up_tokenization_spaces": true,
|
16 |
+
"cls_token": "</s>",
|
17 |
+
"do_lowercase_and_remove_accent": true,
|
18 |
+
"id2lang": {
|
19 |
+
"0": "en",
|
20 |
+
"1": "fr"
|
21 |
+
},
|
22 |
+
"lang2id": {
|
23 |
+
"en": 0,
|
24 |
+
"fr": 1
|
25 |
+
},
|
26 |
+
"mask_token": "<special1>",
|
27 |
+
"model_max_length": 512,
|
28 |
+
"pad_token": "<pad>",
|
29 |
+
"sep_token": "</s>",
|
30 |
+
"tokenizer_class": "XLMTokenizer",
|
31 |
+
"unk_token": "<unk>"
|
32 |
+
}
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|