Commit
·
adc611c
1
Parent(s):
837d4e6
Upload tiny models for ProphetNetModel
Browse files- config.json +34 -0
- prophetnet.tokenizer +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +6 -0
- tokenizer_config.json +16 -0
config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_dropout": 0.1,
|
| 3 |
+
"activation_function": "gelu",
|
| 4 |
+
"add_cross_attention": true,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"ProphetNetModel"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.1,
|
| 9 |
+
"bos_token_id": 1,
|
| 10 |
+
"decoder_ffn_dim": 32,
|
| 11 |
+
"decoder_start_token_id": 0,
|
| 12 |
+
"disable_ngram_loss": false,
|
| 13 |
+
"dropout": 0.1,
|
| 14 |
+
"encoder_ffn_dim": 32,
|
| 15 |
+
"eos_token_id": 2,
|
| 16 |
+
"eps": 0.0,
|
| 17 |
+
"hidden_size": 16,
|
| 18 |
+
"init_std": 0.02,
|
| 19 |
+
"is_encoder_decoder": true,
|
| 20 |
+
"max_position_embeddings": 30,
|
| 21 |
+
"model_type": "prophetnet",
|
| 22 |
+
"ngram": 2,
|
| 23 |
+
"num_buckets": 32,
|
| 24 |
+
"num_decoder_attention_heads": 4,
|
| 25 |
+
"num_decoder_layers": 4,
|
| 26 |
+
"num_encoder_attention_heads": 4,
|
| 27 |
+
"num_encoder_layers": 4,
|
| 28 |
+
"pad_token_id": 0,
|
| 29 |
+
"relative_max_distance": 128,
|
| 30 |
+
"torch_dtype": "float32",
|
| 31 |
+
"transformers_version": "4.28.0.dev0",
|
| 32 |
+
"use_cache": true,
|
| 33 |
+
"vocab_size": 30522
|
| 34 |
+
}
|
prophetnet.tokenizer
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:a698a93d2db6668d9a6a54c66e7f255e50fe0a368ea8732a004479fed60e3c66
|
| 3 |
+
size 2140781
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"mask_token": "[MASK]",
|
| 3 |
+
"pad_token": "[PAD]",
|
| 4 |
+
"sep_token": "[SEP]",
|
| 5 |
+
"unk_token": "[UNK]"
|
| 6 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"clean_up_tokenization_spaces": true,
|
| 3 |
+
"do_basic_tokenize": true,
|
| 4 |
+
"do_lower_case": true,
|
| 5 |
+
"mask_token": "[MASK]",
|
| 6 |
+
"model_max_length": 30,
|
| 7 |
+
"never_split": null,
|
| 8 |
+
"pad_token": "[PAD]",
|
| 9 |
+
"sep_token": "[SEP]",
|
| 10 |
+
"special_tokens_map_file": "/home/runner/.cache/huggingface/hub/models--microsoft--prophetnet-large-uncased/snapshots/fd5b6f7e0cae2f7cd69f33b3da1d316c8f43645e/special_tokens_map.json",
|
| 11 |
+
"strip_accents": null,
|
| 12 |
+
"tokenize_chinese_chars": true,
|
| 13 |
+
"tokenizer_class": "ProphetNetTokenizer",
|
| 14 |
+
"unk_token": "[UNK]",
|
| 15 |
+
"x_sep_token": "[X_SEP]"
|
| 16 |
+
}
|