add files
Browse files- config.json +48 -0
- pytorch_model.bin +3 -0
- sentencepiece.bpe.model +3 -0
- special_tokens_map.json +1 -0
- tokenizer_config.json +1 -0
- vocab.json +183 -0
config.json
ADDED
|
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_dropout": 0.1,
|
| 3 |
+
"activation_function": "relu",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"Speech2TextTransformerForConditionalGeneration"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.1,
|
| 8 |
+
"bos_token_id": 0,
|
| 9 |
+
"classifier_dropout": 0.0,
|
| 10 |
+
"conv_channels": 1024,
|
| 11 |
+
"conv_kernel_sizes": [
|
| 12 |
+
5,
|
| 13 |
+
5
|
| 14 |
+
],
|
| 15 |
+
"d_model": 256,
|
| 16 |
+
"decoder_attention_heads": 4,
|
| 17 |
+
"decoder_ffn_dim": 2048,
|
| 18 |
+
"decoder_layerdrop": 0.0,
|
| 19 |
+
"decoder_layers": 6,
|
| 20 |
+
"decoder_start_token_id": 2,
|
| 21 |
+
"dropout": 0.1,
|
| 22 |
+
"early_stopping": true,
|
| 23 |
+
"encoder_attention_heads": 4,
|
| 24 |
+
"encoder_ffn_dim": 2048,
|
| 25 |
+
"encoder_layerdrop": 0.0,
|
| 26 |
+
"encoder_layers": 12,
|
| 27 |
+
"eos_token_id": 2,
|
| 28 |
+
"gradient_checkpointing": false,
|
| 29 |
+
"init_std": 0.02,
|
| 30 |
+
"input_channels": 1,
|
| 31 |
+
"input_feat_per_channel": 80,
|
| 32 |
+
"is_encoder_decoder": true,
|
| 33 |
+
"max_length": 200,
|
| 34 |
+
"max_position_embeddings": 1024,
|
| 35 |
+
"max_source_positions": 6000,
|
| 36 |
+
"max_target_positions": 1024,
|
| 37 |
+
"model_type": "speech_to_text_transformer",
|
| 38 |
+
"num_beams": 5,
|
| 39 |
+
"num_conv_layers": 2,
|
| 40 |
+
"num_hidden_layers": 12,
|
| 41 |
+
"pad_token_id": 1,
|
| 42 |
+
"scale_embedding": true,
|
| 43 |
+
"share_embeds": false,
|
| 44 |
+
"tie_word_embeddings": false,
|
| 45 |
+
"transformers_version": "4.4.0.dev0",
|
| 46 |
+
"use_cache": true,
|
| 47 |
+
"vocab_size": 181
|
| 48 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:50642832df4ac676d148bd3e5881c4447a1bf7ce8571e6b9a358eec28ab55ba5
|
| 3 |
+
size 108412503
|
sentencepiece.bpe.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4880a4424d0a477e623a5d48e4f0d9e51d6c9794d2d3c76630248fa89f88d990
|
| 3 |
+
size 239648
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}
|
vocab.json
ADDED
|
@@ -0,0 +1,183 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<s>": 0,
|
| 3 |
+
"<pad>": 1,
|
| 4 |
+
"</s>": 2,
|
| 5 |
+
"<unk>": 3,
|
| 6 |
+
"\u2581": 4,
|
| 7 |
+
"e": 5,
|
| 8 |
+
"n": 6,
|
| 9 |
+
"i": 7,
|
| 10 |
+
"r": 8,
|
| 11 |
+
"t": 9,
|
| 12 |
+
"s": 10,
|
| 13 |
+
"a": 11,
|
| 14 |
+
"d": 12,
|
| 15 |
+
"h": 13,
|
| 16 |
+
"u": 14,
|
| 17 |
+
"l": 15,
|
| 18 |
+
"o": 16,
|
| 19 |
+
"c": 17,
|
| 20 |
+
"g": 18,
|
| 21 |
+
"m": 19,
|
| 22 |
+
".": 20,
|
| 23 |
+
"b": 21,
|
| 24 |
+
"f": 22,
|
| 25 |
+
"w": 23,
|
| 26 |
+
"k": 24,
|
| 27 |
+
"z": 25,
|
| 28 |
+
"S": 26,
|
| 29 |
+
"v": 27,
|
| 30 |
+
"p": 28,
|
| 31 |
+
",": 29,
|
| 32 |
+
"D": 30,
|
| 33 |
+
"\u00fc": 31,
|
| 34 |
+
"\u00e4": 32,
|
| 35 |
+
"E": 33,
|
| 36 |
+
"A": 34,
|
| 37 |
+
"B": 35,
|
| 38 |
+
"M": 36,
|
| 39 |
+
"G": 37,
|
| 40 |
+
"F": 38,
|
| 41 |
+
"K": 39,
|
| 42 |
+
"P": 40,
|
| 43 |
+
"W": 41,
|
| 44 |
+
"T": 42,
|
| 45 |
+
"y": 43,
|
| 46 |
+
"H": 44,
|
| 47 |
+
"\u00f6": 45,
|
| 48 |
+
"I": 46,
|
| 49 |
+
"R": 47,
|
| 50 |
+
"L": 48,
|
| 51 |
+
"-": 49,
|
| 52 |
+
"C": 50,
|
| 53 |
+
"V": 51,
|
| 54 |
+
"N": 52,
|
| 55 |
+
"\u00df": 53,
|
| 56 |
+
"\u201e": 54,
|
| 57 |
+
"Z": 55,
|
| 58 |
+
"\u201c": 56,
|
| 59 |
+
"J": 57,
|
| 60 |
+
"U": 58,
|
| 61 |
+
"j": 59,
|
| 62 |
+
"O": 60,
|
| 63 |
+
"x": 61,
|
| 64 |
+
"?": 62,
|
| 65 |
+
"\"": 63,
|
| 66 |
+
"!": 64,
|
| 67 |
+
"q": 65,
|
| 68 |
+
"Y": 66,
|
| 69 |
+
"\u00dc": 67,
|
| 70 |
+
":": 68,
|
| 71 |
+
"Q": 69,
|
| 72 |
+
"'": 70,
|
| 73 |
+
"\u00c4": 71,
|
| 74 |
+
"\u2018": 72,
|
| 75 |
+
"\u00d6": 73,
|
| 76 |
+
";": 74,
|
| 77 |
+
"(": 75,
|
| 78 |
+
")": 76,
|
| 79 |
+
"\u2013": 77,
|
| 80 |
+
"X": 78,
|
| 81 |
+
"\u201a": 79,
|
| 82 |
+
"\u201d": 80,
|
| 83 |
+
"0": 81,
|
| 84 |
+
"1": 82,
|
| 85 |
+
"\u00e9": 83,
|
| 86 |
+
"2": 84,
|
| 87 |
+
"\u2019": 85,
|
| 88 |
+
"&": 86,
|
| 89 |
+
"3": 87,
|
| 90 |
+
"5": 88,
|
| 91 |
+
"4": 89,
|
| 92 |
+
"7": 90,
|
| 93 |
+
"\u2014": 91,
|
| 94 |
+
"9": 92,
|
| 95 |
+
"8": 93,
|
| 96 |
+
"6": 94,
|
| 97 |
+
"/": 95,
|
| 98 |
+
"\u0301": 96,
|
| 99 |
+
"[": 97,
|
| 100 |
+
"]": 98,
|
| 101 |
+
"\u00e1": 99,
|
| 102 |
+
"\u014d": 100,
|
| 103 |
+
"\u00f3": 101,
|
| 104 |
+
"\u00f1": 102,
|
| 105 |
+
"\u00fa": 103,
|
| 106 |
+
"\u00ed": 104,
|
| 107 |
+
"\u0101": 105,
|
| 108 |
+
"\u00e8": 106,
|
| 109 |
+
"*": 107,
|
| 110 |
+
"\u0107": 108,
|
| 111 |
+
"\u00e0": 109,
|
| 112 |
+
"\u00ea": 110,
|
| 113 |
+
"\u00eb": 111,
|
| 114 |
+
"\u00a1": 112,
|
| 115 |
+
"\u00e7": 113,
|
| 116 |
+
"\u00f0": 114,
|
| 117 |
+
"\u00e3": 115,
|
| 118 |
+
"\u010d": 116,
|
| 119 |
+
"\u016b": 117,
|
| 120 |
+
"%": 118,
|
| 121 |
+
"\u00c9": 119,
|
| 122 |
+
"\u00e2": 120,
|
| 123 |
+
"\u00f8": 121,
|
| 124 |
+
"\u0161": 122,
|
| 125 |
+
"`": 123,
|
| 126 |
+
"\u00ad": 124,
|
| 127 |
+
"\u00e5": 125,
|
| 128 |
+
"\u00f4": 126,
|
| 129 |
+
"\u0142": 127,
|
| 130 |
+
"\u0153": 128,
|
| 131 |
+
"\u015f": 129,
|
| 132 |
+
"\u0160": 130,
|
| 133 |
+
"_": 131,
|
| 134 |
+
"\u00ce": 132,
|
| 135 |
+
"\u00d3": 133,
|
| 136 |
+
"\u00e6": 134,
|
| 137 |
+
"\u00ef": 135,
|
| 138 |
+
"\u0103": 136,
|
| 139 |
+
"\u011b": 137,
|
| 140 |
+
"\u012b": 138,
|
| 141 |
+
"\u0131": 139,
|
| 142 |
+
"\u02bb": 140,
|
| 143 |
+
"\u02bf": 141,
|
| 144 |
+
"\u03c0": 142,
|
| 145 |
+
"\u0438": 143,
|
| 146 |
+
"\u043a": 144,
|
| 147 |
+
"=": 145,
|
| 148 |
+
"\u00ab": 146,
|
| 149 |
+
"\u00bb": 147,
|
| 150 |
+
"\u00c3": 148,
|
| 151 |
+
"\u00d8": 149,
|
| 152 |
+
"\u00ee": 150,
|
| 153 |
+
"\u00fb": 151,
|
| 154 |
+
"\u00fe": 152,
|
| 155 |
+
"\u010b": 153,
|
| 156 |
+
"\u010c": 154,
|
| 157 |
+
"\u0119": 155,
|
| 158 |
+
"\u011f": 156,
|
| 159 |
+
"\u0144": 157,
|
| 160 |
+
"\u014c": 158,
|
| 161 |
+
"\u0151": 159,
|
| 162 |
+
"\u0159": 160,
|
| 163 |
+
"\u017e": 161,
|
| 164 |
+
"\u01ce": 162,
|
| 165 |
+
"\u03b1": 163,
|
| 166 |
+
"\u0412": 164,
|
| 167 |
+
"\u0435": 165,
|
| 168 |
+
"\u0437": 166,
|
| 169 |
+
"\u0439": 167,
|
| 170 |
+
"\u043b": 168,
|
| 171 |
+
"\u043d": 169,
|
| 172 |
+
"\u044c": 170,
|
| 173 |
+
"\u044f": 171,
|
| 174 |
+
"\u1e43": 172,
|
| 175 |
+
"\u1ea1": 173,
|
| 176 |
+
"\u1ee5": 174,
|
| 177 |
+
"\u2192": 175,
|
| 178 |
+
"\u2261": 176,
|
| 179 |
+
"\u4eac": 177,
|
| 180 |
+
"\u5927": 178,
|
| 181 |
+
"\u90fd": 179,
|
| 182 |
+
"\u962a": 180
|
| 183 |
+
}
|