Training done
Browse files- added_tokens.json +126 -0
- preprocessor_config.json +26 -0
- sentencepiece.bpe.model +3 -0
- special_tokens_map.json +19 -0
- tokenizer.json +0 -0
- tokenizer_config.json +21 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,126 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</s_1990>": 57644,
|
| 3 |
+
"</s_1991>": 57636,
|
| 4 |
+
"</s_1992>": 57618,
|
| 5 |
+
"</s_1993>": 57616,
|
| 6 |
+
"</s_1994>": 57586,
|
| 7 |
+
"</s_1995>": 57584,
|
| 8 |
+
"</s_1996>": 57526,
|
| 9 |
+
"</s_1997>": 57606,
|
| 10 |
+
"</s_1998>": 57604,
|
| 11 |
+
"</s_1999>": 57594,
|
| 12 |
+
"</s_2000>": 57592,
|
| 13 |
+
"</s_2001>": 57590,
|
| 14 |
+
"</s_2002>": 57588,
|
| 15 |
+
"</s_2003>": 57624,
|
| 16 |
+
"</s_2004>": 57622,
|
| 17 |
+
"</s_2005>": 57614,
|
| 18 |
+
"</s_2006>": 57612,
|
| 19 |
+
"</s_2007>": 57610,
|
| 20 |
+
"</s_2008>": 57608,
|
| 21 |
+
"</s_2009>": 57638,
|
| 22 |
+
"</s_2010>": 57626,
|
| 23 |
+
"</s_2011>": 57620,
|
| 24 |
+
"</s_2012>": 57602,
|
| 25 |
+
"</s_2013>": 57600,
|
| 26 |
+
"</s_2014>": 57598,
|
| 27 |
+
"</s_2015>": 57596,
|
| 28 |
+
"</s_2016>": 57640,
|
| 29 |
+
"</s_2017>": 57634,
|
| 30 |
+
"</s_2018>": 57632,
|
| 31 |
+
"</s_2019>": 57630,
|
| 32 |
+
"</s_2020>": 57628,
|
| 33 |
+
"</s_2021>": 57642,
|
| 34 |
+
"</s_Additional Paid-In Capital>": 57576,
|
| 35 |
+
"</s_Assets>": 57528,
|
| 36 |
+
"</s_Average accounts payable>": 57546,
|
| 37 |
+
"</s_Average accounts receivable>": 57540,
|
| 38 |
+
"</s_Average inventory>": 57542,
|
| 39 |
+
"</s_Common Stock>": 57564,
|
| 40 |
+
"</s_Credit Balance>": 57580,
|
| 41 |
+
"</s_Current Liabilities>": 57552,
|
| 42 |
+
"</s_Debt Period>": 57558,
|
| 43 |
+
"</s_Equity>": 57562,
|
| 44 |
+
"</s_Fixed Assets>": 57532,
|
| 45 |
+
"</s_General Reserves>": 57578,
|
| 46 |
+
"</s_Intangible Assets>": 57536,
|
| 47 |
+
"</s_Inventory>": 57538,
|
| 48 |
+
"</s_Issued Capital>": 57572,
|
| 49 |
+
"</s_Liabilities>": 57548,
|
| 50 |
+
"</s_Long Term Debt>": 57556,
|
| 51 |
+
"</s_Other Comprehensive Income>": 57568,
|
| 52 |
+
"</s_Paid-up Capital>": 57574,
|
| 53 |
+
"</s_Purchases>": 57544,
|
| 54 |
+
"</s_Retained Earnings>": 57582,
|
| 55 |
+
"</s_Short-term Liabilities>": 57554,
|
| 56 |
+
"</s_Tangible Assets>": 57534,
|
| 57 |
+
"</s_Term Liabilities>": 57560,
|
| 58 |
+
"</s_Total Assets>": 57530,
|
| 59 |
+
"</s_Total Equity>": 57570,
|
| 60 |
+
"</s_Total Liabilities>": 57550,
|
| 61 |
+
"</s_Treasury Stock>": 57566,
|
| 62 |
+
"<s_1990>": 57643,
|
| 63 |
+
"<s_1991>": 57635,
|
| 64 |
+
"<s_1992>": 57617,
|
| 65 |
+
"<s_1993>": 57615,
|
| 66 |
+
"<s_1994>": 57585,
|
| 67 |
+
"<s_1995>": 57583,
|
| 68 |
+
"<s_1996>": 57525,
|
| 69 |
+
"<s_1997>": 57605,
|
| 70 |
+
"<s_1998>": 57603,
|
| 71 |
+
"<s_1999>": 57593,
|
| 72 |
+
"<s_2000>": 57591,
|
| 73 |
+
"<s_2001>": 57589,
|
| 74 |
+
"<s_2002>": 57587,
|
| 75 |
+
"<s_2003>": 57623,
|
| 76 |
+
"<s_2004>": 57621,
|
| 77 |
+
"<s_2005>": 57613,
|
| 78 |
+
"<s_2006>": 57611,
|
| 79 |
+
"<s_2007>": 57609,
|
| 80 |
+
"<s_2008>": 57607,
|
| 81 |
+
"<s_2009>": 57637,
|
| 82 |
+
"<s_2010>": 57625,
|
| 83 |
+
"<s_2011>": 57619,
|
| 84 |
+
"<s_2012>": 57601,
|
| 85 |
+
"<s_2013>": 57599,
|
| 86 |
+
"<s_2014>": 57597,
|
| 87 |
+
"<s_2015>": 57595,
|
| 88 |
+
"<s_2016>": 57639,
|
| 89 |
+
"<s_2017>": 57633,
|
| 90 |
+
"<s_2018>": 57631,
|
| 91 |
+
"<s_2019>": 57629,
|
| 92 |
+
"<s_2020>": 57627,
|
| 93 |
+
"<s_2021>": 57641,
|
| 94 |
+
"<s_Additional Paid-In Capital>": 57575,
|
| 95 |
+
"<s_Assets>": 57527,
|
| 96 |
+
"<s_Average accounts payable>": 57545,
|
| 97 |
+
"<s_Average accounts receivable>": 57539,
|
| 98 |
+
"<s_Average inventory>": 57541,
|
| 99 |
+
"<s_Common Stock>": 57563,
|
| 100 |
+
"<s_Credit Balance>": 57579,
|
| 101 |
+
"<s_Current Liabilities>": 57551,
|
| 102 |
+
"<s_Debt Period>": 57557,
|
| 103 |
+
"<s_Equity>": 57561,
|
| 104 |
+
"<s_Fixed Assets>": 57531,
|
| 105 |
+
"<s_General Reserves>": 57577,
|
| 106 |
+
"<s_Intangible Assets>": 57535,
|
| 107 |
+
"<s_Inventory>": 57537,
|
| 108 |
+
"<s_Issued Capital>": 57571,
|
| 109 |
+
"<s_Liabilities>": 57547,
|
| 110 |
+
"<s_Long Term Debt>": 57555,
|
| 111 |
+
"<s_Other Comprehensive Income>": 57567,
|
| 112 |
+
"<s_Paid-up Capital>": 57573,
|
| 113 |
+
"<s_Purchases>": 57543,
|
| 114 |
+
"<s_Retained Earnings>": 57581,
|
| 115 |
+
"<s_Short-term Liabilities>": 57553,
|
| 116 |
+
"<s_Tangible Assets>": 57533,
|
| 117 |
+
"<s_Term Liabilities>": 57559,
|
| 118 |
+
"<s_Total Assets>": 57529,
|
| 119 |
+
"<s_Total Equity>": 57569,
|
| 120 |
+
"<s_Total Liabilities>": 57549,
|
| 121 |
+
"<s_Treasury Stock>": 57565,
|
| 122 |
+
"<s_cord-v2>": 57645,
|
| 123 |
+
"<s_iitcdip>": 57523,
|
| 124 |
+
"<s_synthdog>": 57524,
|
| 125 |
+
"<sep/>": 57522
|
| 126 |
+
}
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_align_long_axis": false,
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"do_pad": true,
|
| 5 |
+
"do_rescale": true,
|
| 6 |
+
"do_resize": true,
|
| 7 |
+
"do_thumbnail": true,
|
| 8 |
+
"image_mean": [
|
| 9 |
+
0.5,
|
| 10 |
+
0.5,
|
| 11 |
+
0.5
|
| 12 |
+
],
|
| 13 |
+
"image_processor_type": "DonutImageProcessor",
|
| 14 |
+
"image_std": [
|
| 15 |
+
0.5,
|
| 16 |
+
0.5,
|
| 17 |
+
0.5
|
| 18 |
+
],
|
| 19 |
+
"processor_class": "DonutProcessor",
|
| 20 |
+
"resample": 2,
|
| 21 |
+
"rescale_factor": 0.00392156862745098,
|
| 22 |
+
"size": [
|
| 23 |
+
960,
|
| 24 |
+
1280
|
| 25 |
+
]
|
| 26 |
+
}
|
sentencepiece.bpe.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cb9e3dce4c326195d08fc3dd0f7e2eee1da8595c847bf4c1a9c78b7a82d47e2d
|
| 3 |
+
size 1296245
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<s_iitcdip>",
|
| 4 |
+
"<s_synthdog>"
|
| 5 |
+
],
|
| 6 |
+
"bos_token": "<s>",
|
| 7 |
+
"cls_token": "<s>",
|
| 8 |
+
"eos_token": "</s>",
|
| 9 |
+
"mask_token": {
|
| 10 |
+
"content": "<mask>",
|
| 11 |
+
"lstrip": true,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "<pad>",
|
| 17 |
+
"sep_token": "</s>",
|
| 18 |
+
"unk_token": "<unk>"
|
| 19 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"clean_up_tokenization_spaces": true,
|
| 4 |
+
"cls_token": "<s>",
|
| 5 |
+
"eos_token": "</s>",
|
| 6 |
+
"mask_token": {
|
| 7 |
+
"__type": "AddedToken",
|
| 8 |
+
"content": "<mask>",
|
| 9 |
+
"lstrip": true,
|
| 10 |
+
"normalized": true,
|
| 11 |
+
"rstrip": false,
|
| 12 |
+
"single_word": false
|
| 13 |
+
},
|
| 14 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 15 |
+
"pad_token": "<pad>",
|
| 16 |
+
"processor_class": "DonutProcessor",
|
| 17 |
+
"sep_token": "</s>",
|
| 18 |
+
"sp_model_kwargs": {},
|
| 19 |
+
"tokenizer_class": "XLMRobertaTokenizer",
|
| 20 |
+
"unk_token": "<unk>"
|
| 21 |
+
}
|