Initial commit

1_Pooling/config.json

@@ -0,0 +1,7 @@
+{
+  "word_embedding_dimension": 768,
+  "pooling_mode_cls_token": false,
+  "pooling_mode_mean_tokens": true,
+  "pooling_mode_max_tokens": false,
+  "pooling_mode_mean_sqrt_len_tokens": false
+}

2_Dense/config.json

@@ -0,0 +1 @@
+{"in_features": 768, "out_features": 1024, "bias": true, "activation_function": "torch.nn.modules.activation.Tanh"}

2_Dense/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a47bed97476cabb2875f74ac357fd65e70c80c767dd194096e7e2f03ab13186
+size 3151041

added_tokens.json

@@ -0,0 +1 @@
+{"🍧": 248692, "🧵": 249169, "🫨": 248747, "⚛": 248761, "🌓": 248598, "🦷": 248771, "🍭": 248923, "🙉": 248633, "🔕": 248780, "🎡": 248932, "🤾": 248721, "🫏": 248826, "♍": 248609, "🤬": 249219, "🦑": 248746, "🚜": 249076, "🥫": 248509, "🕹": 248871, "🪸": 248492, "🧣": 249155, "💹": 249158, "‼": 248733, "🪖": 248447, "😸": 248908, "🔳": 248985, "🛤": 248973, "🦔": 248641, "🦭": 249159, "🐙": 248701, "🛖": 248435, "🧄": 248723, "🏪": 248999, "🦀": 249201, "🍗": 248433, "🏔": 248588, "🍵": 248873, "🚁": 248677, "🤸": 248809, "⏯": 248576, "🎃": 248736, "🦓": 248789, "📊": 248473, "🔉": 248590, "🧤": 248740, "🥘": 248445, "🚯": 248659, "🎻": 249148, "🥡": 248720, "🪽": 249136, "🕕": 248494, "🐛": 248540, "🦵": 248885, "🐕": 248756, "🆕": 248655, "🏷": 248427, "🪁": 249183, "📈": 248438, "🎴": 248769, "🧘": 248663, "🫙": 249054, "📿": 249181, "🍱": 248715, "🫰": 249012, "♉": 248638, "🪈": 249038, "🫃": 248637, "🛕": 248859, "🏩": 249182, "🏉": 248434, "🚧": 249143, "🌮": 248976, "🆖": 248837, "🚢": 248547, "🥲": 248429, "㊗": 248521, "🟡": 248460, "🛀": 249097, "🐟": 248912, "🐀": 249257, "🦇": 248543, "🛅": 249041, "🔀": 248946, "🏇": 248612, "🛏": 249078, "📂": 248496, "🤢": 249023, "🟨": 248765, "🔪": 248906, "🪼": 248952, "🔲": 248988, "🌂": 248584, "📑": 249256, "🍡": 248591, "🚐": 248731, "🩶": 249243, "⬛": 249000, "🛶": 248940, "🈸": 249079, "🦹": 248958, "🥓": 248751, "🪔": 248919, "🛬": 248951, "🤕": 248475, "🫒": 248703, "🪞": 248687, "🈂": 248844, "🐠": 249253, "🚟": 249154, "🩰": 248852, "🧫": 248975, "👢": 248910, "🪠": 248499, "🪡": 248841, "😺": 248851, "🏞": 249200, "🦘": 248825, "🫄": 248658, "💈": 248554, "🗾": 248464, "🫦": 248526, "🧍": 249218, "🎚": 249103, "🧮": 249049, "🐇": 249015, "🤼": 249115, "📔": 248909, "🪨": 248673, "🏭": 248934, "🪛": 249077, "🥎": 248645, "🪱": 249162, "🚭": 248836, "🍥": 248722, "👾": 249059, "👡": 248549, "🥠": 248942, "🛂": 248770, "🃏": 248468, "👲": 249001, "🛄": 248879, "🕚": 248607, "🐂": 248621, "🔼": 248754, "🫢": 248783, "⏲": 248922, "🪜": 248597, "🦱": 248613, "🟠": 249195, "⏭": 249212, "🦥": 248939, "🎲": 248455, "🥾": 249056, "🏙": 248567, "🥸": 248486, "🥧": 248595, "🖨": 248947, "🪢": 249124, "🏯": 249175, "🟣": 248737, "🏰": 248774, "🎾": 249108, "🌒": 249187, "📪": 248565, "🍙": 249057, "🧳": 249179, "🪹": 248592, "🤒": 248610, "🕠": 249109, "⌛": 248707, "🏘": 249199, "🧹": 249233, "🈯": 248568, "🐭": 248615, "🦻": 248729, "🚾": 249205, "🚸": 248550, "🍞": 248697, "🫔": 249126, "🚕": 248608, "🦲": 249047, "🛐": 248830, "🔨": 248849, "🍝": 249087, "🐐": 249198, "🐁": 249252, "🕌": 249241, "🪵": 248877, "🛋": 249203, "🍤": 248777, "🖇": 248563, "🐊": 248574, "🐩": 248580, "⏮": 248920, "🪕": 248759, "🔭": 249261, "♊": 248741, "💩": 248508, "🚺": 249044, "🤏": 248699, "🧢": 249035, "🖌": 249209, "🏸": 248804, "🥏": 248557, "🚝": 248882, "🛸": 248562, "🧁": 248840, "🛡": 248811, "🗞": 248867, "🧱": 249112, "🛥": 248436, "🦸": 248538, "🤚": 249011, "🪅": 248440, "🏜": 248685, "🕙": 248539, "🫣": 248481, "🛵": 248857, "🥐": 249131, "🫀": 248442, "🧌": 249061, "🤱": 248925, "🚋": 248878, "⛈": 249232, "🏂": 248792, "🈳": 248785, "⚙": 248676, "👴": 248820, "🥌": 249177, "🩻": 248474, "🌬": 248530, "🗻": 249133, "🐧": 249254, "🫱": 249090, "⌨": 248450, "📭": 248998, "🚦": 248717, "🕣": 248854, "💨": 248913, "🚰": 249202, "💾": 249215, "🤍": 248967, "⛩": 248953, "🐓": 248680, "♓": 248790, "🪚": 249165, "🍸": 249251, "🔆": 249046, "🛰": 248656, "🦶": 248876, "🤎": 249116, "🌉": 248964, "🧓": 249145, "🏍": 248525, "🥶": 249147, "🥷": 248702, "🕦": 248938, "🫖": 248724, "🐬": 249002, "📛": 248484, "🧊": 249092, "👽": 248883, "🐉": 248462, "🕢": 248896, "🆗": 248917, "🏑": 249144, "🌠": 248623, "🥙": 248463, "🥻": 249039, "🏧": 248642, "🪯": 248822, "🧼": 248993, "🏣": 248828, "🦴": 248691, "🤫": 248706, "🐿": 248823, "😽": 249121, "🥊": 248768, "🛗": 248551, "🥢": 249168, "🥦": 248432, "🔛": 248671, "🥚": 248520, "🕖": 248847, "🥁": 249118, "🫶": 249157, "🕋": 249027, "🐑": 249122, "🔠": 249075, "🎰": 248683, "🌵": 249081, "➗": 248803, "🩹": 248477, "🚼": 249214, "🏕": 248916, "🦠": 249085, "🔢": 249142, "🪦": 248902, "🧨": 248793, "🈷": 249153, "🔏": 248622, "🛁": 248430, "🥥": 248956, "🪻": 248898, "👵": 248960, "🚉": 248856, "👹": 249099, "🚆": 248974, "🈴": 249003, "🎇": 248589, "🦞": 249045, "🩴": 248705, "🔄": 248995, "🫁": 248449, "🫠": 249247, "🌇": 248528, "🚞": 248596, "🚏": 249119, "🪘": 248643, "🕞": 248601, "🦳": 248927, "����": 248519, "🌔": 248514, "🚔": 249259, "🥵": 249029, "🥜": 249083, "🪬": 248564, "🥿": 249138, "⛷": 248649, "🧦": 248900, "🎒": 249072, "🚪": 248861, "🧥": 248657, "🪝": 248714, "🧠": 248893, "🌚": 249190, "🧒": 248536, "🛣": 249040, "🫗": 248652, "🦨": 248776, "⚧": 248971, "🤰": 249207, "🫎": 248894, "🚹": 249024, "💱": 249100, "🌦": 249172, "🤶": 249094, "🪒": 248467, "🐫": 248505, "🏗": 249098, "🟥": 249239, "🧆": 248571, "🐽": 248602, "⚱": 249010, "🤐": 248806, "🎌": 248805, "🧋": 248696, "⏸": 249193, "🦗": 249235, "🐢": 249210, "🫕": 249174, "📁": 248444, "🍈": 249173, "🪮": 248630, "🖖": 248864, "🥄": 248944, "😿": 248796, "🦡": 249246, "🏤": 248773, "🏒": 248866, "🛝": 248510, "🥴": 248860, "🦌": 248625, "♎": 248611, "🦺": 248948, "📉": 248617, "🥪": 248963, "📆": 248487, "🔈": 248559, "🌌": 248587, "🥍": 249051, "🛟": 248981, "🤴": 248573, "🤛": 249107, "🛩": 249156, "🚒": 249225, "🗒": 249048, "🚎": 249163, "⛹": 249260, "🎫": 249258, "🧴": 249104, "🎎": 249231, "🦈": 249139, "🗺": 248493, "🕜": 248513, "🍛": 248869, "📤": 248718, "🔩": 248781, "⏹": 249213, "🌭": 248868, "🚊": 249178, "🔑": 248937, "🪳": 249037, "🚃": 249184, "🟰": 248874, "🦍": 248713, "🚬": 248918, "🐵": 248533, "🈶": 248502, "⏏": 248572, "🩺": 248983, "🖍": 248560, "🏦": 248553, "🛞": 248897, "🍐": 248933, "🥼": 248832, "㊙": 249060, "🧟": 249197, "🍘": 248647, "🏛": 248914, "🔒": 248498, "🧏": 248544, "🪐": 249032, "🦬": 249245, "👞": 248728, "🗨": 249150, "🧩": 249234, "🈵": 248752, "🕍": 248719, "🐡": 248480, "🐜": 249249, "🪤": 248668, "🤳": 248541, "🧗": 248853, "🟪": 248787, "🧿": 249114, "📫": 248469, "🟫": 249255, "💴": 248931, "🍠": 248735, "🍟": 248748, "🥗": 248745, "🏮": 248644, "🎦": 248858, "👺": 249106, "🥃": 248788, "🧉": 248684, "🎣": 249137, "🥖": 248470, "🥳": 249128, "🤧": 249064, "🆒": 248798, "🧇": 248767, "🗝": 248629, "⛑": 248575, "👘": 248542, "🏎": 248992, "🦖": 248686, "📠": 249065, "🌫": 248734, "🕳": 248700, "🦐": 248950, "🍼": 248650, "🏥": 248835, "🪥": 248599, "🦫": 248566, "🚮": 248583, "🍩": 248935, "🔃": 249050, "🥩": 248928, "🟤": 248936, "📀": 248518, "🎽": 248577, "🕴": 248483, "🚓": 248817, "🧝": 248524, "📓": 248632, "🦧": 249030, "📇": 249007, "🐅": 249117, "🌡": 248904, "🤽": 248794, "🗂": 248631, "📳": 248471, "🩷": 248945, "🥤": 248949, "🤹": 248570, "🕗": 249221, "🕰": 248810, "🩳": 248970, "🧺": 248921, "🥯": 248757, "🐌": 248880, "🌤": 248812, "😧": 249227, "🧯": 249167, "🧎": 248537, "🤺": 248443, "🚽": 248726, "🙎": 248808, "🌜": 248635, "👝": 248709, "🐏": 248865, "🧙": 248605, "🚂": 248504, "🪟": 249226, "🚡": 249020, "🏁": 249031, "🔌": 248842, "🔓": 249058, "🏏": 248515, "🚣": 249186, "🐆": 249026, "🧕": 249176, "🚳": 248758, "🕑": 248818, "📼": 249230, "🤜": 249152, "🥔": 248905, "🚄": 248648, "🪷": 249036, "🛠": 248966, "🍜": 248979, "🪲": 248522, "⛸": 248693, "🔦": 248813, "💒": 248594, "🧾": 248827, "🔡": 248600, "♌": 249062, "🦣": 248578, "🥞": 248453, "🧅": 248895, "🏐": 248660, "🎏": 248670, "🩸": 248839, "👛": 249217, "🕡": 248982, "🚈": 248901, "🌁": 248694, "✝": 248664, "🥑": 248511, "🚷": 249149, "🛻": 249017, "🛼": 248991, "🧖": 249025, "🕥": 248627, "🕒": 248485, "🔽": 248581, "👂": 249140, "🔐": 248546, "🗃": 248497, "🏚": 248888, "🧐": 249095, "🫳": 249228, "🟢": 248489, "🧶": 248651, "🕶": 248688, "🌧": 248666, "🌘": 248782, "🪄": 248653, "🅿": 248755, "⬜": 248821, "🧧": 248428, "🤯": 249096, "🧑": 248672, "🐨": 249019, "🀄": 248929, "🏟": 248815, "🚑": 249188, "🐺": 249055, "🦮": 248517, "🎟": 248802, "🦕": 248654, "🥨": 248799, "🥕": 248838, "🔯": 248674, "🏈": 248891, "🥈": 248899, "🪺": 248646, "🖼": 248506, "🫛": 248698, "🖱": 248784, "🪧": 248535, "🧈": 249211, "🦂": 249034, "⛽": 249248, "🔤": 248997, "🎪": 249086, "〽": 248791, "🎱": 248961, "🎺": 248586, "♑": 248994, "🫓": 248743, "⛲": 249141, "🚤": 249204, "👓": 249170, "📏": 248556, "🚻": 248636, "🫵": 248531, "🎑": 248954, "🐔": 249068, "⛴": 248779, "🧽": 248545, "🈹": 248690, "🉑": 249224, "🔟": 249189, "🐚": 248886, "💷": 248711, "🆙": 248978, "🤿": 249191, "🚥": 249164, "🆎": 248619, "🗯": 248585, "🦒": 248843, "🟩": 248439, "🩼": 249127, "🍣": 248744, "🥋": 248772, "🧲": 249082, "🐮": 249074, "🗑": 249088, "🏬": 248996, "🪂": 248448, "🧸": 248488, "🦦": 249161, "🕝": 248763, "🧪": 249192, "🦽": 248959, "🔂": 248863, "🫐": 248500, "🏴": 248501, "🪴": 248457, "📄": 248801, "🥱": 249244, "📐": 248679, "🦾": 248507, "🧔": 249110, "🫤": 248986, "🚍": 249016, "➿": 248458, "🦆": 249042, "🫡": 248968, "🛺": 248569, "🧃": 249125, "🤥": 248628, "🕛": 248889, "🖲": 248766, "⛪": 249120, "🦃": 248834, "🫂": 248606, "😾": 248675, "🛷": 248824, "📒": 248710, "🈲": 248512, "🙍": 249066, "🪗": 249006, "🛃": 248732, "🥛": 248582, "🫲": 248924, "🏊": 249052, "🌩": 248795, "🌯": 248716, "🟧": 249166, "🪆": 248616, "🎿": 248915, "🎠": 249005, "🔧": 248618, "⚖": 248534, "🌃": 248739, "🧂": 249018, "⚒": 248845, "🗿": 249135, "🫑": 248456, "🦜": 248816, "🕷": 248437, "🕐": 248855, "🎳": 249073, "🩵": 248989, "🦢": 248965, "💿": 248604, "🪿": 249071, "💤": 248466, "🦎": 248911, "🏅": 249063, "🤮": 248819, "🦏": 248555, "🫚": 248800, "🚱": 248786, "🦙": 248523, "💂": 248778, "📙": 249160, "🛎": 249113, "🧭": 249180, "🐋": 249208, "🪑": 249185, "🦩": 248624, "🐎": 248479, "🧰": 248593, "🗄": 248980, "⚗": 248943, "📵": 249013, "🌗": 248478, "🦝": 248661, "🛴": 249111, "🥅": 249101, "🆚": 248972, "🪰": 248829, "🦼": 248639, "🥬": 249237, "🌪": 249022, "🕎": 248626, "🕟": 248884, "🎐": 249014, "🎆": 248678, "🕧": 248482, "🧻": 249236, "🦰": 249132, "🦯": 249089, "🏹": 249130, "🈚": 249053, "🦟": 248753, "🌛": 248984, "♾": 248725, "🌰": 248872, "⛏": 248490, "😼": 248614, "🎍": 248892, "🪓": 249028, "🔋": 248907, "🗡": 249134, "🍶": 248814, "🌨": 248730, "🏺": 249250, "🧀": 248712, "👷": 249146, "🗳": 248461, "🪀": 249216, "🎛": 248704, "⏫": 248890, "🦛": 249080, "🪇": 249238, "Ⓜ": 248957, "🥹": 248431, "🥟": 249033, "🚖": 248667, "🎹": 248503, "🪭": 248926, "📨": 249196, "🖊": 248760, "📴": 249004, "🉐": 249070, "📯": 249220, "🚿": 248941, "🩲": 249171, "🐗": 248750, "🐃": 248695, "🕤": 248682, "🖕": 249105, "🫷": 249091, "⁉": 248850, "🧞": 248558, "🕸": 248749, "🦚": 248797, "🈺": 248472, "🎷": 248451, "💽": 249021, "🍖": 248742, "🔚": 248903, "🥮": 248446, "🥭": 248634, "🫅": 248846, "📟": 248727, "🦉": 249008, "🪶": 249151, "♏": 249129, "🍮": 249067, "🕓": 248465, "🌥": 248881, "🤠": 248862, "📗": 248807, "⚰": 248930, "💺": 248831, "🫘": 249093, "🥝": 248848, "⛎": 248516, "🔇": 249009, "🐪": 248775, "🪃": 248689, "🛢": 249240, "🦤": 248662, "🌑": 249069, "🩱": 248529, "🤌": 248620, "🥽": 248875, "🟦": 248491, "♿": 248454, "🥉": 248762, "🏓": 248459, "🦪": 248738, "🫧": 248548, "🪪": 249223, "🔙": 249206, "🐖": 248603, "🌕": 248452, "🫴": 248987, "🛹": 248669, "🕘": 248552, "🐈": 248527, "🪩": 248962, "🧛": 248887, "🈁": 248495, "📶": 248969, "🪫": 248990, "🪙": 248476, "🛳": 249123, "📻": 249084, "🍢": 248833, "🪣": 248955, "📡": 248532, "🗜": 248764, "🫥": 249043, "🧬": 248681, "🫸": 248870, "🌖": 249242, "🥒": 249229, "🦿": 249222, "🏨": 248561, "🔣": 248579, "📘": 248708, "🚅": 249102, "🚠": 248640, "➰": 248665, "🐹": 248441, "🧷": 248977, "🐄": 249194}

config.json

@@ -0,0 +1,27 @@
+{
+  "_name_or_path": "../ConGen-Thai/phayathaibert",
+  "architectures": [
+    "CamembertModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "camembert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.11.0",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 249262
+}

config_sentence_transformers.json

@@ -0,0 +1,7 @@
+{
+  "__version__": {
+    "sentence_transformers": "2.1.0",
+    "transformers": "4.11.0",
+    "pytorch": "1.8.1+cu111"
+  }
+}

modules.json

@@ -0,0 +1,20 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  },
+  {
+    "idx": 2,
+    "name": "2",
+    "path": "2_Dense",
+    "type": "sentence_transformers.models.Dense"
+  }
+]

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5023b72e9d6e4a863f24248292927b23f59d58cd79f8c37c623c8dc802330ef
+size 1109981751

sentence_bert_config.json

@@ -0,0 +1,4 @@
+{
+  "max_seq_length": 128,
+  "do_lower_case": false
+}

sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e295c936bc0d8b6669ae769a2f8a0363e6d3abcfd8d0869134aa1e903a447d26
+size 5261686

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true}, "additional_special_tokens": ["<s>NOTUSED", "</s>NOTUSED", "<_>"]}

tokenizer_config.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "additional_special_tokens": ["<s>NOTUSED", "</s>NOTUSED", "<_>"], "clean_up_tokenization_spaces": true, "model_max_length": 510, "sp_model_kwargs": {}, "special_tokens_map_file": "../ConGen-Thai/phayathaibert/special_tokens_map.json", "name_or_path": "../ConGen-Thai/phayathaibert", "tokenizer_class": "CamembertTokenizer"}