Training in progress, epoch 1
Browse files- config.json +3 -3
- model.safetensors +2 -2
- preprocessor_config.json +2 -2
- training_args.bin +2 -2
- vocab.json +1 -149
config.json
CHANGED
|
@@ -76,7 +76,7 @@
|
|
| 76 |
"num_hidden_layers": 48,
|
| 77 |
"num_negatives": 100,
|
| 78 |
"output_hidden_size": 1280,
|
| 79 |
-
"pad_token_id":
|
| 80 |
"proj_codevector_dim": 1024,
|
| 81 |
"tdnn_dilation": [
|
| 82 |
1,
|
|
@@ -99,9 +99,9 @@
|
|
| 99 |
1,
|
| 100 |
1
|
| 101 |
],
|
| 102 |
-
"torch_dtype": "
|
| 103 |
"transformers_version": "4.50.3",
|
| 104 |
"use_weighted_layer_sum": false,
|
| 105 |
-
"vocab_size":
|
| 106 |
"xvector_output_dim": 512
|
| 107 |
}
|
|
|
|
| 76 |
"num_hidden_layers": 48,
|
| 77 |
"num_negatives": 100,
|
| 78 |
"output_hidden_size": 1280,
|
| 79 |
+
"pad_token_id": 0,
|
| 80 |
"proj_codevector_dim": 1024,
|
| 81 |
"tdnn_dilation": [
|
| 82 |
1,
|
|
|
|
| 99 |
1,
|
| 100 |
1
|
| 101 |
],
|
| 102 |
+
"torch_dtype": "float32",
|
| 103 |
"transformers_version": "4.50.3",
|
| 104 |
"use_weighted_layer_sum": false,
|
| 105 |
+
"vocab_size": 121,
|
| 106 |
"xvector_output_dim": 512
|
| 107 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f76d4e30618e2bf7900aa7b5926be1dda0b8fc2261323751f36fa19efd32c5ed
|
| 3 |
+
size 3859352084
|
preprocessor_config.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"cache_dir": "/l/users/karima.kadaoui/.cache/huggingface/datasets",
|
| 3 |
"do_normalize": true,
|
| 4 |
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 5 |
"feature_size": 1,
|
| 6 |
"padding_side": "right",
|
| 7 |
-
"padding_value": 0
|
|
|
|
| 8 |
"return_attention_mask": true,
|
| 9 |
"sampling_rate": 16000
|
| 10 |
}
|
|
|
|
| 1 |
{
|
|
|
|
| 2 |
"do_normalize": true,
|
| 3 |
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
"feature_size": 1,
|
| 5 |
"padding_side": "right",
|
| 6 |
+
"padding_value": 0,
|
| 7 |
+
"processor_class": "Wav2Vec2Processor",
|
| 8 |
"return_attention_mask": true,
|
| 9 |
"sampling_rate": 16000
|
| 10 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:91933d355eab57566d8a75d093e7e84f1fe5586de25f0537ec24104ad4f4cbd2
|
| 3 |
+
size 5368
|
vocab.json
CHANGED
|
@@ -1,149 +1 @@
|
|
| 1 |
-
{
|
| 2 |
-
"ara": {
|
| 3 |
-
"!": 1,
|
| 4 |
-
"\"": 2,
|
| 5 |
-
"#": 3,
|
| 6 |
-
"%": 4,
|
| 7 |
-
"&": 5,
|
| 8 |
-
"'": 6,
|
| 9 |
-
"(": 7,
|
| 10 |
-
")": 8,
|
| 11 |
-
"+": 9,
|
| 12 |
-
",": 10,
|
| 13 |
-
"-": 11,
|
| 14 |
-
".": 12,
|
| 15 |
-
"/": 13,
|
| 16 |
-
"0": 14,
|
| 17 |
-
"1": 15,
|
| 18 |
-
"2": 16,
|
| 19 |
-
"3": 17,
|
| 20 |
-
"4": 18,
|
| 21 |
-
"5": 19,
|
| 22 |
-
"6": 20,
|
| 23 |
-
"7": 21,
|
| 24 |
-
"8": 22,
|
| 25 |
-
"9": 23,
|
| 26 |
-
":": 24,
|
| 27 |
-
";": 25,
|
| 28 |
-
"?": 26,
|
| 29 |
-
"@": 27,
|
| 30 |
-
"A": 28,
|
| 31 |
-
"B": 29,
|
| 32 |
-
"C": 30,
|
| 33 |
-
"D": 31,
|
| 34 |
-
"E": 32,
|
| 35 |
-
"F": 33,
|
| 36 |
-
"G": 34,
|
| 37 |
-
"H": 35,
|
| 38 |
-
"I": 36,
|
| 39 |
-
"J": 37,
|
| 40 |
-
"K": 38,
|
| 41 |
-
"L": 39,
|
| 42 |
-
"M": 40,
|
| 43 |
-
"N": 41,
|
| 44 |
-
"O": 42,
|
| 45 |
-
"P": 43,
|
| 46 |
-
"Q": 44,
|
| 47 |
-
"R": 45,
|
| 48 |
-
"S": 46,
|
| 49 |
-
"T": 47,
|
| 50 |
-
"U": 48,
|
| 51 |
-
"V": 49,
|
| 52 |
-
"W": 50,
|
| 53 |
-
"X": 51,
|
| 54 |
-
"Y": 52,
|
| 55 |
-
"Z": 53,
|
| 56 |
-
"[PAD]": 144,
|
| 57 |
-
"[UNK]": 143,
|
| 58 |
-
"\\": 54,
|
| 59 |
-
"a": 55,
|
| 60 |
-
"b": 56,
|
| 61 |
-
"c": 57,
|
| 62 |
-
"d": 58,
|
| 63 |
-
"e": 59,
|
| 64 |
-
"f": 60,
|
| 65 |
-
"g": 61,
|
| 66 |
-
"h": 62,
|
| 67 |
-
"i": 63,
|
| 68 |
-
"j": 64,
|
| 69 |
-
"k": 65,
|
| 70 |
-
"l": 66,
|
| 71 |
-
"m": 67,
|
| 72 |
-
"n": 68,
|
| 73 |
-
"o": 69,
|
| 74 |
-
"p": 70,
|
| 75 |
-
"q": 71,
|
| 76 |
-
"r": 72,
|
| 77 |
-
"s": 73,
|
| 78 |
-
"t": 74,
|
| 79 |
-
"u": 75,
|
| 80 |
-
"v": 76,
|
| 81 |
-
"w": 77,
|
| 82 |
-
"x": 78,
|
| 83 |
-
"y": 79,
|
| 84 |
-
"z": 80,
|
| 85 |
-
"|": 0,
|
| 86 |
-
"·": 81,
|
| 87 |
-
"à": 82,
|
| 88 |
-
"å": 83,
|
| 89 |
-
"ç": 84,
|
| 90 |
-
"è": 85,
|
| 91 |
-
"é": 86,
|
| 92 |
-
"ê": 87,
|
| 93 |
-
"û": 88,
|
| 94 |
-
"İ": 89,
|
| 95 |
-
"،": 90,
|
| 96 |
-
"؛": 91,
|
| 97 |
-
"؟": 92,
|
| 98 |
-
"ء": 93,
|
| 99 |
-
"آ": 94,
|
| 100 |
-
"أ": 95,
|
| 101 |
-
"ؤ": 96,
|
| 102 |
-
"إ": 97,
|
| 103 |
-
"ئ": 98,
|
| 104 |
-
"ا": 99,
|
| 105 |
-
"ب": 100,
|
| 106 |
-
"ة": 101,
|
| 107 |
-
"ت": 102,
|
| 108 |
-
"ث": 103,
|
| 109 |
-
"ج": 104,
|
| 110 |
-
"ح": 105,
|
| 111 |
-
"خ": 106,
|
| 112 |
-
"د": 107,
|
| 113 |
-
"ذ": 108,
|
| 114 |
-
"ر": 109,
|
| 115 |
-
"ز": 110,
|
| 116 |
-
"س": 111,
|
| 117 |
-
"ش": 112,
|
| 118 |
-
"ص": 113,
|
| 119 |
-
"ض": 114,
|
| 120 |
-
"ط": 115,
|
| 121 |
-
"ظ": 116,
|
| 122 |
-
"ع": 117,
|
| 123 |
-
"غ": 118,
|
| 124 |
-
"ـ": 119,
|
| 125 |
-
"ف": 120,
|
| 126 |
-
"ق": 121,
|
| 127 |
-
"ك": 122,
|
| 128 |
-
"ل": 123,
|
| 129 |
-
"م": 124,
|
| 130 |
-
"ن": 125,
|
| 131 |
-
"ه": 126,
|
| 132 |
-
"و": 127,
|
| 133 |
-
"ى": 128,
|
| 134 |
-
"ي": 129,
|
| 135 |
-
"ً": 130,
|
| 136 |
-
"ٌ": 131,
|
| 137 |
-
"ٍ": 132,
|
| 138 |
-
"َ": 133,
|
| 139 |
-
"ُ": 134,
|
| 140 |
-
"ِ": 135,
|
| 141 |
-
"ّ": 136,
|
| 142 |
-
"٣": 137,
|
| 143 |
-
"٥": 138,
|
| 144 |
-
"٧": 139,
|
| 145 |
-
"’": 140,
|
| 146 |
-
"“": 141,
|
| 147 |
-
"”": 142
|
| 148 |
-
}
|
| 149 |
-
}
|
|
|
|
| 1 |
+
{"ara": {"!": 1, "\"": 2, "#": 3, "%": 4, "&": 5, "'": 6, "(": 7, ")": 8, "+": 9, ",": 10, "-": 11, ".": 12, "/": 13, "0": 14, "1": 15, "2": 16, "3": 17, "4": 18, "5": 19, "6": 20, "7": 21, "8": 22, "9": 23, ":": 24, ";": 25, "?": 26, "@": 27, "A": 28, "B": 29, "C": 30, "D": 31, "E": 32, "F": 33, "G": 34, "H": 35, "I": 36, "J": 37, "K": 38, "L": 39, "M": 40, "N": 41, "O": 42, "P": 43, "Q": 44, "R": 45, "S": 46, "T": 47, "U": 48, "V": 49, "W": 50, "X": 51, "Y": 52, "Z": 53, "\\": 54, "a": 55, "b": 56, "c": 57, "d": 58, "e": 59, "f": 60, "g": 61, "h": 62, "i": 63, "j": 64, "k": 65, "l": 66, "m": 67, "n": 68, "o": 69, "p": 70, "q": 71, "r": 72, "s": 73, "t": 74, "u": 75, "v": 76, "w": 77, "x": 78, "y": 79, "z": 80, "\u00b7": 81, "\u00e0": 82, "\u00e5": 83, "\u00e7": 84, "\u00e8": 85, "\u00e9": 86, "\u00ea": 87, "\u00fb": 88, "\u0130": 89, "\u060c": 90, "\u061b": 91, "\u061f": 92, "\u0621": 93, "\u0622": 94, "\u0623": 95, "\u0624": 96, "\u0625": 97, "\u0626": 98, "\u0627": 99, "\u0628": 100, "\u0629": 101, "\u062a": 102, "\u062b": 103, "\u062c": 104, "\u062d": 105, "\u062e": 106, "\u062f": 107, "\u0630": 108, "\u0631": 109, "\u0632": 110, "\u0633": 111, "\u0634": 112, "\u0635": 113, "\u0636": 114, "\u0637": 115, "\u0638": 116, "\u0639": 117, "\u063a": 118, "\u0640": 119, "\u0641": 120, "\u0642": 121, "\u0643": 122, "\u0644": 123, "\u0645": 124, "\u0646": 125, "\u0647": 126, "\u0648": 127, "\u0649": 128, "\u064a": 129, "\u064b": 130, "\u064c": 131, "\u064d": 132, "\u064e": 133, "\u064f": 134, "\u0650": 135, "\u0651": 136, "\u0663": 137, "\u0665": 138, "\u0667": 139, "\u2019": 140, "\u201c": 141, "\u201d": 142, "|": 0, "[UNK]": 143, "[PAD]": 144}}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|