Training in progress, epoch 1
Browse files- config.json +62 -0
- model.safetensors +3 -0
- special_tokens_map.json +12 -0
- tokenizer.json +139 -0
- tokenizer_config.json +43 -0
- training_args.bin +3 -0
config.json
ADDED
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "Helsinki-NLP/opus-mt-en-de",
|
3 |
+
"_num_labels": 3,
|
4 |
+
"activation_dropout": 0.0,
|
5 |
+
"activation_function": "swish",
|
6 |
+
"add_bias_logits": false,
|
7 |
+
"add_final_layer_norm": false,
|
8 |
+
"architectures": [
|
9 |
+
"MarianMTModel"
|
10 |
+
],
|
11 |
+
"attention_dropout": 0.0,
|
12 |
+
"bad_words_ids": [
|
13 |
+
[
|
14 |
+
58100
|
15 |
+
]
|
16 |
+
],
|
17 |
+
"bos_token_id": 0,
|
18 |
+
"classif_dropout": 0.0,
|
19 |
+
"classifier_dropout": 0.0,
|
20 |
+
"d_model": 512,
|
21 |
+
"decoder_attention_heads": 8,
|
22 |
+
"decoder_ffn_dim": 2048,
|
23 |
+
"decoder_layerdrop": 0.0,
|
24 |
+
"decoder_layers": 6,
|
25 |
+
"decoder_start_token_id": 58100,
|
26 |
+
"decoder_vocab_size": 58101,
|
27 |
+
"dropout": 0.1,
|
28 |
+
"encoder_attention_heads": 8,
|
29 |
+
"encoder_ffn_dim": 2048,
|
30 |
+
"encoder_layerdrop": 0.0,
|
31 |
+
"encoder_layers": 6,
|
32 |
+
"eos_token_id": 0,
|
33 |
+
"forced_eos_token_id": 0,
|
34 |
+
"gradient_checkpointing": false,
|
35 |
+
"id2label": {
|
36 |
+
"0": "LABEL_0",
|
37 |
+
"1": "LABEL_1",
|
38 |
+
"2": "LABEL_2"
|
39 |
+
},
|
40 |
+
"init_std": 0.02,
|
41 |
+
"is_encoder_decoder": true,
|
42 |
+
"label2id": {
|
43 |
+
"LABEL_0": 0,
|
44 |
+
"LABEL_1": 1,
|
45 |
+
"LABEL_2": 2
|
46 |
+
},
|
47 |
+
"max_length": 512,
|
48 |
+
"max_position_embeddings": 512,
|
49 |
+
"model_type": "marian",
|
50 |
+
"normalize_before": false,
|
51 |
+
"normalize_embedding": false,
|
52 |
+
"num_beams": 4,
|
53 |
+
"num_hidden_layers": 6,
|
54 |
+
"pad_token_id": 58100,
|
55 |
+
"scale_embedding": true,
|
56 |
+
"share_encoder_decoder_embeddings": true,
|
57 |
+
"static_position_embeddings": true,
|
58 |
+
"torch_dtype": "float32",
|
59 |
+
"transformers_version": "4.41.1",
|
60 |
+
"use_cache": true,
|
61 |
+
"vocab_size": 58101
|
62 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8fe58ca31c92a7d5fad3a3321b41d12373faf1e3b663a09cc19c812627730e8f
|
3 |
+
size 295806292
|
special_tokens_map.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cls_token": "<s>",
|
3 |
+
"pad_token": {
|
4 |
+
"content": "[PAD]",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false
|
9 |
+
},
|
10 |
+
"sep_token": "</s>",
|
11 |
+
"unk_token": "[UNK]"
|
12 |
+
}
|
tokenizer.json
ADDED
@@ -0,0 +1,139 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"version": "1.0",
|
3 |
+
"truncation": {
|
4 |
+
"direction": "Right",
|
5 |
+
"max_length": 128,
|
6 |
+
"strategy": "LongestFirst",
|
7 |
+
"stride": 0
|
8 |
+
},
|
9 |
+
"padding": {
|
10 |
+
"strategy": {
|
11 |
+
"Fixed": 128
|
12 |
+
},
|
13 |
+
"direction": "Right",
|
14 |
+
"pad_to_multiple_of": null,
|
15 |
+
"pad_id": 70,
|
16 |
+
"pad_type_id": 0,
|
17 |
+
"pad_token": "[PAD]"
|
18 |
+
},
|
19 |
+
"added_tokens": [
|
20 |
+
{
|
21 |
+
"id": 0,
|
22 |
+
"content": "[UNK]",
|
23 |
+
"single_word": false,
|
24 |
+
"lstrip": false,
|
25 |
+
"rstrip": false,
|
26 |
+
"normalized": false,
|
27 |
+
"special": true
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"id": 1,
|
31 |
+
"content": "<s>",
|
32 |
+
"single_word": false,
|
33 |
+
"lstrip": false,
|
34 |
+
"rstrip": false,
|
35 |
+
"normalized": false,
|
36 |
+
"special": true
|
37 |
+
},
|
38 |
+
{
|
39 |
+
"id": 2,
|
40 |
+
"content": "</s>",
|
41 |
+
"single_word": false,
|
42 |
+
"lstrip": false,
|
43 |
+
"rstrip": false,
|
44 |
+
"normalized": false,
|
45 |
+
"special": true
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"id": 70,
|
49 |
+
"content": "[PAD]",
|
50 |
+
"single_word": false,
|
51 |
+
"lstrip": false,
|
52 |
+
"rstrip": false,
|
53 |
+
"normalized": false,
|
54 |
+
"special": true
|
55 |
+
}
|
56 |
+
],
|
57 |
+
"normalizer": null,
|
58 |
+
"pre_tokenizer": {
|
59 |
+
"type": "Whitespace"
|
60 |
+
},
|
61 |
+
"post_processor": null,
|
62 |
+
"decoder": null,
|
63 |
+
"model": {
|
64 |
+
"type": "WordLevel",
|
65 |
+
"vocab": {
|
66 |
+
"[UNK]": 0,
|
67 |
+
"<s>": 1,
|
68 |
+
"</s>": 2,
|
69 |
+
"\"": 3,
|
70 |
+
"{{": 4,
|
71 |
+
"\":\"": 5,
|
72 |
+
",": 6,
|
73 |
+
"}": 7,
|
74 |
+
"\":": 8,
|
75 |
+
"name": 9,
|
76 |
+
"{": 10,
|
77 |
+
"\":{": 11,
|
78 |
+
"id": 12,
|
79 |
+
"ktl": 13,
|
80 |
+
"pli": 14,
|
81 |
+
"translation": 15,
|
82 |
+
"}}": 16,
|
83 |
+
"},": 17,
|
84 |
+
"}}\",": 18,
|
85 |
+
"value": 19,
|
86 |
+
"}}\"": 20,
|
87 |
+
":": 21,
|
88 |
+
"param0": 22,
|
89 |
+
"type0": 23,
|
90 |
+
"\":\"{{": 24,
|
91 |
+
"PROCEDURE": 25,
|
92 |
+
"fun": 26,
|
93 |
+
"return": 27,
|
94 |
+
"\",": 28,
|
95 |
+
"CALL": 29,
|
96 |
+
"type": 30,
|
97 |
+
"type1": 31,
|
98 |
+
"}}()\"": 32,
|
99 |
+
"}}({{": 33,
|
100 |
+
"}})\"": 34,
|
101 |
+
"\":\"{\"": 35,
|
102 |
+
"\":\"}\"": 36,
|
103 |
+
"(": 37,
|
104 |
+
"({{": 38,
|
105 |
+
"0": 39,
|
106 |
+
"1": 40,
|
107 |
+
"10": 41,
|
108 |
+
"2": 42,
|
109 |
+
"3": 43,
|
110 |
+
"4": 44,
|
111 |
+
"5": 45,
|
112 |
+
"6": 46,
|
113 |
+
"7": 47,
|
114 |
+
"8": 48,
|
115 |
+
"9": 49,
|
116 |
+
"<{{": 50,
|
117 |
+
"=": 51,
|
118 |
+
"ASSIGN": 52,
|
119 |
+
"DECLARE": 53,
|
120 |
+
"DO": 54,
|
121 |
+
"END": 55,
|
122 |
+
"IF": 56,
|
123 |
+
"MAIN": 57,
|
124 |
+
"RETURN": 58,
|
125 |
+
"THEN": 59,
|
126 |
+
"[": 60,
|
127 |
+
"]": 61,
|
128 |
+
"args": 62,
|
129 |
+
"data": 63,
|
130 |
+
"if": 64,
|
131 |
+
"main": 65,
|
132 |
+
"var": 66,
|
133 |
+
"}})": 67,
|
134 |
+
"}},{{": 68,
|
135 |
+
"}}>)\"": 69
|
136 |
+
},
|
137 |
+
"unk_token": "[UNK]"
|
138 |
+
}
|
139 |
+
}
|
tokenizer_config.json
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "[UNK]",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"1": {
|
12 |
+
"content": "<s>",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"2": {
|
20 |
+
"content": "</s>",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"70": {
|
28 |
+
"content": "[PAD]",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
}
|
35 |
+
},
|
36 |
+
"clean_up_tokenization_spaces": true,
|
37 |
+
"cls_token": "<s>",
|
38 |
+
"model_max_length": 1000000000000000019884624838656,
|
39 |
+
"pad_token": "[PAD]",
|
40 |
+
"sep_token": "</s>",
|
41 |
+
"tokenizer_class": "PreTrainedTokenizerFast",
|
42 |
+
"unk_token": "[UNK]"
|
43 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:187d230b4d9c34fab1ea69380f0ea5922f9a30d4c63bf026c4f1cb31425789d7
|
3 |
+
size 5240
|