Joshua Lochner
commited on
Commit
·
a9f8f3e
1
Parent(s):
ef7da33
Add Helsinki-NLP model for ru-en translation
Browse files- quantized/Helsinki-NLP/opus-mt-ru-en/default/config.json +60 -0
- quantized/Helsinki-NLP/opus-mt-ru-en/default/model.onnx +3 -0
- quantized/Helsinki-NLP/opus-mt-ru-en/default/special_tokens_map.json +5 -0
- quantized/Helsinki-NLP/opus-mt-ru-en/default/tokenizer.json +0 -0
- quantized/Helsinki-NLP/opus-mt-ru-en/default/tokenizer_config.json +12 -0
- quantized/Helsinki-NLP/opus-mt-ru-en/default/vocab.json +0 -0
- quantized/Helsinki-NLP/opus-mt-ru-en/seq2seq-lm-with-past/config.json +60 -0
- quantized/Helsinki-NLP/opus-mt-ru-en/seq2seq-lm-with-past/decoder_model.onnx +3 -0
- quantized/Helsinki-NLP/opus-mt-ru-en/seq2seq-lm-with-past/decoder_model_merged.onnx +3 -0
- quantized/Helsinki-NLP/opus-mt-ru-en/seq2seq-lm-with-past/decoder_with_past_model.onnx +3 -0
- quantized/Helsinki-NLP/opus-mt-ru-en/seq2seq-lm-with-past/encoder_model.onnx +3 -0
- quantized/Helsinki-NLP/opus-mt-ru-en/seq2seq-lm-with-past/generation_config.json +16 -0
- quantized/Helsinki-NLP/opus-mt-ru-en/seq2seq-lm-with-past/special_tokens_map.json +5 -0
- quantized/Helsinki-NLP/opus-mt-ru-en/seq2seq-lm-with-past/tokenizer.json +0 -0
- quantized/Helsinki-NLP/opus-mt-ru-en/seq2seq-lm-with-past/tokenizer_config.json +12 -0
- quantized/Helsinki-NLP/opus-mt-ru-en/seq2seq-lm-with-past/vocab.json +0 -0
quantized/Helsinki-NLP/opus-mt-ru-en/default/config.json
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Helsinki-NLP/opus-mt-ru-en",
|
| 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 |
+
62517
|
| 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": 62517,
|
| 26 |
+
"decoder_vocab_size": 62518,
|
| 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 |
+
"id2label": {
|
| 35 |
+
"0": "LABEL_0",
|
| 36 |
+
"1": "LABEL_1",
|
| 37 |
+
"2": "LABEL_2"
|
| 38 |
+
},
|
| 39 |
+
"init_std": 0.02,
|
| 40 |
+
"is_encoder_decoder": true,
|
| 41 |
+
"label2id": {
|
| 42 |
+
"LABEL_0": 0,
|
| 43 |
+
"LABEL_1": 1,
|
| 44 |
+
"LABEL_2": 2
|
| 45 |
+
},
|
| 46 |
+
"max_length": 512,
|
| 47 |
+
"max_position_embeddings": 512,
|
| 48 |
+
"model_type": "marian",
|
| 49 |
+
"normalize_before": false,
|
| 50 |
+
"normalize_embedding": false,
|
| 51 |
+
"num_beams": 6,
|
| 52 |
+
"num_hidden_layers": 6,
|
| 53 |
+
"pad_token_id": 62517,
|
| 54 |
+
"scale_embedding": true,
|
| 55 |
+
"share_encoder_decoder_embeddings": true,
|
| 56 |
+
"static_position_embeddings": true,
|
| 57 |
+
"transformers_version": "4.27.2",
|
| 58 |
+
"use_cache": true,
|
| 59 |
+
"vocab_size": 62518
|
| 60 |
+
}
|
quantized/Helsinki-NLP/opus-mt-ru-en/default/model.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b3bf349ae4eb3bd73c892bf25cf850ad3f127c69b61a9c4693602d93f9ea05ad
|
| 3 |
+
size 78651734
|
quantized/Helsinki-NLP/opus-mt-ru-en/default/special_tokens_map.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eos_token": "</s>",
|
| 3 |
+
"pad_token": "<pad>",
|
| 4 |
+
"unk_token": "<unk>"
|
| 5 |
+
}
|
quantized/Helsinki-NLP/opus-mt-ru-en/default/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
quantized/Helsinki-NLP/opus-mt-ru-en/default/tokenizer_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eos_token": "</s>",
|
| 3 |
+
"model_max_length": 512,
|
| 4 |
+
"pad_token": "<pad>",
|
| 5 |
+
"separate_vocabs": false,
|
| 6 |
+
"source_lang": "ru",
|
| 7 |
+
"sp_model_kwargs": {},
|
| 8 |
+
"special_tokens_map_file": null,
|
| 9 |
+
"target_lang": "en",
|
| 10 |
+
"tokenizer_class": "MarianTokenizer",
|
| 11 |
+
"unk_token": "<unk>"
|
| 12 |
+
}
|
quantized/Helsinki-NLP/opus-mt-ru-en/default/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
quantized/Helsinki-NLP/opus-mt-ru-en/seq2seq-lm-with-past/config.json
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Helsinki-NLP/opus-mt-ru-en",
|
| 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 |
+
62517
|
| 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": 62517,
|
| 26 |
+
"decoder_vocab_size": 62518,
|
| 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 |
+
"id2label": {
|
| 35 |
+
"0": "LABEL_0",
|
| 36 |
+
"1": "LABEL_1",
|
| 37 |
+
"2": "LABEL_2"
|
| 38 |
+
},
|
| 39 |
+
"init_std": 0.02,
|
| 40 |
+
"is_encoder_decoder": true,
|
| 41 |
+
"label2id": {
|
| 42 |
+
"LABEL_0": 0,
|
| 43 |
+
"LABEL_1": 1,
|
| 44 |
+
"LABEL_2": 2
|
| 45 |
+
},
|
| 46 |
+
"max_length": 512,
|
| 47 |
+
"max_position_embeddings": 512,
|
| 48 |
+
"model_type": "marian",
|
| 49 |
+
"normalize_before": false,
|
| 50 |
+
"normalize_embedding": false,
|
| 51 |
+
"num_beams": 6,
|
| 52 |
+
"num_hidden_layers": 6,
|
| 53 |
+
"pad_token_id": 62517,
|
| 54 |
+
"scale_embedding": true,
|
| 55 |
+
"share_encoder_decoder_embeddings": true,
|
| 56 |
+
"static_position_embeddings": true,
|
| 57 |
+
"transformers_version": "4.27.2",
|
| 58 |
+
"use_cache": true,
|
| 59 |
+
"vocab_size": 62518
|
| 60 |
+
}
|
quantized/Helsinki-NLP/opus-mt-ru-en/seq2seq-lm-with-past/decoder_model.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1c6ca147cd18bc7d70ab6175c9ee2b310fade970642a2b659cc7b9f2a7f0eeb7
|
| 3 |
+
size 90852795
|
quantized/Helsinki-NLP/opus-mt-ru-en/seq2seq-lm-with-past/decoder_model_merged.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:954c0307c7559d551a78825645f1c6fb3a60a995b665d9bf9173fa61f8a1c61b
|
| 3 |
+
size 91004032
|
quantized/Helsinki-NLP/opus-mt-ru-en/seq2seq-lm-with-past/decoder_with_past_model.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:356d75bcba7444e399924fc72c361c0aea639e4de635feb7d8e8489e8d08c9e9
|
| 3 |
+
size 87583366
|
quantized/Helsinki-NLP/opus-mt-ru-en/seq2seq-lm-with-past/encoder_model.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dfb64d896fe5567065f40f5f745508b2969045fccb98077c4fff6e4a3490c1fa
|
| 3 |
+
size 51628190
|
quantized/Helsinki-NLP/opus-mt-ru-en/seq2seq-lm-with-past/generation_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bad_words_ids": [
|
| 4 |
+
[
|
| 5 |
+
62517
|
| 6 |
+
]
|
| 7 |
+
],
|
| 8 |
+
"bos_token_id": 0,
|
| 9 |
+
"decoder_start_token_id": 62517,
|
| 10 |
+
"eos_token_id": 0,
|
| 11 |
+
"forced_eos_token_id": 0,
|
| 12 |
+
"max_length": 512,
|
| 13 |
+
"num_beams": 6,
|
| 14 |
+
"pad_token_id": 62517,
|
| 15 |
+
"transformers_version": "4.27.0.dev0"
|
| 16 |
+
}
|
quantized/Helsinki-NLP/opus-mt-ru-en/seq2seq-lm-with-past/special_tokens_map.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eos_token": "</s>",
|
| 3 |
+
"pad_token": "<pad>",
|
| 4 |
+
"unk_token": "<unk>"
|
| 5 |
+
}
|
quantized/Helsinki-NLP/opus-mt-ru-en/seq2seq-lm-with-past/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
quantized/Helsinki-NLP/opus-mt-ru-en/seq2seq-lm-with-past/tokenizer_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eos_token": "</s>",
|
| 3 |
+
"model_max_length": 512,
|
| 4 |
+
"pad_token": "<pad>",
|
| 5 |
+
"separate_vocabs": false,
|
| 6 |
+
"source_lang": "ru",
|
| 7 |
+
"sp_model_kwargs": {},
|
| 8 |
+
"special_tokens_map_file": null,
|
| 9 |
+
"target_lang": "en",
|
| 10 |
+
"tokenizer_class": "MarianTokenizer",
|
| 11 |
+
"unk_token": "<unk>"
|
| 12 |
+
}
|
quantized/Helsinki-NLP/opus-mt-ru-en/seq2seq-lm-with-past/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|