Commit
·
482d17b
1
Parent(s):
a0cfb98
upload
Browse files- config.json +12 -8
- pytorch_model.bin +1 -1
config.json
CHANGED
|
@@ -2,8 +2,7 @@
|
|
| 2 |
"architectures": [
|
| 3 |
"SpeechEncoderDecoderModel"
|
| 4 |
],
|
| 5 |
-
"decoder": {
|
| 6 |
-
"_name_or_path": "/home/suraj/projects/mbart-50/hf_models/mbart-50-large-one-to-many/",
|
| 7 |
"_num_labels": 3,
|
| 8 |
"activation_dropout": 0.0,
|
| 9 |
"activation_function": "relu",
|
|
@@ -94,13 +93,13 @@
|
|
| 94 |
"use_cache": true,
|
| 95 |
"vocab_size": 250054
|
| 96 |
},
|
| 97 |
-
"decoder_start_token_id":
|
| 98 |
"encoder": {
|
| 99 |
"_name_or_path": "",
|
| 100 |
"activation_dropout": 0.0,
|
| 101 |
-
"
|
| 102 |
-
"
|
| 103 |
-
"
|
| 104 |
"add_cross_attention": false,
|
| 105 |
"apply_spec_augment": true,
|
| 106 |
"architectures": [
|
|
@@ -188,7 +187,7 @@
|
|
| 188 |
"min_length": 0,
|
| 189 |
"model_type": "wav2vec2",
|
| 190 |
"no_repeat_ngram_size": 0,
|
| 191 |
-
"
|
| 192 |
"num_attention_heads": 16,
|
| 193 |
"num_beam_groups": 1,
|
| 194 |
"num_beams": 1,
|
|
@@ -201,6 +200,7 @@
|
|
| 201 |
"num_negatives": 100,
|
| 202 |
"num_return_sequences": 1,
|
| 203 |
"output_attentions": false,
|
|
|
|
| 204 |
"output_hidden_states": false,
|
| 205 |
"output_scores": false,
|
| 206 |
"pad_token_id": 0,
|
|
@@ -229,11 +229,15 @@
|
|
| 229 |
},
|
| 230 |
"eos_token_id": 2,
|
| 231 |
"feature_extractor_type": "wav2vec2",
|
|
|
|
|
|
|
| 232 |
"is_encoder_decoder": true,
|
| 233 |
"model_type": "speech-encoder-decoder",
|
| 234 |
"pad_token_id": 1,
|
| 235 |
"tie_word_embeddings": false,
|
| 236 |
"tokenizer_class": "mbart50",
|
| 237 |
"torch_dtype": "float32",
|
| 238 |
-
"transformers_version": null
|
|
|
|
|
|
|
| 239 |
}
|
|
|
|
| 2 |
"architectures": [
|
| 3 |
"SpeechEncoderDecoderModel"
|
| 4 |
],
|
| 5 |
+
"decoder": { "_name_or_path": "/home/suraj/projects/mbart-50/hf_models/mbart-50-large-one-to-many/",
|
|
|
|
| 6 |
"_num_labels": 3,
|
| 7 |
"activation_dropout": 0.0,
|
| 8 |
"activation_function": "relu",
|
|
|
|
| 93 |
"use_cache": true,
|
| 94 |
"vocab_size": 250054
|
| 95 |
},
|
| 96 |
+
"decoder_start_token_id": 2,
|
| 97 |
"encoder": {
|
| 98 |
"_name_or_path": "",
|
| 99 |
"activation_dropout": 0.0,
|
| 100 |
+
"adapter_kernel_size": 3,
|
| 101 |
+
"adapter_stride": 2,
|
| 102 |
+
"add_adapter": true,
|
| 103 |
"add_cross_attention": false,
|
| 104 |
"apply_spec_augment": true,
|
| 105 |
"architectures": [
|
|
|
|
| 187 |
"min_length": 0,
|
| 188 |
"model_type": "wav2vec2",
|
| 189 |
"no_repeat_ngram_size": 0,
|
| 190 |
+
"num_adapter_layers": 3,
|
| 191 |
"num_attention_heads": 16,
|
| 192 |
"num_beam_groups": 1,
|
| 193 |
"num_beams": 1,
|
|
|
|
| 200 |
"num_negatives": 100,
|
| 201 |
"num_return_sequences": 1,
|
| 202 |
"output_attentions": false,
|
| 203 |
+
"output_hidden_size": 1024,
|
| 204 |
"output_hidden_states": false,
|
| 205 |
"output_scores": false,
|
| 206 |
"pad_token_id": 0,
|
|
|
|
| 229 |
},
|
| 230 |
"eos_token_id": 2,
|
| 231 |
"feature_extractor_type": "wav2vec2",
|
| 232 |
+
"forced_bos_token_id": 250004,
|
| 233 |
+
"forced_eos_token_id": 2,
|
| 234 |
"is_encoder_decoder": true,
|
| 235 |
"model_type": "speech-encoder-decoder",
|
| 236 |
"pad_token_id": 1,
|
| 237 |
"tie_word_embeddings": false,
|
| 238 |
"tokenizer_class": "mbart50",
|
| 239 |
"torch_dtype": "float32",
|
| 240 |
+
"transformers_version": null,
|
| 241 |
+
"num_beams": 5,
|
| 242 |
+
"max_length": 200
|
| 243 |
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3172257129
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:02b827faddcf3cc3f0412f90423f5215e49f4dd42ef5765affd297e495334679
|
| 3 |
size 3172257129
|