Commit
·
5083c21
1
Parent(s):
d708f56
replacing 16 files
Browse files- config.json +24 -11
- preprocessor_config.json +1 -1
config.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "
|
| 3 |
"activation_dropout": 0.0,
|
| 4 |
"activation_function": "gelu",
|
| 5 |
"apply_spec_augment": false,
|
| 6 |
"architectures": [
|
| 7 |
-
"
|
| 8 |
],
|
| 9 |
"attention_dropout": 0.0,
|
| 10 |
"begin_suppress_tokens": [
|
|
@@ -14,18 +14,31 @@
|
|
| 14 |
"bos_token_id": 50257,
|
| 15 |
"classifier_proj_size": 256,
|
| 16 |
"d_model": 384,
|
| 17 |
-
"decoder_attention_heads":
|
| 18 |
"decoder_ffn_dim": 1536,
|
| 19 |
"decoder_layerdrop": 0.0,
|
| 20 |
-
"decoder_layers":
|
| 21 |
"decoder_start_token_id": 50258,
|
| 22 |
"dropout": 0.0,
|
| 23 |
-
"encoder_attention_heads":
|
| 24 |
"encoder_ffn_dim": 1536,
|
| 25 |
"encoder_layerdrop": 0.0,
|
| 26 |
-
"encoder_layers":
|
| 27 |
"eos_token_id": 50257,
|
| 28 |
-
"forced_decoder_ids":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
"init_std": 0.02,
|
| 30 |
"is_encoder_decoder": true,
|
| 31 |
"mask_feature_length": 10,
|
|
@@ -38,14 +51,14 @@
|
|
| 38 |
"max_source_positions": 1500,
|
| 39 |
"max_target_positions": 448,
|
| 40 |
"model_type": "whisper",
|
| 41 |
-
"num_hidden_layers":
|
| 42 |
"num_mel_bins": 80,
|
| 43 |
"pad_token_id": 50257,
|
| 44 |
"scale_embedding": false,
|
| 45 |
"suppress_tokens": [],
|
| 46 |
-
"torch_dtype": "
|
| 47 |
-
"transformers_version": "4.
|
| 48 |
-
"use_cache":
|
| 49 |
"use_weighted_layer_sum": false,
|
| 50 |
"vocab_size": 51865
|
| 51 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "gramvaani_whisper-tiny-hi/",
|
| 3 |
"activation_dropout": 0.0,
|
| 4 |
"activation_function": "gelu",
|
| 5 |
"apply_spec_augment": false,
|
| 6 |
"architectures": [
|
| 7 |
+
"PoptorchPipelinedWhisperForConditionalGeneration"
|
| 8 |
],
|
| 9 |
"attention_dropout": 0.0,
|
| 10 |
"begin_suppress_tokens": [
|
|
|
|
| 14 |
"bos_token_id": 50257,
|
| 15 |
"classifier_proj_size": 256,
|
| 16 |
"d_model": 384,
|
| 17 |
+
"decoder_attention_heads": 6,
|
| 18 |
"decoder_ffn_dim": 1536,
|
| 19 |
"decoder_layerdrop": 0.0,
|
| 20 |
+
"decoder_layers": 4,
|
| 21 |
"decoder_start_token_id": 50258,
|
| 22 |
"dropout": 0.0,
|
| 23 |
+
"encoder_attention_heads": 6,
|
| 24 |
"encoder_ffn_dim": 1536,
|
| 25 |
"encoder_layerdrop": 0.0,
|
| 26 |
+
"encoder_layers": 4,
|
| 27 |
"eos_token_id": 50257,
|
| 28 |
+
"forced_decoder_ids": [
|
| 29 |
+
[
|
| 30 |
+
1,
|
| 31 |
+
50306
|
| 32 |
+
],
|
| 33 |
+
[
|
| 34 |
+
2,
|
| 35 |
+
50359
|
| 36 |
+
],
|
| 37 |
+
[
|
| 38 |
+
3,
|
| 39 |
+
50363
|
| 40 |
+
]
|
| 41 |
+
],
|
| 42 |
"init_std": 0.02,
|
| 43 |
"is_encoder_decoder": true,
|
| 44 |
"mask_feature_length": 10,
|
|
|
|
| 51 |
"max_source_positions": 1500,
|
| 52 |
"max_target_positions": 448,
|
| 53 |
"model_type": "whisper",
|
| 54 |
+
"num_hidden_layers": 4,
|
| 55 |
"num_mel_bins": 80,
|
| 56 |
"pad_token_id": 50257,
|
| 57 |
"scale_embedding": false,
|
| 58 |
"suppress_tokens": [],
|
| 59 |
+
"torch_dtype": "float16",
|
| 60 |
+
"transformers_version": "4.29.2",
|
| 61 |
+
"use_cache": true,
|
| 62 |
"use_weighted_layer_sum": false,
|
| 63 |
"vocab_size": 51865
|
| 64 |
}
|
preprocessor_config.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"chunk_length": 30,
|
| 3 |
-
"feature_extractor_type": "
|
| 4 |
"feature_size": 80,
|
| 5 |
"hop_length": 160,
|
| 6 |
"n_fft": 400,
|
|
|
|
| 1 |
{
|
| 2 |
"chunk_length": 30,
|
| 3 |
+
"feature_extractor_type": "WhisperFeatureExtractorTorch",
|
| 4 |
"feature_size": 80,
|
| 5 |
"hop_length": 160,
|
| 6 |
"n_fft": 400,
|