End of training
Browse files- README.md +117 -0
- adapter_config.json +32 -0
- adapter_model.safetensors +3 -0
- preprocessor_config.json +14 -0
- runs/Oct20_10-45-40_cb4d8a5b1987/events.out.tfevents.1729421160.cb4d8a5b1987.735.0 +3 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,117 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
language:
|
4 |
+
- ms
|
5 |
+
license: apache-2.0
|
6 |
+
base_model: openai/whisper-large-v3
|
7 |
+
tags:
|
8 |
+
- generated_from_trainer
|
9 |
+
datasets:
|
10 |
+
- clt013/malay-speech-3k-rows-dataset_v2
|
11 |
+
model-index:
|
12 |
+
- name: Whisper Large v3 FT Malay - CLT013
|
13 |
+
results: []
|
14 |
+
---
|
15 |
+
|
16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
17 |
+
should probably proofread and complete it, then remove this comment. -->
|
18 |
+
|
19 |
+
# Whisper Large v3 FT Malay - CLT013
|
20 |
+
|
21 |
+
This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on the Malay Speech 3k dataset.
|
22 |
+
It achieves the following results on the evaluation set:
|
23 |
+
- Loss: 0.4149
|
24 |
+
|
25 |
+
## Model description
|
26 |
+
|
27 |
+
More information needed
|
28 |
+
|
29 |
+
## Intended uses & limitations
|
30 |
+
|
31 |
+
More information needed
|
32 |
+
|
33 |
+
## Training and evaluation data
|
34 |
+
|
35 |
+
More information needed
|
36 |
+
|
37 |
+
## Training procedure
|
38 |
+
|
39 |
+
### Training hyperparameters
|
40 |
+
|
41 |
+
The following hyperparameters were used during training:
|
42 |
+
- learning_rate: 1e-05
|
43 |
+
- train_batch_size: 8
|
44 |
+
- eval_batch_size: 8
|
45 |
+
- seed: 42
|
46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
47 |
+
- lr_scheduler_type: linear
|
48 |
+
- lr_scheduler_warmup_steps: 100
|
49 |
+
- num_epochs: 20
|
50 |
+
- mixed_precision_training: Native AMP
|
51 |
+
|
52 |
+
### Training results
|
53 |
+
|
54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
55 |
+
|:-------------:|:-------:|:----:|:---------------:|
|
56 |
+
| 2.6257 | 0.3731 | 100 | 2.4258 |
|
57 |
+
| 2.3394 | 0.7463 | 200 | 1.7629 |
|
58 |
+
| 1.5256 | 1.1194 | 300 | 1.1983 |
|
59 |
+
| 1.0648 | 1.4925 | 400 | 0.9111 |
|
60 |
+
| 0.7458 | 1.8657 | 500 | 0.5754 |
|
61 |
+
| 0.5287 | 2.2388 | 600 | 0.5316 |
|
62 |
+
| 1.0221 | 2.6119 | 700 | 1.6308 |
|
63 |
+
| 0.9045 | 2.9851 | 800 | 0.4655 |
|
64 |
+
| 0.3766 | 3.3582 | 900 | 0.4186 |
|
65 |
+
| 0.3767 | 3.7313 | 1000 | 0.4082 |
|
66 |
+
| 0.355 | 4.1045 | 1100 | 0.4022 |
|
67 |
+
| 0.3284 | 4.4776 | 1200 | 0.3996 |
|
68 |
+
| 0.3229 | 4.8507 | 1300 | 0.3928 |
|
69 |
+
| 0.318 | 5.2239 | 1400 | 0.3925 |
|
70 |
+
| 0.3083 | 5.5970 | 1500 | 0.3905 |
|
71 |
+
| 0.3035 | 5.9701 | 1600 | 0.3900 |
|
72 |
+
| 0.2792 | 6.3433 | 1700 | 0.3897 |
|
73 |
+
| 0.2948 | 6.7164 | 1800 | 0.3872 |
|
74 |
+
| 0.2537 | 7.0896 | 1900 | 0.3869 |
|
75 |
+
| 0.2703 | 7.4627 | 2000 | 0.3907 |
|
76 |
+
| 0.2696 | 7.8358 | 2100 | 0.3891 |
|
77 |
+
| 0.2672 | 8.2090 | 2200 | 0.3923 |
|
78 |
+
| 0.2645 | 8.5821 | 2300 | 0.3900 |
|
79 |
+
| 0.2419 | 8.9552 | 2400 | 0.3894 |
|
80 |
+
| 0.2361 | 9.3284 | 2500 | 0.3928 |
|
81 |
+
| 0.2361 | 9.7015 | 2600 | 0.3915 |
|
82 |
+
| 0.2229 | 10.0746 | 2700 | 0.3922 |
|
83 |
+
| 0.2151 | 10.4478 | 2800 | 0.3962 |
|
84 |
+
| 0.248 | 10.8209 | 2900 | 0.3945 |
|
85 |
+
| 0.2152 | 11.1940 | 3000 | 0.3981 |
|
86 |
+
| 0.2144 | 11.5672 | 3100 | 0.4012 |
|
87 |
+
| 0.2278 | 11.9403 | 3200 | 0.3970 |
|
88 |
+
| 0.2017 | 12.3134 | 3300 | 0.4002 |
|
89 |
+
| 0.216 | 12.6866 | 3400 | 0.4018 |
|
90 |
+
| 0.2166 | 13.0597 | 3500 | 0.4013 |
|
91 |
+
| 0.2176 | 13.4328 | 3600 | 0.4043 |
|
92 |
+
| 0.2002 | 13.8060 | 3700 | 0.4054 |
|
93 |
+
| 0.1752 | 14.1791 | 3800 | 0.4066 |
|
94 |
+
| 0.1892 | 14.5522 | 3900 | 0.4048 |
|
95 |
+
| 0.2121 | 14.9254 | 4000 | 0.4078 |
|
96 |
+
| 0.209 | 15.2985 | 4100 | 0.4069 |
|
97 |
+
| 0.1857 | 15.6716 | 4200 | 0.4079 |
|
98 |
+
| 0.1867 | 16.0448 | 4300 | 0.4097 |
|
99 |
+
| 0.1833 | 16.4179 | 4400 | 0.4093 |
|
100 |
+
| 0.1918 | 16.7910 | 4500 | 0.4130 |
|
101 |
+
| 0.1815 | 17.1642 | 4600 | 0.4123 |
|
102 |
+
| 0.1854 | 17.5373 | 4700 | 0.4135 |
|
103 |
+
| 0.181 | 17.9104 | 4800 | 0.4131 |
|
104 |
+
| 0.1756 | 18.2836 | 4900 | 0.4130 |
|
105 |
+
| 0.177 | 18.6567 | 5000 | 0.4143 |
|
106 |
+
| 0.1829 | 19.0299 | 5100 | 0.4137 |
|
107 |
+
| 0.1809 | 19.4030 | 5200 | 0.4163 |
|
108 |
+
| 0.1777 | 19.7761 | 5300 | 0.4149 |
|
109 |
+
|
110 |
+
|
111 |
+
### Framework versions
|
112 |
+
|
113 |
+
- PEFT 0.13.2
|
114 |
+
- Transformers 4.44.2
|
115 |
+
- Pytorch 2.4.1+cu121
|
116 |
+
- Datasets 3.0.1
|
117 |
+
- Tokenizers 0.19.1
|
adapter_config.json
ADDED
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": {
|
4 |
+
"base_model_class": "WhisperForConditionalGeneration",
|
5 |
+
"parent_library": "transformers.models.whisper.modeling_whisper"
|
6 |
+
},
|
7 |
+
"base_model_name_or_path": "openai/whisper-large-v3",
|
8 |
+
"bias": "none",
|
9 |
+
"fan_in_fan_out": false,
|
10 |
+
"inference_mode": true,
|
11 |
+
"init_lora_weights": true,
|
12 |
+
"layer_replication": null,
|
13 |
+
"layers_pattern": null,
|
14 |
+
"layers_to_transform": null,
|
15 |
+
"loftq_config": {},
|
16 |
+
"lora_alpha": 64,
|
17 |
+
"lora_dropout": 0.05,
|
18 |
+
"megatron_config": null,
|
19 |
+
"megatron_core": "megatron.core",
|
20 |
+
"modules_to_save": null,
|
21 |
+
"peft_type": "LORA",
|
22 |
+
"r": 32,
|
23 |
+
"rank_pattern": {},
|
24 |
+
"revision": null,
|
25 |
+
"target_modules": [
|
26 |
+
"v_proj",
|
27 |
+
"q_proj"
|
28 |
+
],
|
29 |
+
"task_type": null,
|
30 |
+
"use_dora": false,
|
31 |
+
"use_rslora": false
|
32 |
+
}
|
adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a6c389622c1c0b6b8af298050d85522a4a289ed1cec0a7f97448b15177abc72a
|
3 |
+
size 62969640
|
preprocessor_config.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"chunk_length": 30,
|
3 |
+
"feature_extractor_type": "WhisperFeatureExtractor",
|
4 |
+
"feature_size": 128,
|
5 |
+
"hop_length": 160,
|
6 |
+
"n_fft": 400,
|
7 |
+
"n_samples": 480000,
|
8 |
+
"nb_max_frames": 3000,
|
9 |
+
"padding_side": "right",
|
10 |
+
"padding_value": 0.0,
|
11 |
+
"processor_class": "WhisperProcessor",
|
12 |
+
"return_attention_mask": false,
|
13 |
+
"sampling_rate": 16000
|
14 |
+
}
|
runs/Oct20_10-45-40_cb4d8a5b1987/events.out.tfevents.1729421160.cb4d8a5b1987.735.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:025259177d2a75be4c986526a3d3be5317efc8e55b002625c68b118d87336d26
|
3 |
+
size 32142
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7fda5afae55f682e1f272b26919b036ea6ea345baf0f168f20f8168e314f784d
|
3 |
+
size 5368
|