Hugging Face's logo

JunWorks
/
whisper-small-LIDmetric-4060-2lang-properLORAsettings

PEFT
TensorBoard
Safetensors
multilingual
hf-asr-leaderboard
Generated from Trainer
Model card Files
xet
Metrics Community
JunWorks commited on
Commit
bc924c4
·
verified ·
1 Parent(s): 13a5273

End of training

Browse files
Files changed (3) hide show
  1. README.md +147 -0
  2. adapter_model.safetensors +1 -1
  3. runs/Jul23_18-08-18_Edge-Ai/events.out.tfevents.1721729301.Edge-Ai +2 -2
README.md ADDED
@@ -0,0 +1,147 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: openai/whisper-small
3
+ datasets:
4
+ - mozilla-foundation/common_voice_13_0
5
+ language:
6
+ - multilingual
7
+ library_name: peft
8
+ license: apache-2.0
9
+ tags:
10
+ - hf-asr-leaderboard
11
+ - generated_from_trainer
12
+ model-index:
13
+ - name: Whisper Small 2 lang LORA 2nd Settings
14
+ results: []
15
+ ---
16
+
17
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
18
+ should probably proofread and complete it, then remove this comment. -->
19
+
20
+ # Whisper Small 2 lang LORA 2nd Settings
21
+
22
+ This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 13.0 dataset.
23
+ It achieves the following results on the evaluation set:
24
+ - Loss: 0.2569
25
+
26
+ ## Model description
27
+
28
+ More information needed
29
+
30
+ ## Intended uses & limitations
31
+
32
+ More information needed
33
+
34
+ ## Training and evaluation data
35
+
36
+ More information needed
37
+
38
+ ## Training procedure
39
+
40
+ ### Training hyperparameters
41
+
42
+ The following hyperparameters were used during training:
43
+ - learning_rate: 1e-05
44
+ - train_batch_size: 4
45
+ - eval_batch_size: 1
46
+ - seed: 42
47
+ - gradient_accumulation_steps: 2
48
+ - total_train_batch_size: 8
49
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
+ - lr_scheduler_type: linear
51
+ - lr_scheduler_warmup_steps: 500
52
+ - training_steps: 20000
53
+ - mixed_precision_training: Native AMP
54
+
55
+ ### Training results
56
+
57
+ | Training Loss | Epoch | Step | Validation Loss |
58
+ |:-------------:|:------:|:-----:|:---------------:|
59
+ | 1.9564 | 0.1002 | 250 | 2.1385 |
60
+ | 1.2046 | 0.2004 | 500 | 1.1713 |
61
+ | 0.8719 | 0.3006 | 750 | 0.9414 |
62
+ | 0.7799 | 0.4008 | 1000 | 0.8261 |
63
+ | 0.6956 | 0.5010 | 1250 | 0.6536 |
64
+ | 0.3536 | 0.6012 | 1500 | 0.2990 |
65
+ | 0.3167 | 0.7014 | 1750 | 0.2902 |
66
+ | 0.3252 | 0.8016 | 2000 | 0.2850 |
67
+ | 0.3133 | 0.9018 | 2250 | 0.2821 |
68
+ | 0.3764 | 1.0020 | 2500 | 0.2800 |
69
+ | 0.2942 | 1.1022 | 2750 | 0.2775 |
70
+ | 0.2937 | 1.2024 | 3000 | 0.2770 |
71
+ | 0.3478 | 1.3026 | 3250 | 0.2745 |
72
+ | 0.3776 | 1.4028 | 3500 | 0.2729 |
73
+ | 0.3099 | 1.5030 | 3750 | 0.2713 |
74
+ | 0.3087 | 1.6032 | 4000 | 0.2705 |
75
+ | 0.2998 | 1.7034 | 4250 | 0.2699 |
76
+ | 0.3226 | 1.8036 | 4500 | 0.2683 |
77
+ | 0.3589 | 1.9038 | 4750 | 0.2676 |
78
+ | 0.3411 | 2.0040 | 5000 | 0.2673 |
79
+ | 0.3084 | 2.1042 | 5250 | 0.2674 |
80
+ | 0.31 | 2.2044 | 5500 | 0.2663 |
81
+ | 0.3388 | 2.3046 | 5750 | 0.2657 |
82
+ | 0.2716 | 2.4048 | 6000 | 0.2652 |
83
+ | 0.3059 | 2.5050 | 6250 | 0.2652 |
84
+ | 0.27 | 2.6052 | 6500 | 0.2648 |
85
+ | 0.2954 | 2.7054 | 6750 | 0.2639 |
86
+ | 0.336 | 2.8056 | 7000 | 0.2641 |
87
+ | 0.2833 | 2.9058 | 7250 | 0.2631 |
88
+ | 0.2777 | 3.0060 | 7500 | 0.2624 |
89
+ | 0.2418 | 3.1062 | 7750 | 0.2618 |
90
+ | 0.3194 | 3.2064 | 8000 | 0.2623 |
91
+ | 0.3319 | 3.3066 | 8250 | 0.2623 |
92
+ | 0.3551 | 3.4068 | 8500 | 0.2615 |
93
+ | 0.3421 | 3.5070 | 8750 | 0.2619 |
94
+ | 0.3862 | 3.6072 | 9000 | 0.2616 |
95
+ | 0.2437 | 3.7074 | 9250 | 0.2609 |
96
+ | 0.2995 | 3.8076 | 9500 | 0.2604 |
97
+ | 0.3535 | 3.9078 | 9750 | 0.2603 |
98
+ | 0.2871 | 4.0080 | 10000 | 0.2601 |
99
+ | 0.2908 | 4.1082 | 10250 | 0.2604 |
100
+ | 0.3203 | 4.2084 | 10500 | 0.2599 |
101
+ | 0.2598 | 4.3086 | 10750 | 0.2594 |
102
+ | 0.2942 | 4.4088 | 11000 | 0.2593 |
103
+ | 0.3302 | 4.5090 | 11250 | 0.2590 |
104
+ | 0.3615 | 4.6092 | 11500 | 0.2584 |
105
+ | 0.3291 | 4.7094 | 11750 | 0.2582 |
106
+ | 0.2781 | 4.8096 | 12000 | 0.2588 |
107
+ | 0.3106 | 4.9098 | 12250 | 0.2585 |
108
+ | 0.2484 | 5.0100 | 12500 | 0.2583 |
109
+ | 0.2645 | 5.1102 | 12750 | 0.2583 |
110
+ | 0.3034 | 5.2104 | 13000 | 0.2581 |
111
+ | 0.2865 | 5.3106 | 13250 | 0.2576 |
112
+ | 0.3301 | 5.4108 | 13500 | 0.2580 |
113
+ | 0.3759 | 5.5110 | 13750 | 0.2579 |
114
+ | 0.3318 | 5.6112 | 14000 | 0.2581 |
115
+ | 0.2825 | 5.7114 | 14250 | 0.2579 |
116
+ | 0.2976 | 5.8116 | 14500 | 0.2578 |
117
+ | 0.2976 | 5.9118 | 14750 | 0.2577 |
118
+ | 0.3681 | 6.0120 | 15000 | 0.2575 |
119
+ | 0.3274 | 6.1122 | 15250 | 0.2575 |
120
+ | 0.2948 | 6.2124 | 15500 | 0.2577 |
121
+ | 0.2932 | 6.3126 | 15750 | 0.2576 |
122
+ | 0.2587 | 6.4128 | 16000 | 0.2578 |
123
+ | 0.2564 | 6.5130 | 16250 | 0.2573 |
124
+ | 0.2776 | 6.6132 | 16500 | 0.2569 |
125
+ | 0.2954 | 6.7134 | 16750 | 0.2569 |
126
+ | 0.2891 | 6.8136 | 17000 | 0.2568 |
127
+ | 0.2373 | 6.9138 | 17250 | 0.2569 |
128
+ | 0.3532 | 7.0140 | 17500 | 0.2569 |
129
+ | 0.2676 | 7.1142 | 17750 | 0.2569 |
130
+ | 0.2763 | 7.2144 | 18000 | 0.2569 |
131
+ | 0.2692 | 7.3146 | 18250 | 0.2571 |
132
+ | 0.3198 | 7.4148 | 18500 | 0.2570 |
133
+ | 0.2158 | 7.5150 | 18750 | 0.2571 |
134
+ | 0.277 | 7.6152 | 19000 | 0.2572 |
135
+ | 0.2308 | 7.7154 | 19250 | 0.2572 |
136
+ | 0.3166 | 7.8156 | 19500 | 0.2569 |
137
+ | 0.3064 | 7.9158 | 19750 | 0.2570 |
138
+ | 0.2743 | 8.0160 | 20000 | 0.2569 |
139
+
140
+
141
+ ### Framework versions
142
+
143
+ - PEFT 0.11.2.dev0
144
+ - Transformers 4.43.0.dev0
145
+ - Pytorch 2.3.0+cu121
146
+ - Datasets 2.20.0
147
+ - Tokenizers 0.19.1
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f3fe5c1683dc5f51772589ee91d56f369cc43b375034dffb66659efe50cce090
3
  size 14176064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d81468fd56fc6d0aa4dbbe7c70751a6e95716060d6f7728b5a5a807597a736f2
3
  size 14176064
runs/Jul23_18-08-18_Edge-Ai/events.out.tfevents.1721729301.Edge-Ai CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9f6bdb4b559f1d59a48048836cb52b2abd688aa32dcb73c3222669ab3804266e
3
- size 429520
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1965eefb90d1b454d8098f5f91d3d1101a5161629119de40e1f6af2af1c1c5c6
3
+ size 452760