End of training
Browse files
README.md
ADDED
@@ -0,0 +1,128 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: Akashpb13/Swahili_xlsr
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
datasets:
|
7 |
+
- ml-superb-subset
|
8 |
+
metrics:
|
9 |
+
- wer
|
10 |
+
model-index:
|
11 |
+
- name: ssw_finetune
|
12 |
+
results:
|
13 |
+
- task:
|
14 |
+
name: Automatic Speech Recognition
|
15 |
+
type: automatic-speech-recognition
|
16 |
+
dataset:
|
17 |
+
name: ml-superb-subset
|
18 |
+
type: ml-superb-subset
|
19 |
+
config: ssw
|
20 |
+
split: test
|
21 |
+
args: ssw
|
22 |
+
metrics:
|
23 |
+
- name: Wer
|
24 |
+
type: wer
|
25 |
+
value: 42.14876033057851
|
26 |
+
---
|
27 |
+
|
28 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
29 |
+
should probably proofread and complete it, then remove this comment. -->
|
30 |
+
|
31 |
+
# ssw_finetune
|
32 |
+
|
33 |
+
This model is a fine-tuned version of [Akashpb13/Swahili_xlsr](https://huggingface.co/Akashpb13/Swahili_xlsr) on the ml-superb-subset dataset.
|
34 |
+
It achieves the following results on the evaluation set:
|
35 |
+
- Loss: 0.4301
|
36 |
+
- Wer: 42.1488
|
37 |
+
|
38 |
+
## Model description
|
39 |
+
|
40 |
+
More information needed
|
41 |
+
|
42 |
+
## Intended uses & limitations
|
43 |
+
|
44 |
+
More information needed
|
45 |
+
|
46 |
+
## Training and evaluation data
|
47 |
+
|
48 |
+
More information needed
|
49 |
+
|
50 |
+
## Training procedure
|
51 |
+
|
52 |
+
### Training hyperparameters
|
53 |
+
|
54 |
+
The following hyperparameters were used during training:
|
55 |
+
- learning_rate: 9.6e-05
|
56 |
+
- train_batch_size: 32
|
57 |
+
- eval_batch_size: 8
|
58 |
+
- seed: 42
|
59 |
+
- gradient_accumulation_steps: 2
|
60 |
+
- total_train_batch_size: 64
|
61 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
62 |
+
- lr_scheduler_type: cosine
|
63 |
+
- lr_scheduler_warmup_steps: 25
|
64 |
+
- training_steps: 500
|
65 |
+
- mixed_precision_training: Native AMP
|
66 |
+
|
67 |
+
### Training results
|
68 |
+
|
69 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
70 |
+
|:-------------:|:-------:|:----:|:---------------:|:--------:|
|
71 |
+
| 22.1208 | 0.8333 | 10 | 25.1031 | 100.5510 |
|
72 |
+
| 12.838 | 1.6667 | 20 | 10.4898 | 100.0 |
|
73 |
+
| 4.2236 | 2.5 | 30 | 3.9356 | 100.0 |
|
74 |
+
| 3.4491 | 3.3333 | 40 | 3.4590 | 100.0 |
|
75 |
+
| 3.2593 | 4.1667 | 50 | 3.3211 | 100.0 |
|
76 |
+
| 3.1611 | 5.0 | 60 | 3.1737 | 100.0 |
|
77 |
+
| 3.1157 | 5.8333 | 70 | 3.1089 | 100.0 |
|
78 |
+
| 3.0472 | 6.6667 | 80 | 3.0868 | 100.0 |
|
79 |
+
| 3.0291 | 7.5 | 90 | 3.0445 | 100.0 |
|
80 |
+
| 2.9996 | 8.3333 | 100 | 3.0058 | 100.0 |
|
81 |
+
| 2.9187 | 9.1667 | 110 | 2.9600 | 100.0 |
|
82 |
+
| 2.7708 | 10.0 | 120 | 2.7274 | 100.0 |
|
83 |
+
| 2.5396 | 10.8333 | 130 | 2.4602 | 100.0 |
|
84 |
+
| 2.0911 | 11.6667 | 140 | 1.8863 | 100.0 |
|
85 |
+
| 1.4477 | 12.5 | 150 | 1.2924 | 95.8678 |
|
86 |
+
| 1.042 | 13.3333 | 160 | 0.9620 | 80.1653 |
|
87 |
+
| 0.8089 | 14.1667 | 170 | 0.7520 | 67.4931 |
|
88 |
+
| 0.6621 | 15.0 | 180 | 0.6530 | 53.7190 |
|
89 |
+
| 0.5476 | 15.8333 | 190 | 0.5838 | 50.6887 |
|
90 |
+
| 0.4866 | 16.6667 | 200 | 0.5662 | 50.4132 |
|
91 |
+
| 0.4296 | 17.5 | 210 | 0.5303 | 49.5868 |
|
92 |
+
| 0.3977 | 18.3333 | 220 | 0.5121 | 47.9339 |
|
93 |
+
| 0.392 | 19.1667 | 230 | 0.4895 | 47.3829 |
|
94 |
+
| 0.346 | 20.0 | 240 | 0.4825 | 44.3526 |
|
95 |
+
| 0.3226 | 20.8333 | 250 | 0.4628 | 45.1791 |
|
96 |
+
| 0.3145 | 21.6667 | 260 | 0.4662 | 45.1791 |
|
97 |
+
| 0.2948 | 22.5 | 270 | 0.4492 | 41.8733 |
|
98 |
+
| 0.2857 | 23.3333 | 280 | 0.4484 | 43.2507 |
|
99 |
+
| 0.2571 | 24.1667 | 290 | 0.4511 | 43.2507 |
|
100 |
+
| 0.2706 | 25.0 | 300 | 0.4382 | 41.8733 |
|
101 |
+
| 0.2404 | 25.8333 | 310 | 0.4528 | 42.1488 |
|
102 |
+
| 0.2498 | 26.6667 | 320 | 0.4428 | 41.5978 |
|
103 |
+
| 0.2381 | 27.5 | 330 | 0.4377 | 40.2204 |
|
104 |
+
| 0.2142 | 28.3333 | 340 | 0.4300 | 41.0468 |
|
105 |
+
| 0.2236 | 29.1667 | 350 | 0.4305 | 42.1488 |
|
106 |
+
| 0.2249 | 30.0 | 360 | 0.4253 | 41.0468 |
|
107 |
+
| 0.209 | 30.8333 | 370 | 0.4272 | 42.9752 |
|
108 |
+
| 0.2071 | 31.6667 | 380 | 0.4363 | 43.8017 |
|
109 |
+
| 0.2209 | 32.5 | 390 | 0.4328 | 44.6281 |
|
110 |
+
| 0.2012 | 33.3333 | 400 | 0.4351 | 44.0771 |
|
111 |
+
| 0.1895 | 34.1667 | 410 | 0.4362 | 43.8017 |
|
112 |
+
| 0.1921 | 35.0 | 420 | 0.4383 | 45.1791 |
|
113 |
+
| 0.1805 | 35.8333 | 430 | 0.4381 | 45.1791 |
|
114 |
+
| 0.1963 | 36.6667 | 440 | 0.4331 | 41.3223 |
|
115 |
+
| 0.1829 | 37.5 | 450 | 0.4301 | 41.5978 |
|
116 |
+
| 0.1927 | 38.3333 | 460 | 0.4290 | 41.8733 |
|
117 |
+
| 0.1779 | 39.1667 | 470 | 0.4289 | 42.4242 |
|
118 |
+
| 0.1892 | 40.0 | 480 | 0.4302 | 42.1488 |
|
119 |
+
| 0.2025 | 40.8333 | 490 | 0.4300 | 42.4242 |
|
120 |
+
| 0.2105 | 41.6667 | 500 | 0.4301 | 42.1488 |
|
121 |
+
|
122 |
+
|
123 |
+
### Framework versions
|
124 |
+
|
125 |
+
- Transformers 4.41.1
|
126 |
+
- Pytorch 2.3.0+cu121
|
127 |
+
- Datasets 2.19.1
|
128 |
+
- Tokenizers 0.19.1
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1261942780
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:df6cce419942ebc59b7ef974e32b510b08548dfb7f99ffacf04bdd825990ead4
|
3 |
size 1261942780
|
runs/May22_21-05-55_46a7ce9c3c33/events.out.tfevents.1716412017.46a7ce9c3c33.1173.0
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0d807f7bc0006eae31386d0465abf3e64f6363f59053fbd2da950dfcf418335e
|
3 |
+
size 43384
|