Model save
Browse files- README.md +90 -0
- adapter.ara.safetensors +3 -0
README.md
ADDED
@@ -0,0 +1,90 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: cc-by-nc-4.0
|
3 |
+
base_model: facebook/mms-1b-all
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
metrics:
|
7 |
+
- wer
|
8 |
+
model-index:
|
9 |
+
- name: mms-MGB3
|
10 |
+
results: []
|
11 |
+
---
|
12 |
+
|
13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
14 |
+
should probably proofread and complete it, then remove this comment. -->
|
15 |
+
|
16 |
+
# mms-MGB3
|
17 |
+
|
18 |
+
This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the None dataset.
|
19 |
+
It achieves the following results on the evaluation set:
|
20 |
+
- Loss: 0.9382
|
21 |
+
- Wer: 0.6591
|
22 |
+
|
23 |
+
## Model description
|
24 |
+
|
25 |
+
More information needed
|
26 |
+
|
27 |
+
## Intended uses & limitations
|
28 |
+
|
29 |
+
More information needed
|
30 |
+
|
31 |
+
## Training and evaluation data
|
32 |
+
|
33 |
+
More information needed
|
34 |
+
|
35 |
+
## Training procedure
|
36 |
+
|
37 |
+
### Training hyperparameters
|
38 |
+
|
39 |
+
The following hyperparameters were used during training:
|
40 |
+
- learning_rate: 1e-05
|
41 |
+
- train_batch_size: 14
|
42 |
+
- eval_batch_size: 8
|
43 |
+
- seed: 42
|
44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
45 |
+
- lr_scheduler_type: constant_with_warmup
|
46 |
+
- lr_scheduler_warmup_steps: 50
|
47 |
+
- num_epochs: 4
|
48 |
+
|
49 |
+
### Training results
|
50 |
+
|
51 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
52 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
53 |
+
| 9.7535 | 0.13 | 250 | 8.6735 | 1.0023 |
|
54 |
+
| 3.2385 | 0.27 | 500 | 3.3341 | 1.0003 |
|
55 |
+
| 2.3512 | 0.4 | 750 | 2.0937 | 0.9027 |
|
56 |
+
| 1.4967 | 0.53 | 1000 | 1.3694 | 0.7637 |
|
57 |
+
| 1.3214 | 0.67 | 1250 | 1.2237 | 0.7347 |
|
58 |
+
| 1.2072 | 0.8 | 1500 | 1.1672 | 0.7176 |
|
59 |
+
| 1.1913 | 0.93 | 1750 | 1.1334 | 0.7108 |
|
60 |
+
| 1.1127 | 1.07 | 2000 | 1.1102 | 0.7044 |
|
61 |
+
| 1.1454 | 1.2 | 2250 | 1.0919 | 0.6996 |
|
62 |
+
| 1.1128 | 1.33 | 2500 | 1.0763 | 0.6955 |
|
63 |
+
| 1.086 | 1.47 | 2750 | 1.0629 | 0.6916 |
|
64 |
+
| 1.1285 | 1.6 | 3000 | 1.0503 | 0.6888 |
|
65 |
+
| 1.081 | 1.73 | 3250 | 1.0406 | 0.6886 |
|
66 |
+
| 1.0449 | 1.86 | 3500 | 1.0320 | 0.6857 |
|
67 |
+
| 1.0625 | 2.0 | 3750 | 1.0231 | 0.6849 |
|
68 |
+
| 1.0892 | 2.13 | 4000 | 1.0157 | 0.6824 |
|
69 |
+
| 1.0566 | 2.26 | 4250 | 1.0097 | 0.6795 |
|
70 |
+
| 1.0972 | 2.4 | 4500 | 1.0036 | 0.6747 |
|
71 |
+
| 1.0617 | 2.53 | 4750 | 0.9957 | 0.6744 |
|
72 |
+
| 1.0441 | 2.66 | 5000 | 0.9881 | 0.6756 |
|
73 |
+
| 1.0589 | 2.8 | 5250 | 0.9807 | 0.6718 |
|
74 |
+
| 1.0005 | 2.93 | 5500 | 0.9758 | 0.6713 |
|
75 |
+
| 1.0447 | 3.06 | 5750 | 0.9701 | 0.6694 |
|
76 |
+
| 0.9722 | 3.2 | 6000 | 0.9667 | 0.6664 |
|
77 |
+
| 0.9873 | 3.33 | 6250 | 0.9595 | 0.6675 |
|
78 |
+
| 0.9857 | 3.46 | 6500 | 0.9551 | 0.6633 |
|
79 |
+
| 0.9625 | 3.6 | 6750 | 0.9519 | 0.6633 |
|
80 |
+
| 0.9748 | 3.73 | 7000 | 0.9464 | 0.6607 |
|
81 |
+
| 0.9626 | 3.86 | 7250 | 0.9427 | 0.6617 |
|
82 |
+
| 1.0242 | 4.0 | 7500 | 0.9382 | 0.6591 |
|
83 |
+
|
84 |
+
|
85 |
+
### Framework versions
|
86 |
+
|
87 |
+
- Transformers 4.33.2
|
88 |
+
- Pytorch 2.0.1
|
89 |
+
- Datasets 2.19.1
|
90 |
+
- Tokenizers 0.13.3
|
adapter.ara.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0fdebaa10a3225f1b785d77ef34c9d979ba52f15b3dea58aa4d385b5968079fd
|
3 |
+
size 9121368
|