End of training
Browse files- README.md +94 -10
- adapter.fry.safetensors +1 -1
- pytorch_model.bin +1 -1
README.md
CHANGED
@@ -5,9 +5,24 @@ tags:
|
|
5 |
- generated_from_trainer
|
6 |
datasets:
|
7 |
- common_voice_13_0
|
|
|
|
|
8 |
model-index:
|
9 |
- name: wav2vec2-large-mms-1b-fries-NL_common_voice_13b_other-train-validation
|
10 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
---
|
12 |
|
13 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -17,13 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
17 |
|
18 |
This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the common_voice_13_0 dataset.
|
19 |
It achieves the following results on the evaluation set:
|
20 |
-
-
|
21 |
-
-
|
22 |
-
- eval_runtime: 232.3788
|
23 |
-
- eval_samples_per_second: 13.637
|
24 |
-
- eval_steps_per_second: 1.708
|
25 |
-
- epoch: 1.72
|
26 |
-
- step: 23500
|
27 |
|
28 |
## Model description
|
29 |
|
@@ -48,8 +58,82 @@ The following hyperparameters were used during training:
|
|
48 |
- seed: 42
|
49 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
50 |
- lr_scheduler_type: linear
|
51 |
-
- lr_scheduler_warmup_steps:
|
52 |
-
- num_epochs:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
53 |
|
54 |
### Framework versions
|
55 |
|
|
|
5 |
- generated_from_trainer
|
6 |
datasets:
|
7 |
- common_voice_13_0
|
8 |
+
metrics:
|
9 |
+
- wer
|
10 |
model-index:
|
11 |
- name: wav2vec2-large-mms-1b-fries-NL_common_voice_13b_other-train-validation
|
12 |
+
results:
|
13 |
+
- task:
|
14 |
+
name: Automatic Speech Recognition
|
15 |
+
type: automatic-speech-recognition
|
16 |
+
dataset:
|
17 |
+
name: common_voice_13_0
|
18 |
+
type: common_voice_13_0
|
19 |
+
config: fy-NL
|
20 |
+
split: test
|
21 |
+
args: fy-NL
|
22 |
+
metrics:
|
23 |
+
- name: Wer
|
24 |
+
type: wer
|
25 |
+
value: 0.19910413556026252
|
26 |
---
|
27 |
|
28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
32 |
|
33 |
This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the common_voice_13_0 dataset.
|
34 |
It achieves the following results on the evaluation set:
|
35 |
+
- Loss: 0.1617
|
36 |
+
- Wer: 0.1991
|
|
|
|
|
|
|
|
|
|
|
37 |
|
38 |
## Model description
|
39 |
|
|
|
58 |
- seed: 42
|
59 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
60 |
- lr_scheduler_type: linear
|
61 |
+
- lr_scheduler_warmup_steps: 200
|
62 |
+
- num_epochs: 1
|
63 |
+
|
64 |
+
### Training results
|
65 |
+
|
66 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
67 |
+
|:-------------:|:-----:|:-----:|:---------------:|:------:|
|
68 |
+
| No log | 0.01 | 200 | 0.2768 | 0.2969 |
|
69 |
+
| 2.0359 | 0.03 | 400 | 0.2503 | 0.2754 |
|
70 |
+
| 2.0359 | 0.04 | 600 | 0.2378 | 0.2671 |
|
71 |
+
| 0.4005 | 0.06 | 800 | 0.2259 | 0.2572 |
|
72 |
+
| 0.4005 | 0.07 | 1000 | 0.2387 | 0.2733 |
|
73 |
+
| 0.4051 | 0.09 | 1200 | 0.2382 | 0.2645 |
|
74 |
+
| 0.4051 | 0.1 | 1400 | 0.2231 | 0.2513 |
|
75 |
+
| 0.3982 | 0.12 | 1600 | 0.2146 | 0.2470 |
|
76 |
+
| 0.3982 | 0.13 | 1800 | 0.2167 | 0.2503 |
|
77 |
+
| 0.3646 | 0.15 | 2000 | 0.2177 | 0.2499 |
|
78 |
+
| 0.3646 | 0.16 | 2200 | 0.2228 | 0.2591 |
|
79 |
+
| 0.3538 | 0.18 | 2400 | 0.2117 | 0.2445 |
|
80 |
+
| 0.3538 | 0.19 | 2600 | 0.2097 | 0.2411 |
|
81 |
+
| 0.3687 | 0.21 | 2800 | 0.2073 | 0.2425 |
|
82 |
+
| 0.3687 | 0.22 | 3000 | 0.2138 | 0.2454 |
|
83 |
+
| 0.3586 | 0.23 | 3200 | 0.2040 | 0.2375 |
|
84 |
+
| 0.3586 | 0.25 | 3400 | 0.2059 | 0.2372 |
|
85 |
+
| 0.3453 | 0.26 | 3600 | 0.2060 | 0.2397 |
|
86 |
+
| 0.3453 | 0.28 | 3800 | 0.2015 | 0.2382 |
|
87 |
+
| 0.3741 | 0.29 | 4000 | 0.2088 | 0.2457 |
|
88 |
+
| 0.3741 | 0.31 | 4200 | 0.1948 | 0.2298 |
|
89 |
+
| 0.3454 | 0.32 | 4400 | 0.2014 | 0.2342 |
|
90 |
+
| 0.3454 | 0.34 | 4600 | 0.2031 | 0.2392 |
|
91 |
+
| 0.351 | 0.35 | 4800 | 0.2018 | 0.2401 |
|
92 |
+
| 0.351 | 0.37 | 5000 | 0.1962 | 0.2321 |
|
93 |
+
| 0.3502 | 0.38 | 5200 | 0.1945 | 0.2323 |
|
94 |
+
| 0.3502 | 0.4 | 5400 | 0.1956 | 0.2323 |
|
95 |
+
| 0.3423 | 0.41 | 5600 | 0.1913 | 0.2266 |
|
96 |
+
| 0.3423 | 0.43 | 5800 | 0.1921 | 0.2277 |
|
97 |
+
| 0.3414 | 0.44 | 6000 | 0.1910 | 0.2262 |
|
98 |
+
| 0.3414 | 0.45 | 6200 | 0.1891 | 0.2223 |
|
99 |
+
| 0.3517 | 0.47 | 6400 | 0.1862 | 0.2230 |
|
100 |
+
| 0.3517 | 0.48 | 6600 | 0.1879 | 0.2206 |
|
101 |
+
| 0.3273 | 0.5 | 6800 | 0.1849 | 0.2176 |
|
102 |
+
| 0.3273 | 0.51 | 7000 | 0.1845 | 0.2163 |
|
103 |
+
| 0.321 | 0.53 | 7200 | 0.1831 | 0.2163 |
|
104 |
+
| 0.321 | 0.54 | 7400 | 0.1825 | 0.2163 |
|
105 |
+
| 0.321 | 0.56 | 7600 | 0.1797 | 0.2155 |
|
106 |
+
| 0.321 | 0.57 | 7800 | 0.1787 | 0.2144 |
|
107 |
+
| 0.3382 | 0.59 | 8000 | 0.1804 | 0.2132 |
|
108 |
+
| 0.3382 | 0.6 | 8200 | 0.1789 | 0.2158 |
|
109 |
+
| 0.3285 | 0.62 | 8400 | 0.1778 | 0.2130 |
|
110 |
+
| 0.3285 | 0.63 | 8600 | 0.1753 | 0.2094 |
|
111 |
+
| 0.3103 | 0.65 | 8800 | 0.1786 | 0.2147 |
|
112 |
+
| 0.3103 | 0.66 | 9000 | 0.1799 | 0.2157 |
|
113 |
+
| 0.3184 | 0.67 | 9200 | 0.1747 | 0.2098 |
|
114 |
+
| 0.3184 | 0.69 | 9400 | 0.1740 | 0.2068 |
|
115 |
+
| 0.3037 | 0.7 | 9600 | 0.1728 | 0.2090 |
|
116 |
+
| 0.3037 | 0.72 | 9800 | 0.1732 | 0.2084 |
|
117 |
+
| 0.3145 | 0.73 | 10000 | 0.1725 | 0.2085 |
|
118 |
+
| 0.3145 | 0.75 | 10200 | 0.1691 | 0.2052 |
|
119 |
+
| 0.3063 | 0.76 | 10400 | 0.1699 | 0.2062 |
|
120 |
+
| 0.3063 | 0.78 | 10600 | 0.1694 | 0.2072 |
|
121 |
+
| 0.3104 | 0.79 | 10800 | 0.1692 | 0.2063 |
|
122 |
+
| 0.3104 | 0.81 | 11000 | 0.1674 | 0.2044 |
|
123 |
+
| 0.2991 | 0.82 | 11200 | 0.1677 | 0.2040 |
|
124 |
+
| 0.2991 | 0.84 | 11400 | 0.1664 | 0.2025 |
|
125 |
+
| 0.3146 | 0.85 | 11600 | 0.1666 | 0.2011 |
|
126 |
+
| 0.3146 | 0.87 | 11800 | 0.1666 | 0.2020 |
|
127 |
+
| 0.3162 | 0.88 | 12000 | 0.1647 | 0.2009 |
|
128 |
+
| 0.3162 | 0.89 | 12200 | 0.1642 | 0.2014 |
|
129 |
+
| 0.3156 | 0.91 | 12400 | 0.1634 | 0.1997 |
|
130 |
+
| 0.3156 | 0.92 | 12600 | 0.1630 | 0.1994 |
|
131 |
+
| 0.3075 | 0.94 | 12800 | 0.1625 | 0.2009 |
|
132 |
+
| 0.3075 | 0.95 | 13000 | 0.1621 | 0.1994 |
|
133 |
+
| 0.3121 | 0.97 | 13200 | 0.1619 | 0.1989 |
|
134 |
+
| 0.3121 | 0.98 | 13400 | 0.1619 | 0.1989 |
|
135 |
+
| 0.2909 | 1.0 | 13600 | 0.1617 | 0.1991 |
|
136 |
+
|
137 |
|
138 |
### Framework versions
|
139 |
|
adapter.fry.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 8865152
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:aa7fb35fb5b7e61e20e951f50b8f1cd25cd2c73d54f134e2047be9feb3eb553d
|
3 |
size 8865152
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 3859201165
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9a95c6757610d4da161f4affd7b6ab19fcb1a7238c2db71899696b98c7acbdb3
|
3 |
size 3859201165
|