Hugging Face's logo

tgrhn
/
seamless_turkish

Safetensors
seamless_m4t_v2
Generated from Trainer
Model card Files Community
tgrhn commited on
Commit
dda387d
·
verified ·
1 Parent(s): 886a153

End of training

Browse files
Files changed (3) hide show
  1. README.md +143 -9
  2. model-00001-of-00002.safetensors +1 -1
  3. model-00002-of-00002.safetensors +1 -1
README.md CHANGED
@@ -17,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  This model is a fine-tuned version of [facebook/seamless-m4t-v2-large](https://huggingface.co/facebook/seamless-m4t-v2-large) on the common_voice_17_0 dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 0.8235
21
 
22
  ## Model description
23
 
@@ -43,18 +43,152 @@ The following hyperparameters were used during training:
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: linear
45
  - lr_scheduler_warmup_steps: 500
46
- - training_steps: 5000
47
  - mixed_precision_training: Native AMP
48
 
49
  ### Training results
50
 
51
- | Training Loss | Epoch | Step | Validation Loss |
52
- |:-------------:|:------:|:----:|:---------------:|
53
- | 2.7351 | 0.0215 | 1000 | 1.0948 |
54
- | 1.845 | 0.0431 | 2000 | 0.9485 |
55
- | 0.9842 | 0.0646 | 3000 | 0.8843 |
56
- | 1.1293 | 0.0862 | 4000 | 0.8417 |
57
- | 1.1266 | 0.1077 | 5000 | 0.8235 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
58
 
59
 
60
  ### Framework versions
 
17
 
18
  This model is a fine-tuned version of [facebook/seamless-m4t-v2-large](https://huggingface.co/facebook/seamless-m4t-v2-large) on the common_voice_17_0 dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 0.2392
21
 
22
  ## Model description
23
 
 
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: linear
45
  - lr_scheduler_warmup_steps: 500
46
+ - num_epochs: 3.0
47
  - mixed_precision_training: Native AMP
48
 
49
  ### Training results
50
 
51
+ | Training Loss | Epoch | Step | Validation Loss |
52
+ |:-------------:|:------:|:------:|:---------------:|
53
+ | 5.5858 | 0.0215 | 1000 | 1.0859 |
54
+ | 1.6883 | 0.0431 | 2000 | 0.9220 |
55
+ | 1.3121 | 0.0646 | 3000 | 0.7605 |
56
+ | 0.9259 | 0.0862 | 4000 | 0.5180 |
57
+ | 0.6594 | 0.1077 | 5000 | 0.3576 |
58
+ | 0.4965 | 0.1293 | 6000 | 0.3311 |
59
+ | 0.4696 | 0.1508 | 7000 | 0.3133 |
60
+ | 0.4464 | 0.1724 | 8000 | 0.3030 |
61
+ | 0.4097 | 0.1939 | 9000 | 0.3005 |
62
+ | 0.4476 | 0.2155 | 10000 | 0.2926 |
63
+ | 0.4244 | 0.2370 | 11000 | 0.2897 |
64
+ | 0.4148 | 0.2586 | 12000 | 0.2891 |
65
+ | 0.389 | 0.2801 | 13000 | 0.2854 |
66
+ | 0.3815 | 0.3017 | 14000 | 0.2859 |
67
+ | 0.3892 | 0.3232 | 15000 | 0.2832 |
68
+ | 0.3469 | 0.3448 | 16000 | 0.2874 |
69
+ | 0.3817 | 0.3663 | 17000 | 0.2816 |
70
+ | 0.362 | 0.3879 | 18000 | 0.2803 |
71
+ | 0.3667 | 0.4094 | 19000 | 0.2778 |
72
+ | 0.3779 | 0.4310 | 20000 | 0.2837 |
73
+ | 0.3684 | 0.4525 | 21000 | 0.2747 |
74
+ | 0.3363 | 0.4741 | 22000 | 0.2737 |
75
+ | 0.3451 | 0.4956 | 23000 | 0.2732 |
76
+ | 0.3347 | 0.5172 | 24000 | 0.2712 |
77
+ | 0.3686 | 0.5387 | 25000 | 0.2689 |
78
+ | 0.3681 | 0.5603 | 26000 | 0.2692 |
79
+ | 0.3252 | 0.5818 | 27000 | 0.2684 |
80
+ | 0.3296 | 0.6034 | 28000 | 0.2676 |
81
+ | 0.3292 | 0.6249 | 29000 | 0.2686 |
82
+ | 0.3407 | 0.6465 | 30000 | 0.2663 |
83
+ | 0.3606 | 0.6680 | 31000 | 0.2634 |
84
+ | 0.3057 | 0.6896 | 32000 | 0.2611 |
85
+ | 0.3182 | 0.7111 | 33000 | 0.2641 |
86
+ | 0.329 | 0.7327 | 34000 | 0.2626 |
87
+ | 0.3028 | 0.7542 | 35000 | 0.2604 |
88
+ | 0.3131 | 0.7758 | 36000 | 0.2630 |
89
+ | 0.322 | 0.7973 | 37000 | 0.2606 |
90
+ | 0.3428 | 0.8189 | 38000 | 0.2574 |
91
+ | 0.2972 | 0.8404 | 39000 | 0.2566 |
92
+ | 0.3144 | 0.8620 | 40000 | 0.2581 |
93
+ | 0.3118 | 0.8835 | 41000 | 0.2581 |
94
+ | 0.3112 | 0.9051 | 42000 | 0.2562 |
95
+ | 0.3318 | 0.9266 | 43000 | 0.2547 |
96
+ | 0.3034 | 0.9482 | 44000 | 0.2569 |
97
+ | 0.2872 | 0.9697 | 45000 | 0.2564 |
98
+ | 0.3063 | 0.9913 | 46000 | 0.2560 |
99
+ | 0.2705 | 1.0128 | 47000 | 0.2595 |
100
+ | 0.2867 | 1.0344 | 48000 | 0.2552 |
101
+ | 0.2806 | 1.0559 | 49000 | 0.2558 |
102
+ | 0.2623 | 1.0775 | 50000 | 0.2543 |
103
+ | 0.2867 | 1.0990 | 51000 | 0.2540 |
104
+ | 0.2754 | 1.1206 | 52000 | 0.2541 |
105
+ | 0.2872 | 1.1421 | 53000 | 0.2533 |
106
+ | 0.2731 | 1.1637 | 54000 | 0.2532 |
107
+ | 0.2648 | 1.1852 | 55000 | 0.2527 |
108
+ | 0.2779 | 1.2068 | 56000 | 0.2511 |
109
+ | 0.2485 | 1.2283 | 57000 | 0.2523 |
110
+ | 0.2551 | 1.2499 | 58000 | 0.2522 |
111
+ | 0.2856 | 1.2714 | 59000 | 0.2524 |
112
+ | 0.2696 | 1.2930 | 60000 | 0.2507 |
113
+ | 0.2587 | 1.3145 | 61000 | 0.2510 |
114
+ | 0.2373 | 1.3361 | 62000 | 0.2506 |
115
+ | 0.2719 | 1.3576 | 63000 | 0.2502 |
116
+ | 0.2516 | 1.3792 | 64000 | 0.2484 |
117
+ | 0.2623 | 1.4007 | 65000 | 0.2470 |
118
+ | 0.2548 | 1.4223 | 66000 | 0.2466 |
119
+ | 0.2993 | 1.4438 | 67000 | 0.2480 |
120
+ | 0.2676 | 1.4654 | 68000 | 0.2478 |
121
+ | 0.2518 | 1.4869 | 69000 | 0.2475 |
122
+ | 0.2536 | 1.5085 | 70000 | 0.2478 |
123
+ | 0.2764 | 1.5300 | 71000 | 0.2477 |
124
+ | 0.2606 | 1.5516 | 72000 | 0.2482 |
125
+ | 0.2657 | 1.5731 | 73000 | 0.2454 |
126
+ | 0.2877 | 1.5947 | 74000 | 0.2457 |
127
+ | 0.2769 | 1.6162 | 75000 | 0.2464 |
128
+ | 0.2628 | 1.6378 | 76000 | 0.2466 |
129
+ | 0.2554 | 1.6593 | 77000 | 0.2474 |
130
+ | 0.247 | 1.6809 | 78000 | 0.2469 |
131
+ | 0.2882 | 1.7024 | 79000 | 0.2457 |
132
+ | 0.2574 | 1.7240 | 80000 | 0.2449 |
133
+ | 0.2536 | 1.7455 | 81000 | 0.2450 |
134
+ | 0.2608 | 1.7671 | 82000 | 0.2446 |
135
+ | 0.2726 | 1.7886 | 83000 | 0.2433 |
136
+ | 0.2779 | 1.8101 | 84000 | 0.2443 |
137
+ | 0.2294 | 1.8317 | 85000 | 0.2450 |
138
+ | 0.2488 | 1.8532 | 86000 | 0.2427 |
139
+ | 0.2646 | 1.8748 | 87000 | 0.2429 |
140
+ | 0.2802 | 1.8963 | 88000 | 0.2441 |
141
+ | 0.2655 | 1.9179 | 89000 | 0.2420 |
142
+ | 0.2441 | 1.9394 | 90000 | 0.2420 |
143
+ | 0.26 | 1.9610 | 91000 | 0.2416 |
144
+ | 0.2239 | 1.9825 | 92000 | 0.2417 |
145
+ | 0.2601 | 2.0041 | 93000 | 0.2415 |
146
+ | 0.2208 | 2.0256 | 94000 | 0.2418 |
147
+ | 0.2309 | 2.0472 | 95000 | 0.2439 |
148
+ | 0.2541 | 2.0687 | 96000 | 0.2424 |
149
+ | 0.2288 | 2.0903 | 97000 | 0.2423 |
150
+ | 0.2102 | 2.1118 | 98000 | 0.2418 |
151
+ | 0.2288 | 2.1334 | 99000 | 0.2426 |
152
+ | 0.2429 | 2.1549 | 100000 | 0.2425 |
153
+ | 0.2123 | 2.1765 | 101000 | 0.2425 |
154
+ | 0.2396 | 2.1980 | 102000 | 0.2408 |
155
+ | 0.2249 | 2.2196 | 103000 | 0.2412 |
156
+ | 0.2053 | 2.2411 | 104000 | 0.2429 |
157
+ | 0.2055 | 2.2627 | 105000 | 0.2421 |
158
+ | 0.1957 | 2.2842 | 106000 | 0.2407 |
159
+ | 0.2444 | 2.3058 | 107000 | 0.2414 |
160
+ | 0.1988 | 2.3273 | 108000 | 0.2417 |
161
+ | 0.2241 | 2.3489 | 109000 | 0.2428 |
162
+ | 0.2575 | 2.3704 | 110000 | 0.2418 |
163
+ | 0.254 | 2.3920 | 111000 | 0.2412 |
164
+ | 0.2535 | 2.4135 | 112000 | 0.2401 |
165
+ | 0.234 | 2.4351 | 113000 | 0.2402 |
166
+ | 0.2387 | 2.4566 | 114000 | 0.2408 |
167
+ | 0.2406 | 2.4782 | 115000 | 0.2408 |
168
+ | 0.2145 | 2.4997 | 116000 | 0.2406 |
169
+ | 0.235 | 2.5213 | 117000 | 0.2401 |
170
+ | 0.226 | 2.5428 | 118000 | 0.2403 |
171
+ | 0.2241 | 2.5644 | 119000 | 0.2399 |
172
+ | 0.2411 | 2.5859 | 120000 | 0.2396 |
173
+ | 0.2495 | 2.6075 | 121000 | 0.2390 |
174
+ | 0.2244 | 2.6290 | 122000 | 0.2396 |
175
+ | 0.2306 | 2.6506 | 123000 | 0.2401 |
176
+ | 0.242 | 2.6721 | 124000 | 0.2404 |
177
+ | 0.233 | 2.6937 | 125000 | 0.2398 |
178
+ | 0.1819 | 2.7152 | 126000 | 0.2406 |
179
+ | 0.2267 | 2.7368 | 127000 | 0.2402 |
180
+ | 0.2227 | 2.7583 | 128000 | 0.2395 |
181
+ | 0.2155 | 2.7799 | 129000 | 0.2396 |
182
+ | 0.2369 | 2.8014 | 130000 | 0.2395 |
183
+ | 0.1982 | 2.8230 | 131000 | 0.2397 |
184
+ | 0.2245 | 2.8445 | 132000 | 0.2398 |
185
+ | 0.238 | 2.8661 | 133000 | 0.2394 |
186
+ | 0.2503 | 2.8876 | 134000 | 0.2392 |
187
+ | 0.2533 | 2.9092 | 135000 | 0.2388 |
188
+ | 0.236 | 2.9307 | 136000 | 0.2391 |
189
+ | 0.2404 | 2.9523 | 137000 | 0.2392 |
190
+ | 0.2374 | 2.9738 | 138000 | 0.2392 |
191
+ | 0.199 | 2.9954 | 139000 | 0.2392 |
192
 
193
 
194
  ### Framework versions
model-00001-of-00002.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3d1c775ced017c03a64e091d21a11fbc70399fbc18283de1f2a5f876b64c48e8
3
  size 4999163080
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d1abece29116d7847f2069726d5c9d92ca1f991406b863a45347e7649176576c
3
  size 4999163080
model-00002-of-00002.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:44d719d7925a3f24c277433f226efc8ebbfb7f01747fa2d7f1a97b30197b1dab
3
  size 4238114628
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ea31cc7e6f35096591ffcd378acf9eb1e2c65ba3a62c93b67b4911bc98dc2e38
3
  size 4238114628