Hugging Face's logo

summervent
/
speller-t5-8

Text2Text Generation
Transformers
PyTorch
TensorBoard
t5
Generated from Trainer
text-generation-inference
Inference Endpoints
Model card Files Metrics Community
summervent commited on
Commit
0563629
•
1 Parent(s): e03ab1a

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +33 -33
README.md CHANGED
@@ -15,12 +15,12 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  This model is a fine-tuned version of [sberbank-ai/ruT5-base](https://huggingface.co/sberbank-ai/ruT5-base) on the None dataset.
17
  It achieves the following results on the evaluation set:
18
- - Loss: 0.1909
19
- - Rouge1: 28.0644
20
- - Rouge2: 13.5886
21
- - Rougel: 27.9279
22
- - Rougelsum: 28.1805
23
- - Gen Len: 37.3063
24
 
25
  ## Model description
26
 
@@ -52,33 +52,33 @@ The following hyperparameters were used during training:
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
54
  |:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
55
- | 1.1356 | 0.04 | 500 | 0.5617 | 23.6812 | 11.3814 | 23.476 | 23.6937 | 48.6126 |
56
- | 0.763 | 0.07 | 1000 | 0.4267 | 25.6757 | 12.0621 | 25.5562 | 25.9692 | 39.6486 |
57
- | 0.6036 | 0.11 | 1500 | 0.3609 | 27.027 | 13.5886 | 27.027 | 27.3273 | 38.4054 |
58
- | 0.5877 | 0.14 | 2000 | 0.3398 | 26.2285 | 13.5886 | 26.1603 | 26.4128 | 39.1712 |
59
- | 0.5303 | 0.18 | 2500 | 0.3104 | 27.027 | 13.5886 | 27.027 | 27.3273 | 37.3604 |
60
- | 0.5248 | 0.22 | 3000 | 0.2936 | 27.027 | 13.5886 | 27.027 | 27.3273 | 37.4775 |
61
- | 0.5093 | 0.25 | 3500 | 0.2816 | 27.027 | 13.5886 | 27.027 | 27.3273 | 37.5495 |
62
- | 0.462 | 0.29 | 4000 | 0.2767 | 28.0644 | 13.5886 | 27.9279 | 28.1805 | 37.8018 |
63
- | 0.4414 | 0.33 | 4500 | 0.2622 | 28.0644 | 13.5886 | 27.9279 | 28.1805 | 37.6126 |
64
- | 0.3651 | 0.36 | 5000 | 0.2547 | 29.0404 | 14.4895 | 28.9585 | 29.0404 | 37.5946 |
65
- | 0.4351 | 0.4 | 5500 | 0.2464 | 29.0404 | 14.4895 | 28.9585 | 29.0404 | 37.8108 |
66
- | 0.4058 | 0.43 | 6000 | 0.2423 | 28.0644 | 13.5886 | 27.9279 | 28.1805 | 37.4414 |
67
- | 0.3649 | 0.47 | 6500 | 0.2391 | 29.0404 | 14.4895 | 28.9585 | 29.0404 | 37.7207 |
68
- | 0.4332 | 0.51 | 7000 | 0.2306 | 29.0404 | 14.4895 | 28.9585 | 29.0404 | 37.7207 |
69
- | 0.3359 | 0.54 | 7500 | 0.2262 | 29.0404 | 14.4895 | 28.9585 | 29.0404 | 37.7117 |
70
- | 0.3227 | 0.58 | 8000 | 0.2254 | 28.0644 | 13.5886 | 27.9279 | 28.1805 | 37.4685 |
71
- | 0.3254 | 0.61 | 8500 | 0.2218 | 28.0644 | 13.5886 | 27.9279 | 28.1805 | 37.3604 |
72
- | 0.3405 | 0.65 | 9000 | 0.2160 | 29.0404 | 14.4895 | 28.9585 | 29.0404 | 37.5405 |
73
- | 0.3272 | 0.69 | 9500 | 0.2151 | 29.0404 | 14.4895 | 28.9585 | 29.0404 | 37.4955 |
74
- | 0.3133 | 0.72 | 10000 | 0.2098 | 28.0644 | 13.5886 | 27.9279 | 28.1805 | 37.2162 |
75
- | 0.2889 | 0.76 | 10500 | 0.2125 | 29.0404 | 14.4895 | 28.9585 | 29.0404 | 37.5856 |
76
- | 0.3006 | 0.79 | 11000 | 0.2030 | 29.0404 | 14.4895 | 28.9585 | 29.0404 | 37.4865 |
77
- | 0.3403 | 0.83 | 11500 | 0.2071 | 28.0644 | 13.5886 | 27.9279 | 28.1805 | 37.4414 |
78
- | 0.3563 | 0.87 | 12000 | 0.2020 | 29.0404 | 14.4895 | 28.9585 | 29.0404 | 37.3694 |
79
- | 0.2706 | 0.9 | 12500 | 0.1988 | 28.0644 | 13.5886 | 27.9279 | 28.1805 | 37.2252 |
80
- | 0.3384 | 0.94 | 13000 | 0.1963 | 28.0644 | 13.5886 | 27.9279 | 28.1805 | 37.3694 |
81
- | 0.325 | 0.98 | 13500 | 0.1909 | 28.0644 | 13.5886 | 27.9279 | 28.1805 | 37.3063 |
82
 
83
 
84
  ### Framework versions
 
15
 
16
  This model is a fine-tuned version of [sberbank-ai/ruT5-base](https://huggingface.co/sberbank-ai/ruT5-base) on the None dataset.
17
  It achieves the following results on the evaluation set:
18
+ - Loss: 0.1580
19
+ - Rouge1: 28.0246
20
+ - Rouge2: 14.6131
21
+ - Rougel: 28.0357
22
+ - Rougelsum: 28.1585
23
+ - Gen Len: 41.6429
24
 
25
  ## Model description
26
 
 
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
54
  |:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
55
+ | 1.0444 | 0.04 | 500 | 0.5084 | 23.8669 | 11.0011 | 23.8074 | 23.9137 | 48.2589 |
56
+ | 0.9248 | 0.07 | 1000 | 0.3787 | 27.1237 | 13.047 | 27.128 | 27.286 | 44.6518 |
57
+ | 0.6515 | 0.11 | 1500 | 0.3261 | 26.7152 | 12.8472 | 26.7857 | 27.0912 | 44.6429 |
58
+ | 0.5533 | 0.14 | 2000 | 0.2929 | 27.5391 | 13.497 | 27.5056 | 27.7679 | 42.2232 |
59
+ | 0.5962 | 0.18 | 2500 | 0.2678 | 27.2948 | 13.5317 | 27.3026 | 27.4828 | 42.7411 |
60
+ | 0.5667 | 0.22 | 3000 | 0.2549 | 27.4315 | 13.7922 | 27.4858 | 27.6141 | 42.5 |
61
+ | 0.4532 | 0.25 | 3500 | 0.2460 | 28.0246 | 14.6131 | 28.0357 | 28.1585 | 42.3929 |
62
+ | 0.4552 | 0.29 | 4000 | 0.2320 | 27.5 | 13.9435 | 27.567 | 27.6953 | 42.1161 |
63
+ | 0.4245 | 0.32 | 4500 | 0.2264 | 27.3214 | 13.3333 | 27.3884 | 27.5502 | 42.1607 |
64
+ | 0.4109 | 0.36 | 5000 | 0.2182 | 27.5 | 13.9435 | 27.567 | 27.6953 | 42.2411 |
65
+ | 0.3826 | 0.4 | 5500 | 0.2115 | 27.5 | 13.9435 | 27.567 | 27.6953 | 41.9196 |
66
+ | 0.4112 | 0.43 | 6000 | 0.2066 | 27.5 | 13.9435 | 27.567 | 27.6953 | 41.9018 |
67
+ | 0.4006 | 0.47 | 6500 | 0.1980 | 28.0246 | 14.6131 | 28.0357 | 28.1585 | 41.8304 |
68
+ | 0.3494 | 0.5 | 7000 | 0.1944 | 27.5 | 13.9435 | 27.567 | 27.6953 | 41.9554 |
69
+ | 0.3225 | 0.54 | 7500 | 0.1928 | 27.5 | 13.9435 | 27.567 | 27.6953 | 41.8929 |
70
+ | 0.3525 | 0.58 | 8000 | 0.1869 | 27.5 | 13.9435 | 27.567 | 27.6953 | 41.6071 |
71
+ | 0.3828 | 0.61 | 8500 | 0.1821 | 27.5 | 13.9435 | 27.567 | 27.6953 | 41.5446 |
72
+ | 0.3287 | 0.65 | 9000 | 0.1781 | 27.5 | 13.9435 | 27.567 | 27.6953 | 41.5714 |
73
+ | 0.3276 | 0.68 | 9500 | 0.1778 | 27.5 | 13.9435 | 27.567 | 27.6953 | 41.6786 |
74
+ | 0.3054 | 0.72 | 10000 | 0.1727 | 27.5 | 13.9435 | 27.567 | 27.6953 | 41.4375 |
75
+ | 0.3685 | 0.76 | 10500 | 0.1728 | 27.5 | 13.9435 | 27.567 | 27.6953 | 41.6964 |
76
+ | 0.3454 | 0.79 | 11000 | 0.1700 | 27.5 | 13.9435 | 27.567 | 27.6953 | 41.75 |
77
+ | 0.3056 | 0.83 | 11500 | 0.1641 | 27.5 | 13.9435 | 27.567 | 27.6953 | 41.7143 |
78
+ | 0.3399 | 0.86 | 12000 | 0.1606 | 27.5 | 13.9435 | 27.567 | 27.6953 | 41.7143 |
79
+ | 0.3079 | 0.9 | 12500 | 0.1600 | 27.5 | 13.9435 | 27.567 | 27.6953 | 41.6429 |
80
+ | 0.2646 | 0.94 | 13000 | 0.1591 | 28.0246 | 14.6131 | 28.0357 | 28.1585 | 41.5446 |
81
+ | 0.2297 | 0.97 | 13500 | 0.1580 | 28.0246 | 14.6131 | 28.0357 | 28.1585 | 41.6429 |
82
 
83
 
84
  ### Framework versions