madatnlp commited on
Commit
a87282b
·
1 Parent(s): 045a288
Files changed (2) hide show
  1. README.md +39 -47
  2. tf_model.h5 +1 -1
README.md CHANGED
@@ -13,9 +13,9 @@ probably proofread and complete it, then remove this comment. -->
13
 
14
  This model is a fine-tuned version of [facebook/mbart-large-50](https://huggingface.co/facebook/mbart-large-50) on an unknown dataset.
15
  It achieves the following results on the evaluation set:
16
- - Train Loss: 1.7384
17
- - Validation Loss: 1.6859
18
- - Epoch: 43
19
 
20
  ## Model description
21
 
@@ -41,50 +41,42 @@ The following hyperparameters were used during training:
41
 
42
  | Train Loss | Validation Loss | Epoch |
43
  |:----------:|:---------------:|:-----:|
44
- | 12.5168 | 17.3038 | 0 |
45
- | 13.4675 | 12.0726 | 1 |
46
- | 6.7752 | 4.7717 | 2 |
47
- | 4.2218 | 3.7952 | 3 |
48
- | 3.6988 | 3.5533 | 4 |
49
- | 3.5477 | 3.4101 | 5 |
50
- | 3.4128 | 3.3961 | 6 |
51
- | 3.3888 | 3.3118 | 7 |
52
- | 3.3311 | 3.2826 | 8 |
53
- | 3.2614 | 3.2270 | 9 |
54
- | 3.2088 | 3.2100 | 10 |
55
- | 3.1062 | 3.0641 | 11 |
56
- | 3.0133 | 2.9382 | 12 |
57
- | 2.9223 | 2.8687 | 13 |
58
- | 2.8273 | 2.7063 | 14 |
59
- | 2.7913 | 2.8055 | 15 |
60
- | 2.6951 | 2.7133 | 16 |
61
- | 2.6631 | 2.5467 | 17 |
62
- | 2.5664 | 2.5094 | 18 |
63
- | 2.5150 | 2.4914 | 19 |
64
- | 2.4462 | 2.3662 | 20 |
65
- | 2.4081 | 2.3500 | 21 |
66
- | 2.3605 | 2.2733 | 22 |
67
- | 2.3164 | 2.2821 | 23 |
68
- | 2.3014 | 2.2191 | 24 |
69
- | 2.2103 | 2.0691 | 25 |
70
- | 2.1883 | 2.1237 | 26 |
71
- | 2.1814 | 2.1356 | 27 |
72
- | 2.1292 | 2.0151 | 28 |
73
- | 2.1046 | 1.9359 | 29 |
74
- | 2.0542 | 2.0341 | 30 |
75
- | 2.0365 | 1.9446 | 31 |
76
- | 1.9829 | 1.8999 | 32 |
77
- | 1.9801 | 1.8124 | 33 |
78
- | 1.9168 | 1.7824 | 34 |
79
- | 1.9209 | 1.8382 | 35 |
80
- | 1.8850 | 1.8215 | 36 |
81
- | 1.8748 | 1.7645 | 37 |
82
- | 1.8346 | 1.6847 | 38 |
83
- | 1.8076 | 1.7017 | 39 |
84
- | 1.8732 | 1.8328 | 40 |
85
- | 1.8164 | 1.7083 | 41 |
86
- | 1.7784 | 1.7930 | 42 |
87
- | 1.7384 | 1.6859 | 43 |
88
 
89
 
90
  ### Framework versions
 
13
 
14
  This model is a fine-tuned version of [facebook/mbart-large-50](https://huggingface.co/facebook/mbart-large-50) on an unknown dataset.
15
  It achieves the following results on the evaluation set:
16
+ - Train Loss: 0.5342
17
+ - Validation Loss: 0.5633
18
+ - Epoch: 35
19
 
20
  ## Model description
21
 
 
41
 
42
  | Train Loss | Validation Loss | Epoch |
43
  |:----------:|:---------------:|:-----:|
44
+ | 4.5626 | 3.7843 | 0 |
45
+ | 2.5836 | 1.9212 | 1 |
46
+ | 1.6546 | 1.2552 | 2 |
47
+ | 1.2499 | 1.0248 | 3 |
48
+ | 1.0088 | 0.8457 | 4 |
49
+ | 0.9100 | 0.7958 | 5 |
50
+ | 0.8290 | 0.8421 | 6 |
51
+ | 0.7999 | 0.7625 | 7 |
52
+ | 0.7633 | 0.7202 | 8 |
53
+ | 0.7439 | 0.7100 | 9 |
54
+ | 0.7182 | 0.6787 | 10 |
55
+ | 0.7092 | 0.6877 | 11 |
56
+ | 0.6823 | 0.6684 | 12 |
57
+ | 0.6738 | 0.6712 | 13 |
58
+ | 0.6603 | 0.6858 | 14 |
59
+ | 0.6462 | 0.6268 | 15 |
60
+ | 0.6373 | 0.6208 | 16 |
61
+ | 0.6424 | 0.6735 | 17 |
62
+ | 0.6259 | 0.6423 | 18 |
63
+ | 0.6249 | 0.6069 | 19 |
64
+ | 0.6148 | 0.6510 | 20 |
65
+ | 0.6063 | 0.6207 | 21 |
66
+ | 0.5987 | 0.5977 | 22 |
67
+ | 0.5917 | 0.6019 | 23 |
68
+ | 0.5800 | 0.5828 | 24 |
69
+ | 0.5779 | 0.5505 | 25 |
70
+ | 0.5765 | 0.5887 | 26 |
71
+ | 0.5667 | 0.5989 | 27 |
72
+ | 0.5623 | 0.5859 | 28 |
73
+ | 0.5564 | 0.5907 | 29 |
74
+ | 0.5523 | 0.5928 | 30 |
75
+ | 0.5478 | 0.5624 | 31 |
76
+ | 0.5472 | 0.5563 | 32 |
77
+ | 0.5462 | 0.5953 | 33 |
78
+ | 0.5324 | 0.5593 | 34 |
79
+ | 0.5342 | 0.5633 | 35 |
 
 
 
 
 
 
 
 
80
 
81
 
82
  ### Framework versions
tf_model.h5 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d890c143526f0a4a71d390c345c5c53f903e705421c77321e0a90e0e9293d31f
3
  size 2445079280
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b1c2daabd53d121f49bcb1caf09aa2588149a8d7ccc7b42fb4dbf18b34b7fb9
3
  size 2445079280