Hugging Face's logo

Shakhovak
/
flan-t5-xl-absa-multitask-rest

Generated from Trainer
Model card Files Community
Shakhovak commited on
Commit
b92e423
•
1 Parent(s): b8d143c

End of training

Browse files
Files changed (3) hide show
  1. README.md +39 -33
  2. adapter_model.bin +1 -1
  3. training_args.bin +1 -1
README.md CHANGED
@@ -14,7 +14,7 @@ should probably proofread and complete it, then remove this comment. -->
14
 
15
  This model is a fine-tuned version of [ybelkada/flan-t5-xl-sharded-bf16](https://huggingface.co/ybelkada/flan-t5-xl-sharded-bf16) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
- - Loss: 0.1206
18
 
19
  ## Model description
20
 
@@ -40,43 +40,49 @@ The following hyperparameters were used during training:
40
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
41
  - lr_scheduler_type: linear
42
  - lr_scheduler_warmup_ratio: 0.1
43
- - num_epochs: 10
44
 
45
  ### Training results
46
 
47
  | Training Loss | Epoch | Step | Validation Loss |
48
  |:-------------:|:-----:|:----:|:---------------:|
49
- | 4.2698 | 0.32 | 200 | 2.8711 |
50
- | 1.3101 | 0.63 | 400 | 0.4477 |
51
- | 0.4353 | 0.95 | 600 | 0.2831 |
52
- | 0.324 | 1.27 | 800 | 0.2462 |
53
- | 0.2806 | 1.58 | 1000 | 0.2112 |
54
- | 0.2527 | 1.9 | 1200 | 0.1953 |
55
- | 0.2256 | 2.22 | 1400 | 0.1847 |
56
- | 0.2206 | 2.53 | 1600 | 0.1724 |
57
- | 0.2048 | 2.85 | 1800 | 0.1673 |
58
- | 0.1872 | 3.16 | 2000 | 0.1638 |
59
- | 0.177 | 3.48 | 2200 | 0.1607 |
60
- | 0.1749 | 3.8 | 2400 | 0.1511 |
61
- | 0.1736 | 4.11 | 2600 | 0.1436 |
62
- | 0.16 | 4.43 | 2800 | 0.1424 |
63
- | 0.1553 | 4.75 | 3000 | 0.1444 |
64
- | 0.1476 | 5.06 | 3200 | 0.1399 |
65
- | 0.1437 | 5.38 | 3400 | 0.1351 |
66
- | 0.1459 | 5.7 | 3600 | 0.1309 |
67
- | 0.1469 | 6.01 | 3800 | 0.1277 |
68
- | 0.1363 | 6.33 | 4000 | 0.1290 |
69
- | 0.1393 | 6.65 | 4200 | 0.1278 |
70
- | 0.1343 | 6.96 | 4400 | 0.1285 |
71
- | 0.1269 | 7.28 | 4600 | 0.1271 |
72
- | 0.1362 | 7.59 | 4800 | 0.1206 |
73
- | 0.1271 | 7.91 | 5000 | 0.1256 |
74
- | 0.1293 | 8.23 | 5200 | 0.1239 |
75
- | 0.1189 | 8.54 | 5400 | 0.1227 |
76
- | 0.1218 | 8.86 | 5600 | 0.1209 |
77
- | 0.1171 | 9.18 | 5800 | 0.1234 |
78
- | 0.124 | 9.49 | 6000 | 0.1222 |
79
- | 0.1218 | 9.81 | 6200 | 0.1206 |
 
 
 
 
 
 
80
 
81
 
82
  ### Framework versions
 
14
 
15
  This model is a fine-tuned version of [ybelkada/flan-t5-xl-sharded-bf16](https://huggingface.co/ybelkada/flan-t5-xl-sharded-bf16) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
+ - Loss: 0.1127
18
 
19
  ## Model description
20
 
 
40
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
41
  - lr_scheduler_type: linear
42
  - lr_scheduler_warmup_ratio: 0.1
43
+ - num_epochs: 12
44
 
45
  ### Training results
46
 
47
  | Training Loss | Epoch | Step | Validation Loss |
48
  |:-------------:|:-----:|:----:|:---------------:|
49
+ | 4.3549 | 0.32 | 200 | 3.5848 |
50
+ | 1.5908 | 0.63 | 400 | 0.5331 |
51
+ | 0.4981 | 0.95 | 600 | 0.3159 |
52
+ | 0.351 | 1.27 | 800 | 0.2457 |
53
+ | 0.2884 | 1.58 | 1000 | 0.2118 |
54
+ | 0.2592 | 1.9 | 1200 | 0.2000 |
55
+ | 0.2323 | 2.22 | 1400 | 0.1839 |
56
+ | 0.2107 | 2.53 | 1600 | 0.1704 |
57
+ | 0.2071 | 2.85 | 1800 | 0.1649 |
58
+ | 0.1944 | 3.16 | 2000 | 0.1634 |
59
+ | 0.1774 | 3.48 | 2200 | 0.1549 |
60
+ | 0.1796 | 3.8 | 2400 | 0.1505 |
61
+ | 0.1695 | 4.11 | 2600 | 0.1427 |
62
+ | 0.1569 | 4.43 | 2800 | 0.1403 |
63
+ | 0.1662 | 4.75 | 3000 | 0.1395 |
64
+ | 0.15 | 5.06 | 3200 | 0.1351 |
65
+ | 0.1448 | 5.38 | 3400 | 0.1283 |
66
+ | 0.1444 | 5.7 | 3600 | 0.1302 |
67
+ | 0.1506 | 6.01 | 3800 | 0.1237 |
68
+ | 0.1321 | 6.33 | 4000 | 0.1264 |
69
+ | 0.1318 | 6.65 | 4200 | 0.1269 |
70
+ | 0.1298 | 6.96 | 4400 | 0.1207 |
71
+ | 0.1273 | 7.28 | 4600 | 0.1224 |
72
+ | 0.123 | 7.59 | 4800 | 0.1209 |
73
+ | 0.1278 | 7.91 | 5000 | 0.1222 |
74
+ | 0.1236 | 8.23 | 5200 | 0.1165 |
75
+ | 0.1188 | 8.54 | 5400 | 0.1154 |
76
+ | 0.1181 | 8.86 | 5600 | 0.1173 |
77
+ | 0.1126 | 9.18 | 5800 | 0.1177 |
78
+ | 0.113 | 9.49 | 6000 | 0.1194 |
79
+ | 0.1086 | 9.81 | 6200 | 0.1148 |
80
+ | 0.1086 | 10.13 | 6400 | 0.1158 |
81
+ | 0.1118 | 10.44 | 6600 | 0.1145 |
82
+ | 0.105 | 10.76 | 6800 | 0.1125 |
83
+ | 0.1119 | 11.08 | 7000 | 0.1146 |
84
+ | 0.1007 | 11.39 | 7200 | 0.1123 |
85
+ | 0.114 | 11.71 | 7400 | 0.1127 |
86
 
87
 
88
  ### Framework versions
adapter_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:17a66a0ed3d8a8b39cdfc98b26948b9a114b8dae4ee799ec4a6f9dad252da41e
3
  size 37855242
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3c7c285c863b0eb89d589d2134ec5dff9e4914b1bf647c7157d83374f08dd098
3
  size 37855242
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a65946cfc661dcfe81069010e83b2f90523a4b9fd121756ea63c5a142de616f2
3
  size 5112
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6f8a2b551cfe8e422c842435194befece9ce633e04ac83567b0dc00fb94a1eae
3
  size 5112