Hugging Face's logo

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

Generated from Trainer
Model card Files Community
Shakhovak commited on
Commit
4f8f3c3
·
verified ·
1 Parent(s): 6b42652

End of training

Browse files
Files changed (3) hide show
  1. README.md +5 -39
  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: 0.1038
18
 
19
  ## Model description
20
 
@@ -40,49 +40,15 @@ 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: 12
44
 
45
  ### Training results
46
 
47
  | Training Loss | Epoch | Step | Validation Loss |
48
  |:-------------:|:-----:|:----:|:---------------:|
49
- | 4.346 | 0.32 | 200 | 3.4570 |
50
- | 1.5615 | 0.63 | 400 | 0.5494 |
51
- | 0.5001 | 0.95 | 600 | 0.3109 |
52
- | 0.3495 | 1.27 | 800 | 0.2476 |
53
- | 0.2841 | 1.58 | 1000 | 0.2129 |
54
- | 0.2602 | 1.9 | 1200 | 0.1858 |
55
- | 0.2273 | 2.22 | 1400 | 0.1740 |
56
- | 0.2095 | 2.53 | 1600 | 0.1641 |
57
- | 0.2024 | 2.85 | 1800 | 0.1547 |
58
- | 0.1981 | 3.16 | 2000 | 0.1554 |
59
- | 0.1816 | 3.48 | 2200 | 0.1493 |
60
- | 0.1749 | 3.8 | 2400 | 0.1407 |
61
- | 0.1668 | 4.11 | 2600 | 0.1358 |
62
- | 0.156 | 4.43 | 2800 | 0.1420 |
63
- | 0.1581 | 4.75 | 3000 | 0.1336 |
64
- | 0.1525 | 5.06 | 3200 | 0.1266 |
65
- | 0.1391 | 5.38 | 3400 | 0.1277 |
66
- | 0.1477 | 5.7 | 3600 | 0.1215 |
67
- | 0.1418 | 6.01 | 3800 | 0.1193 |
68
- | 0.1355 | 6.33 | 4000 | 0.1216 |
69
- | 0.1261 | 6.65 | 4200 | 0.1225 |
70
- | 0.1298 | 6.96 | 4400 | 0.1181 |
71
- | 0.1266 | 7.28 | 4600 | 0.1168 |
72
- | 0.1215 | 7.59 | 4800 | 0.1158 |
73
- | 0.132 | 7.91 | 5000 | 0.1162 |
74
- | 0.1178 | 8.23 | 5200 | 0.1118 |
75
- | 0.1194 | 8.54 | 5400 | 0.1123 |
76
- | 0.1192 | 8.86 | 5600 | 0.1081 |
77
- | 0.1116 | 9.18 | 5800 | 0.1070 |
78
- | 0.1204 | 9.49 | 6000 | 0.1064 |
79
- | 0.11 | 9.81 | 6200 | 0.1094 |
80
- | 0.1132 | 10.13 | 6400 | 0.1064 |
81
- | 0.1134 | 10.44 | 6600 | 0.1036 |
82
- | 0.1065 | 10.76 | 6800 | 0.1043 |
83
- | 0.1003 | 11.08 | 7000 | 0.1046 |
84
- | 0.1119 | 11.39 | 7200 | 0.1044 |
85
- | 0.1082 | 11.71 | 7400 | 0.1038 |
86
 
87
 
88
  ### 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: 0.2781
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: 1
44
 
45
  ### Training results
46
 
47
  | Training Loss | Epoch | Step | Validation Loss |
48
  |:-------------:|:-----:|:----:|:---------------:|
49
+ | 2.2483 | 0.32 | 200 | 0.4278 |
50
+ | 0.4429 | 0.63 | 400 | 0.3025 |
51
+ | 0.3615 | 0.95 | 600 | 0.2781 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
52
 
53
 
54
  ### Framework versions
adapter_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:af7d4f9e6fb88a814183a4a576f526a1e7b6324f2b1ee642950bcebf8c298820
3
  size 37855242
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:19d0ea9c4a6c29a1a564978be3cb0849cc1755892f74c3e80358da5fd923fc16
3
  size 37855242
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d50f27f1e2e56a7ccdbf5cebc2029a7059ac735aff4b8d1ca890c98f96f4bc0d
3
  size 5112
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8418f134a3051e363af41d1f3cfa63a8c9f99873cee2842aa059d1f0e990955e
3
  size 5112