Hugging Face's logo

makhataei
/
qa-persian-albert-fa-zwnj-base-v2

Question Answering
Transformers
TensorBoard
Safetensors
albert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
makhataei commited on
Commit
04916e9
路
verified 路
1 Parent(s): cfc3004

End of training

Browse files
Files changed (4) hide show
  1. README.md +19 -49
  2. model.safetensors +1 -1
  3. runs/Mar02_19-03-38_Software-AI/events.out.tfevents.1709393618.Software-AI.552873.2 +3 -0
  4. training_args.bin +2 -2
README.md CHANGED
@@ -4,7 +4,7 @@ base_model: makhataei/qa-persian-albert-fa-zwnj-base-v2
4
  tags:
5
  - generated_from_trainer
6
  datasets:
7
- - pquad
8
  model-index:
9
  - name: qa-persian-albert-fa-zwnj-base-v2
10
  results: []
@@ -15,9 +15,9 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # qa-persian-albert-fa-zwnj-base-v2
17
 
18
- This model is a fine-tuned version of [makhataei/qa-persian-albert-fa-zwnj-base-v2](https://huggingface.co/makhataei/qa-persian-albert-fa-zwnj-base-v2) on the pquad dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 2.1293
21
 
22
  ## Model description
23
 
@@ -36,58 +36,28 @@ More information needed
36
  ### Training hyperparameters
37
 
38
  The following hyperparameters were used during training:
39
- - learning_rate: 4.8828125e-08
40
- - train_batch_size: 16
41
- - eval_batch_size: 16
42
  - seed: 42
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: linear
45
- - num_epochs: 5
46
 
47
  ### Training results
48
 
49
- | Training Loss | Epoch | Step | Validation Loss |
50
- |:-------------:|:-----:|:-----:|:---------------:|
51
- | 0.0066 | 0.12 | 500 | 2.1456 |
52
- | 0.0278 | 0.25 | 1000 | 2.2019 |
53
- | 0.0904 | 0.38 | 1500 | 2.2192 |
54
- | 0.1574 | 0.5 | 2000 | 2.2146 |
55
- | 0.1628 | 0.62 | 2500 | 2.2076 |
56
- | 0.1991 | 0.75 | 3000 | 2.1965 |
57
- | 0.2063 | 0.88 | 3500 | 2.1837 |
58
- | 0.2293 | 1.0 | 4000 | 2.1688 |
59
- | 0.1453 | 1.12 | 4500 | 2.1645 |
60
- | 0.1307 | 1.25 | 5000 | 2.1616 |
61
- | 0.1363 | 1.38 | 5500 | 2.1594 |
62
- | 0.1286 | 1.5 | 6000 | 2.1575 |
63
- | 0.1323 | 1.62 | 6500 | 2.1546 |
64
- | 0.1243 | 1.75 | 7000 | 2.1538 |
65
- | 0.13 | 1.88 | 7500 | 2.1506 |
66
- | 0.1245 | 2.0 | 8000 | 2.1489 |
67
- | 0.1295 | 2.12 | 8500 | 2.1478 |
68
- | 0.1318 | 2.25 | 9000 | 2.1450 |
69
- | 0.1382 | 2.38 | 9500 | 2.1433 |
70
- | 0.1237 | 2.5 | 10000 | 2.1430 |
71
- | 0.1369 | 2.62 | 10500 | 2.1404 |
72
- | 0.1226 | 2.75 | 11000 | 2.1390 |
73
- | 0.1314 | 2.88 | 11500 | 2.1377 |
74
- | 0.1241 | 3.0 | 12000 | 2.1370 |
75
- | 0.1223 | 3.12 | 12500 | 2.1368 |
76
- | 0.1286 | 3.25 | 13000 | 2.1358 |
77
- | 0.1289 | 3.38 | 13500 | 2.1350 |
78
- | 0.1207 | 3.5 | 14000 | 2.1341 |
79
- | 0.1374 | 3.62 | 14500 | 2.1330 |
80
- | 0.1331 | 3.75 | 15000 | 2.1319 |
81
- | 0.1322 | 3.88 | 15500 | 2.1311 |
82
- | 0.1264 | 4.0 | 16000 | 2.1308 |
83
- | 0.1271 | 4.12 | 16500 | 2.1302 |
84
- | 0.1233 | 4.25 | 17000 | 2.1301 |
85
- | 0.1357 | 4.38 | 17500 | 2.1296 |
86
- | 0.1245 | 4.5 | 18000 | 2.1295 |
87
- | 0.1329 | 4.62 | 18500 | 2.1294 |
88
- | 0.1193 | 4.75 | 19000 | 2.1293 |
89
- | 0.1345 | 4.88 | 19500 | 2.1293 |
90
- | 0.1275 | 5.0 | 20000 | 2.1293 |
91
 
92
 
93
  ### Framework versions
 
4
  tags:
5
  - generated_from_trainer
6
  datasets:
7
+ - parsinlu_reading_comprehension
8
  model-index:
9
  - name: qa-persian-albert-fa-zwnj-base-v2
10
  results: []
 
15
 
16
  # qa-persian-albert-fa-zwnj-base-v2
17
 
18
+ This model is a fine-tuned version of [makhataei/qa-persian-albert-fa-zwnj-base-v2](https://huggingface.co/makhataei/qa-persian-albert-fa-zwnj-base-v2) on the parsinlu_reading_comprehension dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 8.0574
21
 
22
  ## Model description
23
 
 
36
  ### Training hyperparameters
37
 
38
  The following hyperparameters were used during training:
39
+ - learning_rate: 0.00025
40
+ - train_batch_size: 14
41
+ - eval_batch_size: 14
42
  - seed: 42
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: linear
45
+ - num_epochs: 10
46
 
47
  ### Training results
48
 
49
+ | Training Loss | Epoch | Step | Validation Loss |
50
+ |:-------------:|:-----:|:----:|:---------------:|
51
+ | 3.141 | 1.0 | 44 | 2.8363 |
52
+ | 1.7324 | 2.0 | 88 | 3.0483 |
53
+ | 0.9041 | 3.0 | 132 | 3.4902 |
54
+ | 0.5773 | 4.0 | 176 | 3.9009 |
55
+ | 0.3129 | 5.0 | 220 | 4.7521 |
56
+ | 0.1807 | 6.0 | 264 | 5.5262 |
57
+ | 0.1189 | 7.0 | 308 | 6.3706 |
58
+ | 0.0677 | 8.0 | 352 | 7.2028 |
59
+ | 0.0187 | 9.0 | 396 | 8.0377 |
60
+ | 0.0002 | 10.0 | 440 | 8.0574 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
61
 
62
 
63
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cd5695b68adb87e644391d562943dc4c3aac2c01922e0e6e49edb466fbb03610
3
  size 44381360
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7602a2477a718bb31b4e354ee424902a570d892309b22ee40f08cc8e86780c6
3
  size 44381360
runs/Mar02_19-03-38_Software-AI/events.out.tfevents.1709393618.Software-AI.552873.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6feb51b75f0009d1cea1aca1dc0f6c6964011036ce032faa036697b2fe5264b1
3
+ size 9223
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:58c30cfcc34b83de23a7420b9228d774694bbbc5eaa0715753618de28f2ad9c4
3
- size 4155
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9bd11654c41a1db603a33d8ffb96f93020077a85d4a7d67160ab8ea70525ef1c
3
+ size 4219