Hugging Face's logo

nlparabic
/
res_nw_eg

Safetensors
gpt2
Generated from Trainer
Model card Files Community
nlparabic commited on
Commit
45df639
·
verified ·
1 Parent(s): 13f7733

Model save

Browse files
Files changed (2) hide show
  1. README.md +15 -21
  2. model.safetensors +1 -1
README.md CHANGED
@@ -18,11 +18,11 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [riotu-lab/ArabianGPT-01B](https://huggingface.co/riotu-lab/ArabianGPT-01B) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.7316
22
- - Bleu: 0.2459
23
- - Rouge1: 0.6017
24
- - Rouge2: 0.3519
25
- - Rougel: 0.5983
26
 
27
  ## Model description
28
 
@@ -52,22 +52,16 @@ The following hyperparameters were used during training:
52
 
53
  ### Training results
54
 
55
- | Training Loss | Epoch | Step | Bleu | Validation Loss | Rouge1 | Rouge2 | Rougel |
56
- |:-------------:|:-----:|:-----:|:------:|:---------------:|:------:|:------:|:------:|
57
- | 1.1436 | 1.0 | 7107 | 0.1900 | 0.8277 | 0.5212 | 0.2576 | 0.5169 |
58
- | 0.7508 | 2.0 | 14214 | 0.2214 | 0.7543 | 0.5674 | 0.3108 | 0.5636 |
59
- | 0.6471 | 3.0 | 21321 | 0.2375 | 0.7338 | 0.5881 | 0.3356 | 0.5845 |
60
- | 0.5713 | 4.0 | 28428 | 0.2459 | 0.7316 | 0.6017 | 0.3519 | 0.5983 |
61
- | 0.5097 | 5.0 | 35535 | 0.2475 | 0.7390 | 0.6058 | 0.3572 | 0.6022 |
62
- | 0.4573 | 6.0 | 42642 | 0.2503 | 0.7483 | 0.6103 | 0.3618 | 0.6066 |
63
- | 0.4118 | 7.0 | 49749 | 0.2494 | 0.7636 | 0.6106 | 0.3634 | 0.6070 |
64
- | 0.3725 | 8.0 | 56856 | 0.2507 | 0.7796 | 0.6127 | 0.3660 | 0.6089 |
65
- | 0.3375 | 9.0 | 63963 | 0.2491 | 0.7974 | 0.6112 | 0.3654 | 0.6074 |
66
- | 0.3074 | 10.0 | 71070 | 0.8156 | 0.2478 | 0.6101 | 0.3644 | 0.6063 |
67
- | 0.2813 | 11.0 | 78177 | 0.8326 | 0.2488 | 0.6109 | 0.3657 | 0.6072 |
68
- | 0.2586 | 12.0 | 85284 | 0.8498 | 0.2461 | 0.6082 | 0.3630 | 0.6045 |
69
- | 0.2392 | 13.0 | 92391 | 0.8675 | 0.2470 | 0.6095 | 0.3642 | 0.6055 |
70
- | 0.2227 | 14.0 | 99498 | 0.8827 | 0.2463 | 0.6087 | 0.3631 | 0.6049 |
71
 
72
 
73
  ### Framework versions
 
18
 
19
  This model is a fine-tuned version of [riotu-lab/ArabianGPT-01B](https://huggingface.co/riotu-lab/ArabianGPT-01B) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 0.9003
22
+ - Bleu: 0.3478
23
+ - Rouge1: 0.6166
24
+ - Rouge2: 0.3549
25
+ - Rougel: 0.6125
26
 
27
  ## Model description
28
 
 
52
 
53
  ### Training results
54
 
55
+ | Training Loss | Epoch | Step | Validation Loss | Bleu | Rouge1 | Rouge2 | Rougel |
56
+ |:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|:------:|:------:|
57
+ | 1.1362 | 1.0 | 7105 | 0.9123 | 0.3024 | 0.5276 | 0.2508 | 0.5228 |
58
+ | 0.7501 | 2.0 | 14210 | 0.8436 | 0.3299 | 0.5754 | 0.3055 | 0.5713 |
59
+ | 0.6462 | 3.0 | 21315 | 0.8298 | 0.3390 | 0.5962 | 0.3298 | 0.5921 |
60
+ | 0.5705 | 4.0 | 28420 | 0.8328 | 0.3444 | 0.6048 | 0.3412 | 0.6011 |
61
+ | 0.5087 | 5.0 | 35525 | 0.8406 | 0.3480 | 0.6133 | 0.3512 | 0.6096 |
62
+ | 0.4559 | 6.0 | 42630 | 0.8647 | 0.3478 | 0.6133 | 0.3520 | 0.6094 |
63
+ | 0.4104 | 7.0 | 49735 | 0.8770 | 0.3485 | 0.6148 | 0.3533 | 0.6107 |
64
+ | 0.3708 | 8.0 | 56840 | 0.9003 | 0.3478 | 0.6166 | 0.3549 | 0.6125 |
 
 
 
 
 
 
65
 
66
 
67
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3a3d3de566f031d48da768494183e44f928255878ae190607dd03e5674a706b4
3
  size 539224704
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5931170ee7acc1eee5450c7f9b2e6fcf08eed6838b83fae3cd7327de6596a2c2
3
  size 539224704