KMayanja commited on
Commit
b17bde3
·
verified ·
1 Parent(s): 2055003

End of training

Browse files
Files changed (2) hide show
  1. README.md +79 -0
  2. adapter_model.safetensors +1 -1
README.md ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ license: cc-by-nc-4.0
4
+ base_model: facebook/nllb-200-distilled-600M
5
+ tags:
6
+ - base_model:adapter:facebook/nllb-200-distilled-600M
7
+ - lora
8
+ - transformers
9
+ metrics:
10
+ - bleu
11
+ model-index:
12
+ - name: nllb-600M-medical-luganda-bidirectional
13
+ results: []
14
+ ---
15
+
16
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
+ should probably proofread and complete it, then remove this comment. -->
18
+
19
+ # nllb-600M-medical-luganda-bidirectional
20
+
21
+ This model is a fine-tuned version of [facebook/nllb-200-distilled-600M](https://huggingface.co/facebook/nllb-200-distilled-600M) on an unknown dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 2.5663
24
+ - Bleu: 6.6618
25
+ - Chrf: 22.9805
26
+
27
+ ## Model description
28
+
29
+ More information needed
30
+
31
+ ## Intended uses & limitations
32
+
33
+ More information needed
34
+
35
+ ## Training and evaluation data
36
+
37
+ More information needed
38
+
39
+ ## Training procedure
40
+
41
+ ### Training hyperparameters
42
+
43
+ The following hyperparameters were used during training:
44
+ - learning_rate: 3e-05
45
+ - train_batch_size: 4
46
+ - eval_batch_size: 8
47
+ - seed: 42
48
+ - gradient_accumulation_steps: 8
49
+ - total_train_batch_size: 32
50
+ - optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
51
+ - lr_scheduler_type: linear
52
+ - lr_scheduler_warmup_steps: 100
53
+ - num_epochs: 12
54
+
55
+ ### Training results
56
+
57
+ | Training Loss | Epoch | Step | Validation Loss | Bleu | Chrf |
58
+ |:-------------:|:-------:|:----:|:---------------:|:------:|:-------:|
59
+ | 5.65 | 0.9434 | 400 | 5.0490 | 1.802 | 15.6619 |
60
+ | 4.5322 | 1.8868 | 800 | 4.4031 | 1.3741 | 15.75 |
61
+ | 3.8473 | 2.8302 | 1200 | 3.8525 | 1.2316 | 16.2379 |
62
+ | 3.3184 | 3.7736 | 1600 | 3.4441 | 1.2876 | 17.0497 |
63
+ | 3.0563 | 4.7170 | 2000 | 3.2368 | 1.5506 | 17.5027 |
64
+ | 2.7198 | 5.6604 | 2400 | 3.0402 | 2.255 | 18.2851 |
65
+ | 2.5405 | 6.6038 | 2800 | 2.9127 | 3.5638 | 19.6118 |
66
+ | 2.4426 | 7.5472 | 3200 | 2.8274 | 4.2756 | 20.7453 |
67
+ | 2.2637 | 8.4906 | 3600 | 2.6992 | 5.5678 | 21.8649 |
68
+ | 2.2013 | 9.4340 | 4000 | 2.6395 | 6.2028 | 22.4425 |
69
+ | 2.1493 | 10.3774 | 4400 | 2.6026 | 6.5708 | 22.799 |
70
+ | 2.1372 | 11.3208 | 4800 | 2.5663 | 6.6618 | 22.9805 |
71
+
72
+
73
+ ### Framework versions
74
+
75
+ - PEFT 0.17.1
76
+ - Transformers 4.56.2
77
+ - Pytorch 2.8.0
78
+ - Datasets 4.1.1
79
+ - Tokenizers 0.22.1
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ee25538ae6a46f659306d78ec5746198720046b40921ae9c12a56e8b2969f1e2
3
  size 28342344
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce5f1d3ad7c700a177d0bee2a48f90b0fda0a96728d564bf2dbc8d8810905968
3
  size 28342344