Hugging Face's logo

YaminiP
/
llama3.2-finetuned-newsclassify

Text Classification
PEFT
Safetensors
English
Generated from Trainer
Model card Files Community
1
YaminiP commited on
Commit
c8349d3
·
verified ·
1 Parent(s): 0ce82e7

Model save

Browse files
Files changed (2) hide show
  1. README.md +54 -18
  2. adapter_model.safetensors +1 -1
README.md CHANGED
@@ -18,21 +18,21 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [meta-llama/Llama-3.2-1B](https://huggingface.co/meta-llama/Llama-3.2-1B) on the None dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.0941
22
- - Balanced Accuracy: 0.984
23
- - Accuracy: 0.984
24
- - F1-score: 0.9839
25
  - Classification-report: precision recall f1-score support
26
 
27
- 0 1.00 0.92 0.96 50
28
  1 1.00 1.00 1.00 50
29
- 2 0.98 1.00 0.99 50
30
  3 1.00 1.00 1.00 50
31
- 4 0.94 1.00 0.97 50
32
 
33
- accuracy 0.98 250
34
- macro avg 0.98 0.98 0.98 250
35
- weighted avg 0.98 0.98 0.98 250
36
 
37
 
38
  ## Model description
@@ -58,30 +58,66 @@ The following hyperparameters were used during training:
58
  - seed: 42
59
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
60
  - lr_scheduler_type: linear
61
- - num_epochs: 1
62
 
63
  ### Training results
64
 
65
  | Training Loss | Epoch | Step | Validation Loss | Balanced Accuracy | Accuracy | F1-score | Classification-report |
66
  |:-------------:|:-----:|:----:|:---------------:|:-----------------:|:--------:|:--------:|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------:|
67
- | 0.001 | 1.0 | 157 | 0.0941 | 0.984 | 0.984 | 0.9839 | precision recall f1-score support
68
 
69
- 0 1.00 0.92 0.96 50
70
  1 1.00 1.00 1.00 50
71
- 2 0.98 1.00 0.99 50
72
  3 1.00 1.00 1.00 50
73
  4 0.94 1.00 0.97 50
74
 
75
- accuracy 0.98 250
76
- macro avg 0.98 0.98 0.98 250
77
- weighted avg 0.98 0.98 0.98 250
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
78
  |
79
 
80
 
81
  ### Framework versions
82
 
83
  - PEFT 0.13.2
84
- - Transformers 4.47.0.dev0
85
  - Pytorch 2.4.1+cu121
86
  - Datasets 3.0.2
87
  - Tokenizers 0.20.1
 
18
 
19
  This model is a fine-tuned version of [meta-llama/Llama-3.2-1B](https://huggingface.co/meta-llama/Llama-3.2-1B) on the None dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 0.0205
22
+ - Balanced Accuracy: 0.992
23
+ - Accuracy: 0.992
24
+ - F1-score: 0.9920
25
  - Classification-report: precision recall f1-score support
26
 
27
+ 0 1.00 0.96 0.98 50
28
  1 1.00 1.00 1.00 50
29
+ 2 1.00 1.00 1.00 50
30
  3 1.00 1.00 1.00 50
31
+ 4 0.96 1.00 0.98 50
32
 
33
+ accuracy 0.99 250
34
+ macro avg 0.99 0.99 0.99 250
35
+ weighted avg 0.99 0.99 0.99 250
36
 
37
 
38
  ## Model description
 
58
  - seed: 42
59
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
60
  - lr_scheduler_type: linear
61
+ - num_epochs: 4
62
 
63
  ### Training results
64
 
65
  | Training Loss | Epoch | Step | Validation Loss | Balanced Accuracy | Accuracy | F1-score | Classification-report |
66
  |:-------------:|:-----:|:----:|:---------------:|:-----------------:|:--------:|:--------:|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------:|
67
+ | 0.0 | 1.0 | 157 | 0.0405 | 0.9880 | 0.988 | 0.9880 | precision recall f1-score support
68
 
69
+ 0 1.00 0.94 0.97 50
70
  1 1.00 1.00 1.00 50
71
+ 2 1.00 1.00 1.00 50
72
  3 1.00 1.00 1.00 50
73
  4 0.94 1.00 0.97 50
74
 
75
+ accuracy 0.99 250
76
+ macro avg 0.99 0.99 0.99 250
77
+ weighted avg 0.99 0.99 0.99 250
78
+ |
79
+ | 0.0 | 2.0 | 314 | 0.0300 | 0.9880 | 0.988 | 0.9880 | precision recall f1-score support
80
+
81
+ 0 1.00 0.94 0.97 50
82
+ 1 1.00 1.00 1.00 50
83
+ 2 1.00 1.00 1.00 50
84
+ 3 1.00 1.00 1.00 50
85
+ 4 0.94 1.00 0.97 50
86
+
87
+ accuracy 0.99 250
88
+ macro avg 0.99 0.99 0.99 250
89
+ weighted avg 0.99 0.99 0.99 250
90
+ |
91
+ | 0.0 | 3.0 | 471 | 0.0177 | 0.992 | 0.992 | 0.9920 | precision recall f1-score support
92
+
93
+ 0 1.00 0.96 0.98 50
94
+ 1 1.00 1.00 1.00 50
95
+ 2 1.00 1.00 1.00 50
96
+ 3 1.00 1.00 1.00 50
97
+ 4 0.96 1.00 0.98 50
98
+
99
+ accuracy 0.99 250
100
+ macro avg 0.99 0.99 0.99 250
101
+ weighted avg 0.99 0.99 0.99 250
102
+ |
103
+ | 0.0 | 4.0 | 628 | 0.0205 | 0.992 | 0.992 | 0.9920 | precision recall f1-score support
104
+
105
+ 0 1.00 0.96 0.98 50
106
+ 1 1.00 1.00 1.00 50
107
+ 2 1.00 1.00 1.00 50
108
+ 3 1.00 1.00 1.00 50
109
+ 4 0.96 1.00 0.98 50
110
+
111
+ accuracy 0.99 250
112
+ macro avg 0.99 0.99 0.99 250
113
+ weighted avg 0.99 0.99 0.99 250
114
  |
115
 
116
 
117
  ### Framework versions
118
 
119
  - PEFT 0.13.2
120
+ - Transformers 4.46.0
121
  - Pytorch 2.4.1+cu121
122
  - Datasets 3.0.2
123
  - Tokenizers 0.20.1
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9080c8e5eba6578af7cd3b4ce84410070b320a962d4877536eb5496c52764821
3
  size 13689552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:720c66677c54ed387c184389ac32ba2862fff1ec1f83a88cac2eaed80561fceb
3
  size 13689552