juuwel commited on
Commit
9ce91f2
·
verified ·
1 Parent(s): 8d8cc6b

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +18 -30
README.md CHANGED
@@ -1,34 +1,22 @@
1
  ---
2
  {}
3
  ---
4
-
5
- # Test-Model
6
-
7
- This model is a fine-tuned version of llama, and is specialized for document classification. It achieves the following results on the evaluation set:
8
- - Accuracy: 0.924647
9
- - Hit rate: 0.977464
10
-
11
- ## Training Parameters
12
- - **Epochs**: 5
13
- - **Batch Size**: 20
14
- - **Threshold**: 0.975
15
- - **Token Length**: 512
16
- - **Data Set Size**: 10000
17
- - **Evenly Distributed**: True
18
- - **Municipalities**: Heroey, Kongsberg, Boemlo, Luster, Maalselv
19
- - **Learning Rate**: 0.0001
20
- - **Weight Decay**: 0.01
21
- - **Eval Accumulation Steps**: 4
22
 
23
- | Label | Accuracy | Accuracy Accumulation | Hit Rate | Hit Rate Accumulation |
24
- |:-----:|:--------:|:---------------------:|:--------:|:---------------------:|
25
- | 0 | 0.93 | 0.93 | 0.98 | 0.98 |
26
- | 1 | 0.95 | 0.95 | 0.97 | 0.97 |
27
- | 2 | 0.91 | 0.91 | 0.98 | 0.98 |
28
- | 3 | 0.40 | 0.40 | 0.57 | 0.57 |
29
- | 4 | 0.25 | 0.25 | 0.63 | 0.63 |
30
- | 5 | 0.86 | 0.86 | 0.94 | 0.94 |
31
- | 6 | 0.99 | 0.99 | 1.04 | 1.04 |
32
- | 7 | 0.94 | 0.94 | 0.98 | 0.98 |
33
- | 8 | 0.67 | 0.67 | 0.86 | 0.86 |
34
- | 9 | 0.99 | 0.99 | 0.99 | 0.99 |
 
 
 
 
 
 
 
1
  ---
2
  {}
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4
 
5
+ # Test-Model
6
+
7
+ This model is a fine-tuned version of llama, and is specialized for document classification. It achieves the following results on the evaluation set:
8
+ - Accuracy: ACCURACY
9
+ - Hit rate: HIT RATE
10
+
11
+ ## Training Parameters
12
+
13
+ - **Epochs**: 5
14
+ - **Batch Size**: 20
15
+ - **Threshold**: 0.975
16
+ - **Token Length**: 512
17
+ - **Data Set Size**: 10000
18
+ - **Evenly Distributed**: True
19
+ - **Municipalities**: Heroey, Kongsberg, Boemlo, Luster, Maalselv
20
+ - **Learning Rate**: 0.0001
21
+ - **Weight Decay**: 0.01
22
+ - **Eval Accumulation Steps**: 4