FemkeBakker
commited on
Commit
•
2c40a1e
1
Parent(s):
bcbf876
Update README.md
Browse files
README.md
CHANGED
@@ -8,6 +8,10 @@ tags:
|
|
8 |
model-index:
|
9 |
- name: AmsterdamDocClassificationLlama200T1Epochs
|
10 |
results: []
|
|
|
|
|
|
|
|
|
11 |
---
|
12 |
|
13 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -15,24 +19,24 @@ should probably proofread and complete it, then remove this comment. -->
|
|
15 |
|
16 |
# AmsterdamDocClassificationLlama200T1Epochs
|
17 |
|
18 |
-
|
|
|
|
|
|
|
|
|
19 |
It achieves the following results on the evaluation set:
|
20 |
- Loss: 0.8403
|
21 |
|
22 |
-
## Model description
|
23 |
-
|
24 |
-
More information needed
|
25 |
-
|
26 |
-
## Intended uses & limitations
|
27 |
-
|
28 |
-
More information needed
|
29 |
|
30 |
## Training and evaluation data
|
31 |
|
32 |
-
|
|
|
33 |
|
34 |
## Training procedure
|
35 |
|
|
|
|
|
36 |
### Training hyperparameters
|
37 |
|
38 |
The following hyperparameters were used during training:
|
@@ -57,6 +61,7 @@ The following hyperparameters were used during training:
|
|
57 |
| 0.7408 | 0.7952 | 492 | 0.8413 |
|
58 |
| 0.996 | 0.9939 | 615 | 0.8403 |
|
59 |
|
|
|
60 |
|
61 |
### Framework versions
|
62 |
|
@@ -64,3 +69,7 @@ The following hyperparameters were used during training:
|
|
64 |
- Pytorch 2.3.0+cu121
|
65 |
- Datasets 2.19.1
|
66 |
- Tokenizers 0.19.1
|
|
|
|
|
|
|
|
|
|
8 |
model-index:
|
9 |
- name: AmsterdamDocClassificationLlama200T1Epochs
|
10 |
results: []
|
11 |
+
datasets:
|
12 |
+
- FemkeBakker/AmsterdamBalancedFirst200Tokens
|
13 |
+
language:
|
14 |
+
- nl
|
15 |
---
|
16 |
|
17 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
19 |
|
20 |
# AmsterdamDocClassificationLlama200T1Epochs
|
21 |
|
22 |
+
As part of the Assessing Large Language Models for Document Classification project by the Municipality of Amsterdam, we fine-tune Mistral, Llama, and GEITje for document classification.
|
23 |
+
The fine-tuning is performed using the [AmsterdamBalancedFirst200Tokens](https://huggingface.co/datasets/FemkeBakker/AmsterdamBalancedFirst200Tokens) dataset, which consists of documents truncated to the first 200 tokens.
|
24 |
+
In our research, we evaluate the fine-tuning of these LLMs across one, two, and three epochs.
|
25 |
+
This model is a fine-tuned version of [meta-llama/Llama-2-7b-chat-hf](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf) and has been fine-tuned for one epoch.
|
26 |
+
|
27 |
It achieves the following results on the evaluation set:
|
28 |
- Loss: 0.8403
|
29 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
30 |
|
31 |
## Training and evaluation data
|
32 |
|
33 |
+
- The training data consists of 9900 documents and their labels formatted into conversations.
|
34 |
+
- The evaluation data consists of 1100 documents and their labels formatted into conversations.
|
35 |
|
36 |
## Training procedure
|
37 |
|
38 |
+
See the [GitHub](https://github.com/Amsterdam-Internships/document-classification-using-large-language-models) for specifics about the training and the code.
|
39 |
+
|
40 |
### Training hyperparameters
|
41 |
|
42 |
The following hyperparameters were used during training:
|
|
|
61 |
| 0.7408 | 0.7952 | 492 | 0.8413 |
|
62 |
| 0.996 | 0.9939 | 615 | 0.8403 |
|
63 |
|
64 |
+
Training time: in total it took 39 minutes to fine-tune the model for one epoch.
|
65 |
|
66 |
### Framework versions
|
67 |
|
|
|
69 |
- Pytorch 2.3.0+cu121
|
70 |
- Datasets 2.19.1
|
71 |
- Tokenizers 0.19.1
|
72 |
+
|
73 |
+
|
74 |
+
### Acknowledgements
|
75 |
+
This model was trained as part of [insert thesis info] in collaboration with Amsterdam Intelligence for the City of Amsterdam.
|