Update README.md

README.md

@@ -8,6 +8,10 @@ tags:
 model-index:
 - name: AmsterdamDocClassificationGEITje200T3Epochs
  results: []
+datasets:
+- FemkeBakker/AmsterdamBalancedFirst200Tokens
+language:
+- nl
 ---
 
 <!-- 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. -->
 
 # AmsterdamDocClassificationGEITje200T3Epochs
 
-This model is a fine-tuned version of [Rijgersberg/GEITje-7B-chat-v2](https://huggingface.co/Rijgersberg/GEITje-7B-chat-v2) on the [AmsterdamDocClassification](https://huggingface.co/datasets/FemkeBakker/AmsterdamBalancedFirst200Tokens) dataset.
+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. 
+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. 
+In our research, we evaluate the fine-tuning of these LLMs across one, two, and three epochs. 
+This model is a fine-tuned version of [Rijgersberg/GEITje-7B-chat-v2](https://huggingface.co/Rijgersberg/GEITje-7B-chat-v2) and has been fine-tuned for three epochs.
+
 It achieves the following results on the evaluation set:
 - Loss: 0.5854
 
-## Model description
-
-More information needed
-
-## Intended uses & limitations
-
-More information needed
 
 ## Training and evaluation data
 
-More information needed
+- The training data consists of 9900 documents and their labels formatted into conversations. 
+- The evaluation data consists of 1100 documents and their labels formatted into conversations. 
 
 ## Training procedure
 
+See the [GitHub](https://github.com/Amsterdam-Internships/document-classification-using-large-language-models) for specifics about the training and the code.
+
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
@@ -67,6 +71,7 @@ The following hyperparameters were used during training:
 | 0.6234 | 2.7855 | 1722 | 0.5853 |
 | 0.653 | 2.9842 | 1845 | 0.5854 |
 
+Training time: it took in total 2 hours and 26 minutes to fine-tune the model for three epochs.
 
 ### Framework versions
 
@@ -74,3 +79,7 @@ The following hyperparameters were used during training:
 - Pytorch 2.3.0+cu121
 - Datasets 2.19.1
 - Tokenizers 0.19.1
+
+
+### Acknowledgements
+This model was trained as part of [insert thesis info] in collaboration with Amsterdam Intelligence for the City of Amsterdam.