Update README.md
Browse files
README.md
CHANGED
|
@@ -42,9 +42,7 @@ We use the pretrained [`sentence-transformers/all-mpnet-base-v2`](https://huggin
|
|
| 42 |
|
| 43 |
We use the initial model without modifying its architecture or pre-trained model parameters.
|
| 44 |
However, we reduce the processed sequence length to 384 to reduce the training time of the model.
|
| 45 |
-
|
| 46 |
-
We use constant learning rate of 1e-5, and we train the model for ten epochs.
|
| 47 |
-
|
| 48 |
### Dataset
|
| 49 |
|
| 50 |
We finetuned our model on the combination of the Goodbooks-10k and the MovieLens20M datasets with item descriptions generated with [`meta-llama/Meta-Llama-3.1-8B-Instruct`](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct) model. For details please see the dataset pages: [`beeformer/recsys-movielens-20m`](https://huggingface.co/datasets/beeformer/recsys-movielens-20m) and [`beeformer/recsys-goodbooks-10k`](https://huggingface.co/datasets/beeformer/recsys-goodbooks-10k).
|
|
|
|
| 42 |
|
| 43 |
We use the initial model without modifying its architecture or pre-trained model parameters.
|
| 44 |
However, we reduce the processed sequence length to 384 to reduce the training time of the model.
|
| 45 |
+
|
|
|
|
|
|
|
| 46 |
### Dataset
|
| 47 |
|
| 48 |
We finetuned our model on the combination of the Goodbooks-10k and the MovieLens20M datasets with item descriptions generated with [`meta-llama/Meta-Llama-3.1-8B-Instruct`](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct) model. For details please see the dataset pages: [`beeformer/recsys-movielens-20m`](https://huggingface.co/datasets/beeformer/recsys-movielens-20m) and [`beeformer/recsys-goodbooks-10k`](https://huggingface.co/datasets/beeformer/recsys-goodbooks-10k).
|