Multi-task title + summary model

README.md

@@ -0,0 +1,77 @@
+---
+library_name: transformers
+base_model: rosypossyyyy/bart-email-multitask
+tags:
+- generated_from_trainer
+metrics:
+- rouge
+model-index:
+- name: bart-email-multitask
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# bart-email-multitask
+
+This model is a fine-tuned version of [rosypossyyyy/bart-email-multitask](https://huggingface.co/rosypossyyyy/bart-email-multitask) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.5796
+- Rouge1: 50.71
+- Rouge2: 36.15
+- Rougel: 44.19
+- Rougelsum: 44.64
+- Gen Len: 53.1399
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 1e-05
+- train_batch_size: 4
+- eval_batch_size: 4
+- seed: 42
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 16
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 500
+- num_epochs: 2
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
+|:-------------:|:------:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:-------:|
+| 0.3147        | 0.1726 | 500  | 0.7136          | 48.61  | 33.86  | 41.59  | 42.19     | 57.5974 |
+| 0.3772        | 0.3452 | 1000 | 0.6479          | 49.95  | 35.09  | 43.0   | 43.46     | 53.2927 |
+| 0.5262        | 0.5178 | 1500 | 0.5853          | 49.93  | 35.72  | 43.48  | 43.95     | 55.0490 |
+| 0.5042        | 0.6904 | 2000 | 0.5738          | 50.6   | 35.79  | 43.75  | 44.21     | 52.4930 |
+| 0.4835        | 0.8630 | 2500 | 0.5613          | 50.53  | 36.13  | 43.77  | 44.27     | 53.9771 |
+| 0.4328        | 1.0356 | 3000 | 0.5497          | 50.83  | 36.42  | 44.37  | 44.83     | 52.4157 |
+| 0.4392        | 1.2082 | 3500 | 0.5716          | 51.0   | 36.64  | 44.45  | 44.93     | 53.0539 |
+| 0.4162        | 1.3808 | 4000 | 0.5490          | 50.78  | 36.5   | 44.25  | 44.76     | 53.1325 |
+| 0.4235        | 1.5534 | 4500 | 0.5395          | 50.65  | 36.31  | 44.24  | 44.72     | 53.9557 |
+| 0.408         | 1.7260 | 5000 | 0.5433          | 50.67  | 36.51  | 44.2   | 44.7      | 53.9173 |
+
+
+### Framework versions
+
+- Transformers 4.57.3
+- Pytorch 2.9.0+cu126
+- Datasets 4.0.0
+- Tokenizers 0.22.1

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:19dbc6ae6f60269ceec7b39c84cdbdb02208577af03993df03f2a787c6ec6509
+oid sha256:cd2c140601e90ac2911a8ff61fbbf9bbc97882cb1fc16f05c881d8be2c3c45a4
 size 1625422896

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:056454eb981e2c39e27601a08b10d2d51e18e4770004fc3fba9c9098c8de5d70
+oid sha256:62b142a9a91fb2576bf8599b79905966af79a617b77bd632e3db063b59de2677
 size 6033