End of training

README.md

@@ -3,6 +3,8 @@ license: apache-2.0
 base_model: GuysTrans/bart-base-re-attention-seq-512
 tags:
 - generated_from_trainer
+metrics:
+- rouge
 model-index:
 - name: bart-base-re-attention-seq-512
   results: []
@@ -14,6 +16,17 @@ should probably proofread and complete it, then remove this comment. -->
 # bart-base-re-attention-seq-512
 
 This model is a fine-tuned version of [GuysTrans/bart-base-re-attention-seq-512](https://huggingface.co/GuysTrans/bart-base-re-attention-seq-512) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.1000
+- Rouge1: 36.7782
+- Rouge2: 29.3131
+- Rougel: 35.3321
+- Rougelsum: 36.3224
+- Bleu-1: 12.6541
+- Bleu-2: 11.3004
+- Bleu-3: 10.652
+- Bleu-4: 10.2237
+- Gen Len: 25.8608
 
 ## Model description
 
@@ -40,6 +53,13 @@ The following hyperparameters were used during training:
 - lr_scheduler_type: linear
 - num_epochs: 1
 
+### Training results
+
+| Training Loss | Epoch | Step  | Validation Loss | Rouge1  | Rouge2  | Rougel  | Rougelsum | Bleu-1  | Bleu-2  | Bleu-3 | Bleu-4  | Gen Len |
+|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|:-------:|:------:|:-------:|:-------:|
+| 2.2488        | 1.0   | 18247 | 1.1000          | 36.7782 | 29.3131 | 35.3321 | 36.3224   | 12.6541 | 11.3004 | 10.652 | 10.2237 | 25.8608 |
+
+
 ### Framework versions
 
 - Transformers 4.33.0

generation_config.json

@@ -5,7 +5,7 @@
   "eos_token_id": 2,
   "forced_bos_token_id": 0,
   "forced_eos_token_id": 2,
-  "max_length": 512,
+  "max_length": 26,
   "no_repeat_ngram_size": 3,
   "num_beams": 4,
   "pad_token_id": 1,

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:504806432f245404ba5709c65edd3b45ab2b9a2f309193f2558dff10b886452d
+oid sha256:aa70b161c9fcd104e2d17985d3a3636e3fcc4a5db89ea0d2a65a8d08bd154e04
 size 558018637