bart_summarisation

README.md

@@ -17,12 +17,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [MeetK/bart_summarisation](https://huggingface.co/MeetK/bart_summarisation) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.8662
-- Rouge1: 0.3202
-- Rouge2: 0.1186
-- Rougel: 0.2291
-- Rougelsum: 0.229
-- Gen Len: 63.321
+- Loss: 1.7168
+- Rouge1: 0.3288
+- Rouge2: 0.1335
+- Rougel: 0.2493
+- Rougelsum: 0.2495
+- Gen Len: 61.398
 
 ## Model description
 
@@ -53,8 +53,8 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
 |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:-------:|
-| No log        | 1.0   | 313  | 1.8355          | 0.3173 | 0.1175 | 0.2289 | 0.2288    | 63.492  |
-| 1.5858        | 2.0   | 626  | 1.8662          | 0.3202 | 0.1186 | 0.2291 | 0.229     | 63.321  |
+| No log        | 1.0   | 313  | 1.6942          | 0.3296 | 0.1385 | 0.2507 | 0.2507    | 60.888  |
+| 1.4988        | 2.0   | 626  | 1.7168          | 0.3288 | 0.1335 | 0.2493 | 0.2495    | 61.398  |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4e7b76853e99acedf362dac51bc5c4b244e0ef783f43d3c742f2701b6dcf1923
+oid sha256:da8b50311e633c61c1b0e3b6e365c8e5774bb5c77794ad08597f9a1ed0a4a65b
 size 1625426996