fine_tuned_t5_small_model_sec_5_v13

README.md

@@ -18,13 +18,13 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.9766
-- Rouge1: 0.3717
-- Rouge2: 0.1454
-- Rougel: 0.2451
-- Rougelsum: 0.2447
-- Gen Len: 167.6053
-- Bert F1: 0.86
+- Loss: 2.9971
+- Rouge1: 0.4057
+- Rouge2: 0.155
+- Rougel: 0.2516
+- Rougelsum: 0.252
+- Gen Len: 95.1
+- Bert F1: 0.8758
 
 ## Model description
 
@@ -54,12 +54,12 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len  | Bert F1 |
-|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:--------:|:-------:|
-| 3.5821        | 1.0   | 95   | 3.1263          | 0.3818 | 0.1448 | 0.2426 | 0.2422    | 146.6684 | 0.8607  |
-| 3.1907        | 2.0   | 190  | 3.0156          | 0.3772 | 0.144  | 0.2436 | 0.2433    | 156.6421 | 0.8616  |
-| 3.1261        | 3.0   | 285  | 2.9852          | 0.3717 | 0.143  | 0.2425 | 0.2423    | 165.4684 | 0.86    |
-| 3.1135        | 4.0   | 380  | 2.9766          | 0.3717 | 0.1454 | 0.2451 | 0.2447    | 167.6053 | 0.86    |
+| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len | Bert F1 |
+|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:-------:|:-------:|
+| 3.5508        | 1.0   | 95   | 3.1502          | 0.4016 | 0.1522 | 0.2479 | 0.2476    | 97.6526 | 0.874   |
+| 3.1904        | 2.0   | 190  | 3.0374          | 0.4094 | 0.1578 | 0.2536 | 0.2536    | 97.6474 | 0.8757  |
+| 3.138         | 3.0   | 285  | 3.0059          | 0.4034 | 0.1538 | 0.2486 | 0.2491    | 95.0211 | 0.8752  |
+| 3.1061        | 4.0   | 380  | 2.9971          | 0.4057 | 0.155  | 0.2516 | 0.252     | 95.1    | 0.8758  |
 
 
 ### Framework versions

generation_config.json

@@ -1,8 +1,8 @@
 {
   "decoder_start_token_id": 0,
   "eos_token_id": 1,
-  "max_length": 250,
-  "min_length": 200,
+  "max_length": 150,
+  "min_length": 50,
   "num_beams": 4,
   "pad_token_id": 0,
   "temperature": 3,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fc921c33e2a7cfbb62cbe0773f737f26c5cf7d2cae5406744bf09846e0ca6934
+oid sha256:731c8c6d9dde1fc5820f8b52c120591d129696d9de774e2841b53b59e6ba9b94
 size 242041896