RMWeerasinghe
commited on
Commit
•
0a561ea
1
Parent(s):
a169737
Training complete
Browse files
README.md
ADDED
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
library_name: peft
|
4 |
+
tags:
|
5 |
+
- Summarization
|
6 |
+
- generated_from_trainer
|
7 |
+
datasets:
|
8 |
+
- cnn_dailymail
|
9 |
+
metrics:
|
10 |
+
- rouge
|
11 |
+
base_model: google/flan-t5-base
|
12 |
+
model-index:
|
13 |
+
- name: flan-t5-base-finetuned-QLoRA-10000
|
14 |
+
results: []
|
15 |
+
---
|
16 |
+
|
17 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
18 |
+
should probably proofread and complete it, then remove this comment. -->
|
19 |
+
|
20 |
+
# flan-t5-base-finetuned-QLoRA-10000
|
21 |
+
|
22 |
+
This model is a fine-tuned version of [google/flan-t5-base](https://huggingface.co/google/flan-t5-base) on the cnn_dailymail dataset.
|
23 |
+
It achieves the following results on the evaluation set:
|
24 |
+
- Loss: 1.0625
|
25 |
+
- Rouge1: 0.2397
|
26 |
+
- Rouge2: 0.1107
|
27 |
+
- Rougel: 0.1948
|
28 |
+
- Rougelsum: 0.226
|
29 |
+
|
30 |
+
## Model description
|
31 |
+
|
32 |
+
More information needed
|
33 |
+
|
34 |
+
## Intended uses & limitations
|
35 |
+
|
36 |
+
More information needed
|
37 |
+
|
38 |
+
## Training and evaluation data
|
39 |
+
|
40 |
+
More information needed
|
41 |
+
|
42 |
+
## Training procedure
|
43 |
+
|
44 |
+
### Training hyperparameters
|
45 |
+
|
46 |
+
The following hyperparameters were used during training:
|
47 |
+
- learning_rate: 3e-05
|
48 |
+
- train_batch_size: 8
|
49 |
+
- eval_batch_size: 8
|
50 |
+
- seed: 42
|
51 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
52 |
+
- lr_scheduler_type: linear
|
53 |
+
- num_epochs: 10
|
54 |
+
|
55 |
+
### Training results
|
56 |
+
|
57 |
+
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|
58 |
+
|:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|:------:|:---------:|
|
59 |
+
| 1.3695 | 1.0 | 1250 | 1.1374 | 0.232 | 0.1046 | 0.1884 | 0.218 |
|
60 |
+
| 1.197 | 2.0 | 2500 | 1.0885 | 0.2371 | 0.1093 | 0.1934 | 0.2236 |
|
61 |
+
| 1.1489 | 3.0 | 3750 | 1.0765 | 0.2389 | 0.1098 | 0.1939 | 0.2248 |
|
62 |
+
| 1.156 | 4.0 | 5000 | 1.0693 | 0.2403 | 0.1107 | 0.195 | 0.226 |
|
63 |
+
| 1.1135 | 5.0 | 6250 | 1.0663 | 0.2393 | 0.1102 | 0.1944 | 0.2252 |
|
64 |
+
| 1.1607 | 6.0 | 7500 | 1.0648 | 0.24 | 0.1109 | 0.1951 | 0.2259 |
|
65 |
+
| 1.1222 | 7.0 | 8750 | 1.0635 | 0.2398 | 0.1106 | 0.1947 | 0.2256 |
|
66 |
+
| 1.1619 | 8.0 | 10000 | 1.0629 | 0.2399 | 0.1106 | 0.1949 | 0.2259 |
|
67 |
+
| 1.1366 | 9.0 | 11250 | 1.0626 | 0.2397 | 0.1108 | 0.1948 | 0.226 |
|
68 |
+
| 1.2062 | 10.0 | 12500 | 1.0625 | 0.2397 | 0.1107 | 0.1948 | 0.226 |
|
69 |
+
|
70 |
+
|
71 |
+
### Framework versions
|
72 |
+
|
73 |
+
- PEFT 0.8.2
|
74 |
+
- Transformers 4.37.0
|
75 |
+
- Pytorch 2.1.2
|
76 |
+
- Datasets 2.1.0
|
77 |
+
- Tokenizers 0.15.1
|