Maxinstellar
commited on
Commit
•
3354437
1
Parent(s):
019bb61
End of training
Browse files
README.md
CHANGED
@@ -2,6 +2,8 @@
|
|
2 |
base_model: UrukHan/t5-russian-spell
|
3 |
tags:
|
4 |
- generated_from_trainer
|
|
|
|
|
5 |
model-index:
|
6 |
- name: t5-russian-spell
|
7 |
results: []
|
@@ -13,6 +15,13 @@ should probably proofread and complete it, then remove this comment. -->
|
|
13 |
# t5-russian-spell
|
14 |
|
15 |
This model is a fine-tuned version of [UrukHan/t5-russian-spell](https://huggingface.co/UrukHan/t5-russian-spell) on the None dataset.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
16 |
|
17 |
## Model description
|
18 |
|
@@ -32,16 +41,19 @@ More information needed
|
|
32 |
|
33 |
The following hyperparameters were used during training:
|
34 |
- learning_rate: 5e-05
|
35 |
-
- train_batch_size:
|
36 |
-
- eval_batch_size:
|
37 |
- seed: 42
|
38 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
39 |
- lr_scheduler_type: linear
|
40 |
-
- num_epochs:
|
41 |
- mixed_precision_training: Native AMP
|
42 |
|
43 |
### Training results
|
44 |
|
|
|
|
|
|
|
45 |
|
46 |
|
47 |
### Framework versions
|
|
|
2 |
base_model: UrukHan/t5-russian-spell
|
3 |
tags:
|
4 |
- generated_from_trainer
|
5 |
+
metrics:
|
6 |
+
- rouge
|
7 |
model-index:
|
8 |
- name: t5-russian-spell
|
9 |
results: []
|
|
|
15 |
# t5-russian-spell
|
16 |
|
17 |
This model is a fine-tuned version of [UrukHan/t5-russian-spell](https://huggingface.co/UrukHan/t5-russian-spell) on the None dataset.
|
18 |
+
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 7.9129
|
20 |
+
- Rouge1: 0.176
|
21 |
+
- Rouge2: 0.0
|
22 |
+
- Rougel: 0.1793
|
23 |
+
- Rougelsum: 0.1772
|
24 |
+
- Gen Len: 198.0413
|
25 |
|
26 |
## Model description
|
27 |
|
|
|
41 |
|
42 |
The following hyperparameters were used during training:
|
43 |
- learning_rate: 5e-05
|
44 |
+
- train_batch_size: 16
|
45 |
+
- eval_batch_size: 16
|
46 |
- seed: 42
|
47 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
48 |
- lr_scheduler_type: linear
|
49 |
+
- num_epochs: 10
|
50 |
- mixed_precision_training: Native AMP
|
51 |
|
52 |
### Training results
|
53 |
|
54 |
+
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|
55 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:--------:|
|
56 |
+
| 7.8172 | 9.01 | 5000 | 7.9129 | 0.176 | 0.0 | 0.1793 | 0.1772 | 198.0413 |
|
57 |
|
58 |
|
59 |
### Framework versions
|
generation_config.json
CHANGED
@@ -1,5 +1,4 @@
|
|
1 |
{
|
2 |
-
"_from_model_config": true,
|
3 |
"decoder_start_token_id": 0,
|
4 |
"eos_token_id": 2,
|
5 |
"max_length": 256,
|
|
|
1 |
{
|
|
|
2 |
"decoder_start_token_id": 0,
|
3 |
"eos_token_id": 2,
|
4 |
"max_length": 256,
|
runs/Dec15_05-39-30_f400652e9e69/events.out.tfevents.1702618771.f400652e9e69.773.0
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:93e1363b8677285108ea0e6d8c38c52fbfb88e9f4be61c10940b046f02c9a396
|
3 |
+
size 7889
|