Hugging Face's logo

mishikaa16
/
t5-base-finetuned-arxiver

Summarization
Transformers
Safetensors
t5
text2text-generation
Generated from Trainer
text-generation-inference
Model card Files Community
mishikaa16 commited on
Commit
7503b86
·
verified ·
1 Parent(s): cb8d0ad

Training complete

Browse files
Files changed (4) hide show
  1. README.md +18 -18
  2. config.json +1 -1
  3. model.safetensors +1 -1
  4. training_args.bin +1 -1
README.md CHANGED
@@ -15,18 +15,17 @@ model-index:
15
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
16
  should probably proofread and complete it, then remove this comment. -->
17
 
18
- [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/mishikaj162001-cisco/insightSumm_t5_base_model_summarizer/runs/ltclgb9d)
19
- [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/mishikaj162001-cisco/insightSumm_t5_base_model_summarizer/runs/of33hrc0)
20
- [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/mishikaj162001-cisco/insightSumm_summarizer_t5__model/runs/jaeceva0)
21
  # t5-base-finetuned-arxiver
22
 
23
  This model is a fine-tuned version of [t5-base](https://huggingface.co/t5-base) on the None dataset.
24
  It achieves the following results on the evaluation set:
25
- - Loss: 0.2849
26
- - Rouge1: 88.8805
27
- - Rouge2: 86.2718
28
- - Rougel: 88.6307
29
- - Rougelsum: 88.6258
30
 
31
  ## Model description
32
 
@@ -45,24 +44,25 @@ More information needed
45
  ### Training hyperparameters
46
 
47
  The following hyperparameters were used during training:
48
- - learning_rate: 5e-05
49
  - train_batch_size: 16
50
  - eval_batch_size: 16
51
  - seed: 42
 
 
52
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
53
- - lr_scheduler_type: linear
54
- - num_epochs: 5
55
  - mixed_precision_training: Native AMP
56
 
57
  ### Training results
58
 
59
- | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
60
- |:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|:-------:|:---------:|
61
- | 0.197 | 1.0 | 3168 | 0.3226 | 88.3265 | 85.5982 | 88.0812 | 88.0737 |
62
- | 0.1111 | 2.0 | 6336 | 0.2777 | 88.7534 | 86.1654 | 88.5084 | 88.4973 |
63
- | 0.3472 | 3.0 | 9504 | 0.2804 | 88.871 | 86.268 | 88.6179 | 88.6167 |
64
- | 0.1489 | 4.0 | 12672 | 0.2849 | 88.8805 | 86.2718 | 88.6307 | 88.6258 |
65
- | 0.2105 | 5.0 | 15840 | 0.2887 | 88.8131 | 86.2233 | 88.5704 | 88.5621 |
66
 
67
 
68
  ### Framework versions
 
15
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
16
  should probably proofread and complete it, then remove this comment. -->
17
 
18
+ [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/mishikaj162001-cisco/insightSumm_t5_model_finetuned_model_text_summarizer/runs/k7z2e3j7)
19
+ [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/mishikaj162001-cisco/insightSumm_text_summarizer_t5_model_finetuned_model/runs/8kyi43ik)
 
20
  # t5-base-finetuned-arxiver
21
 
22
  This model is a fine-tuned version of [t5-base](https://huggingface.co/t5-base) on the None dataset.
23
  It achieves the following results on the evaluation set:
24
+ - Loss: 0.4141
25
+ - Rouge1: 88.4527
26
+ - Rouge2: 85.9716
27
+ - Rougel: 88.2515
28
+ - Rougelsum: 88.2522
29
 
30
  ## Model description
31
 
 
44
  ### Training hyperparameters
45
 
46
  The following hyperparameters were used during training:
47
+ - learning_rate: 1e-05
48
  - train_batch_size: 16
49
  - eval_batch_size: 16
50
  - seed: 42
51
+ - gradient_accumulation_steps: 2
52
+ - total_train_batch_size: 32
53
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
54
+ - lr_scheduler_type: cosine_with_restarts
55
+ - num_epochs: 7
56
  - mixed_precision_training: Native AMP
57
 
58
  ### Training results
59
 
60
+ | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
61
+ |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|
62
+ | 0.3076 | 1.0 | 1584 | 0.4287 | 88.3592 | 85.8872 | 88.1554 | 88.1659 |
63
+ | 0.2896 | 2.0 | 3168 | 0.4141 | 88.4527 | 85.9716 | 88.2515 | 88.2522 |
64
+ | 0.5806 | 3.0 | 4752 | 0.4071 | 88.4403 | 85.9628 | 88.2306 | 88.2359 |
65
+ | 0.5059 | 4.0 | 6336 | 0.3993 | 88.4437 | 85.9587 | 88.2291 | 88.2381 |
 
66
 
67
 
68
  ### Framework versions
config.json CHANGED
@@ -9,7 +9,7 @@
9
  "d_model": 768,
10
  "decoder_start_token_id": 0,
11
  "dense_act_fn": "relu",
12
- "dropout_rate": 0.1,
13
  "eos_token_id": 1,
14
  "feed_forward_proj": "relu",
15
  "initializer_factor": 1.0,
 
9
  "d_model": 768,
10
  "decoder_start_token_id": 0,
11
  "dense_act_fn": "relu",
12
+ "dropout_rate": 0.3,
13
  "eos_token_id": 1,
14
  "feed_forward_proj": "relu",
15
  "initializer_factor": 1.0,
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:489aa329283a81ae7681f9b81cd1c124e6e0046858a5c7c74ac49d0181d73a5a
3
  size 891644712
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a45affd4b98235627e070620ed9b9923820da35898c0bf2973d679bef8c4b797
3
  size 891644712
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:71677e63adcf2e3581b4de397789436c1cdef6eb323116f0429e709b5df0d395
3
  size 5905
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:901e0d67ac44453a304e323125aab43caf760d4a6b6f5822a19ab2aaf0f534e7
3
  size 5905