Commit ·
5411ec6
1
Parent(s): ba435b6
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: base-model-with-warmup-fulldata-fairseq-V1
|
| 7 |
+
results: []
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 11 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 12 |
+
|
| 13 |
+
# base-model-with-warmup-fulldata-fairseq-V1
|
| 14 |
+
|
| 15 |
+
This model is a fine-tuned version of [facebook/wav2vec2-base-960h](https://huggingface.co/facebook/wav2vec2-base-960h) on the None dataset.
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 4.6583
|
| 18 |
+
|
| 19 |
+
## Model description
|
| 20 |
+
|
| 21 |
+
More information needed
|
| 22 |
+
|
| 23 |
+
## Intended uses & limitations
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Training and evaluation data
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training procedure
|
| 32 |
+
|
| 33 |
+
### Training hyperparameters
|
| 34 |
+
|
| 35 |
+
The following hyperparameters were used during training:
|
| 36 |
+
- learning_rate: 0.0001
|
| 37 |
+
- train_batch_size: 8
|
| 38 |
+
- eval_batch_size: 16
|
| 39 |
+
- seed: 42
|
| 40 |
+
- gradient_accumulation_steps: 2
|
| 41 |
+
- total_train_batch_size: 16
|
| 42 |
+
- optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-08
|
| 43 |
+
- lr_scheduler_type: linear
|
| 44 |
+
- lr_scheduler_warmup_steps: 200
|
| 45 |
+
- num_epochs: 50
|
| 46 |
+
|
| 47 |
+
### Training results
|
| 48 |
+
|
| 49 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 50 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 51 |
+
| 15.6587 | 2.05 | 200 | 4.4816 |
|
| 52 |
+
| 3.8464 | 4.1 | 400 | 4.7516 |
|
| 53 |
+
| 3.8377 | 6.15 | 600 | 4.4104 |
|
| 54 |
+
| 3.915 | 8.21 | 800 | 4.6832 |
|
| 55 |
+
| 3.8094 | 10.26 | 1000 | 5.1651 |
|
| 56 |
+
| 3.8095 | 12.31 | 1200 | 4.7847 |
|
| 57 |
+
| 3.8427 | 14.36 | 1400 | 4.6613 |
|
| 58 |
+
| 3.7917 | 16.41 | 1600 | 4.7898 |
|
| 59 |
+
| 3.804 | 18.46 | 1800 | 4.7368 |
|
| 60 |
+
| 3.8492 | 20.51 | 2000 | 4.7656 |
|
| 61 |
+
| 3.7943 | 22.56 | 2200 | 4.8043 |
|
| 62 |
+
| 3.7982 | 24.62 | 2400 | 4.5903 |
|
| 63 |
+
| 3.7979 | 26.67 | 2600 | 4.5546 |
|
| 64 |
+
| 3.7938 | 28.72 | 2800 | 4.6392 |
|
| 65 |
+
| 3.7891 | 30.77 | 3000 | 4.8080 |
|
| 66 |
+
| 3.7936 | 32.82 | 3200 | 4.5811 |
|
| 67 |
+
| 3.7926 | 34.87 | 3400 | 4.5891 |
|
| 68 |
+
| 3.7922 | 36.92 | 3600 | 4.6251 |
|
| 69 |
+
| 3.79 | 38.97 | 3800 | 4.7409 |
|
| 70 |
+
| 3.791 | 41.03 | 4000 | 4.6539 |
|
| 71 |
+
| 3.7893 | 43.08 | 4200 | 4.6752 |
|
| 72 |
+
| 3.7882 | 45.13 | 4400 | 4.6751 |
|
| 73 |
+
| 3.7924 | 47.18 | 4600 | 4.6222 |
|
| 74 |
+
| 3.7991 | 49.23 | 4800 | 4.6583 |
|
| 75 |
+
|
| 76 |
+
|
| 77 |
+
### Framework versions
|
| 78 |
+
|
| 79 |
+
- Transformers 4.28.0
|
| 80 |
+
- Pytorch 2.0.1
|
| 81 |
+
- Datasets 2.12.0
|
| 82 |
+
- Tokenizers 0.13.2
|