Commit
•
442d3ee
1
Parent(s):
19a99b2
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- xtreme_s
|
| 7 |
+
metrics:
|
| 8 |
+
- f1
|
| 9 |
+
- accuracy
|
| 10 |
+
model-index:
|
| 11 |
+
- name: xtreme_s_xlsr_300m_minds14.en-US_2
|
| 12 |
+
results: []
|
| 13 |
+
---
|
| 14 |
+
|
| 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 |
+
# xtreme_s_xlsr_300m_minds14.en-US_2
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the xtreme_s dataset.
|
| 21 |
+
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.5685
|
| 23 |
+
- F1: 0.8747
|
| 24 |
+
- Accuracy: 0.8759
|
| 25 |
+
|
| 26 |
+
## Model description
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Intended uses & limitations
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training and evaluation data
|
| 35 |
+
|
| 36 |
+
More information needed
|
| 37 |
+
|
| 38 |
+
## Training procedure
|
| 39 |
+
|
| 40 |
+
### Training hyperparameters
|
| 41 |
+
|
| 42 |
+
The following hyperparameters were used during training:
|
| 43 |
+
- learning_rate: 0.0003
|
| 44 |
+
- train_batch_size: 2
|
| 45 |
+
- eval_batch_size: 8
|
| 46 |
+
- seed: 42
|
| 47 |
+
- distributed_type: multi-GPU
|
| 48 |
+
- num_devices: 2
|
| 49 |
+
- gradient_accumulation_steps: 8
|
| 50 |
+
- total_train_batch_size: 32
|
| 51 |
+
- total_eval_batch_size: 16
|
| 52 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 53 |
+
- lr_scheduler_type: linear
|
| 54 |
+
- lr_scheduler_warmup_steps: 100
|
| 55 |
+
- num_epochs: 50.0
|
| 56 |
+
- mixed_precision_training: Native AMP
|
| 57 |
+
|
| 58 |
+
### Training results
|
| 59 |
+
|
| 60 |
+
| Training Loss | Epoch | Step | Validation Loss | F1 | Accuracy |
|
| 61 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|:--------:|
|
| 62 |
+
| 2.6195 | 3.95 | 20 | 2.6348 | 0.0172 | 0.0816 |
|
| 63 |
+
| 2.5925 | 7.95 | 40 | 2.6119 | 0.0352 | 0.0851 |
|
| 64 |
+
| 2.1271 | 11.95 | 60 | 2.3066 | 0.1556 | 0.1986 |
|
| 65 |
+
| 1.2618 | 15.95 | 80 | 1.3810 | 0.6877 | 0.7128 |
|
| 66 |
+
| 0.5455 | 19.95 | 100 | 1.0403 | 0.6992 | 0.7270 |
|
| 67 |
+
| 0.2571 | 23.95 | 120 | 0.8423 | 0.8160 | 0.8121 |
|
| 68 |
+
| 0.3478 | 27.95 | 140 | 0.6500 | 0.8516 | 0.8440 |
|
| 69 |
+
| 0.0732 | 31.95 | 160 | 0.7066 | 0.8123 | 0.8156 |
|
| 70 |
+
| 0.1092 | 35.95 | 180 | 0.5878 | 0.8767 | 0.8759 |
|
| 71 |
+
| 0.0271 | 39.95 | 200 | 0.5994 | 0.8578 | 0.8617 |
|
| 72 |
+
| 0.4664 | 43.95 | 220 | 0.7830 | 0.8403 | 0.8440 |
|
| 73 |
+
| 0.0192 | 47.95 | 240 | 0.5685 | 0.8747 | 0.8759 |
|
| 74 |
+
|
| 75 |
+
|
| 76 |
+
### Framework versions
|
| 77 |
+
|
| 78 |
+
- Transformers 4.18.0
|
| 79 |
+
- Pytorch 1.11.0+cu113
|
| 80 |
+
- Datasets 2.1.0
|
| 81 |
+
- Tokenizers 0.12.1
|