End of training
Browse files- README.md +79 -0
- model.safetensors +1 -1
README.md
ADDED
|
@@ -0,0 +1,79 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: cc-by-nc-4.0
|
| 4 |
+
base_model: facebook/mms-300m
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
metrics:
|
| 8 |
+
- wer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: mms-300m-tir-Aki
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# mms-300m-tir-Aki
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [facebook/mms-300m](https://huggingface.co/facebook/mms-300m) on an unknown dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 1.4483
|
| 22 |
+
- Wer: 0.4084
|
| 23 |
+
- Cer: 0.1599
|
| 24 |
+
|
| 25 |
+
## Model description
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Intended uses & limitations
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training and evaluation data
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Training procedure
|
| 38 |
+
|
| 39 |
+
### Training hyperparameters
|
| 40 |
+
|
| 41 |
+
The following hyperparameters were used during training:
|
| 42 |
+
- learning_rate: 0.0001
|
| 43 |
+
- train_batch_size: 4
|
| 44 |
+
- eval_batch_size: 8
|
| 45 |
+
- seed: 42
|
| 46 |
+
- gradient_accumulation_steps: 8
|
| 47 |
+
- total_train_batch_size: 32
|
| 48 |
+
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 49 |
+
- lr_scheduler_type: linear
|
| 50 |
+
- lr_scheduler_warmup_steps: 500
|
| 51 |
+
- num_epochs: 30
|
| 52 |
+
- mixed_precision_training: Native AMP
|
| 53 |
+
|
| 54 |
+
### Training results
|
| 55 |
+
|
| 56 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|
| 57 |
+
|:-------------:|:------:|:----:|:---------------:|:------:|:------:|
|
| 58 |
+
| 32.5284 | 0.3932 | 500 | 4.3067 | 1.0 | 1.0 |
|
| 59 |
+
| 31.3983 | 0.7865 | 1000 | 4.1331 | 0.9973 | 0.9949 |
|
| 60 |
+
| 9.5147 | 1.1793 | 1500 | 1.6329 | 0.6428 | 0.2630 |
|
| 61 |
+
| 6.6968 | 1.5726 | 2000 | 1.4294 | 0.5261 | 0.2077 |
|
| 62 |
+
| 6.7439 | 1.9658 | 2500 | 1.3102 | 0.4827 | 0.1897 |
|
| 63 |
+
| 5.5428 | 2.3586 | 3000 | 1.3659 | 0.4639 | 0.1819 |
|
| 64 |
+
| 5.3838 | 2.7519 | 3500 | 1.3011 | 0.4520 | 0.1759 |
|
| 65 |
+
| 5.6637 | 3.1447 | 4000 | 1.2025 | 0.4413 | 0.1731 |
|
| 66 |
+
| 5.2161 | 3.5379 | 4500 | 1.3206 | 0.4345 | 0.1691 |
|
| 67 |
+
| 5.3430 | 3.9312 | 5000 | 1.2392 | 0.4305 | 0.1670 |
|
| 68 |
+
| 5.3286 | 4.3240 | 5500 | 1.1944 | 0.4309 | 0.1689 |
|
| 69 |
+
| 4.6537 | 4.7173 | 6000 | 1.3149 | 0.4183 | 0.1633 |
|
| 70 |
+
| 4.4528 | 5.1101 | 6500 | 1.3301 | 0.4082 | 0.1598 |
|
| 71 |
+
| 5.0576 | 5.5033 | 7000 | 1.4483 | 0.4084 | 0.1599 |
|
| 72 |
+
|
| 73 |
+
|
| 74 |
+
### Framework versions
|
| 75 |
+
|
| 76 |
+
- Transformers 5.0.0
|
| 77 |
+
- Pytorch 2.10.0+cu128
|
| 78 |
+
- Datasets 4.0.0
|
| 79 |
+
- Tokenizers 0.22.2
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1263262940
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5100b347995f47b630647dadb7be24325776a13043fd3d1f47c0503a658ea561
|
| 3 |
size 1263262940
|