update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- generated_from_trainer
|
| 4 |
+
model-index:
|
| 5 |
+
- name: TrOCR-Ar-Small
|
| 6 |
+
results: []
|
| 7 |
+
---
|
| 8 |
+
|
| 9 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 10 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 11 |
+
|
| 12 |
+
# TrOCR-Ar-Small
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [microsoft/trocr-small-stage1](https://huggingface.co/microsoft/trocr-small-stage1) on an unknown dataset.
|
| 15 |
+
It achieves the following results on the evaluation set:
|
| 16 |
+
- Loss: 2.2771
|
| 17 |
+
- Cer: 0.8211
|
| 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: 5e-05
|
| 37 |
+
- train_batch_size: 1
|
| 38 |
+
- eval_batch_size: 1
|
| 39 |
+
- seed: 42
|
| 40 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 41 |
+
- lr_scheduler_type: linear
|
| 42 |
+
- num_epochs: 3.0
|
| 43 |
+
- mixed_precision_training: Native AMP
|
| 44 |
+
|
| 45 |
+
### Training results
|
| 46 |
+
|
| 47 |
+
| Training Loss | Epoch | Step | Validation Loss | Cer |
|
| 48 |
+
|:-------------:|:-----:|:-----:|:---------------:|:------:|
|
| 49 |
+
| 3.6363 | 0.14 | 1000 | 2.7594 | 0.9370 |
|
| 50 |
+
| 2.7508 | 0.29 | 2000 | 2.6589 | 0.8901 |
|
| 51 |
+
| 2.6519 | 0.43 | 3000 | 2.6059 | 0.8647 |
|
| 52 |
+
| 2.5936 | 0.57 | 4000 | 2.5360 | 0.7941 |
|
| 53 |
+
| 2.5069 | 0.72 | 5000 | 2.4701 | 0.8262 |
|
| 54 |
+
| 2.4606 | 0.86 | 6000 | 2.4427 | 0.7552 |
|
| 55 |
+
| 2.4046 | 1.0 | 7000 | 2.4262 | 0.7822 |
|
| 56 |
+
| 2.3628 | 1.15 | 8000 | 2.3880 | 0.8186 |
|
| 57 |
+
| 2.3458 | 1.29 | 9000 | 2.3589 | 0.8262 |
|
| 58 |
+
| 2.3062 | 1.43 | 10000 | 2.3704 | 0.8693 |
|
| 59 |
+
| 2.2884 | 1.58 | 11000 | 2.3065 | 0.8034 |
|
| 60 |
+
| 2.263 | 1.72 | 12000 | 2.3413 | 0.8545 |
|
| 61 |
+
| 2.2473 | 1.86 | 13000 | 2.3314 | 0.7996 |
|
| 62 |
+
| 2.2318 | 2.01 | 14000 | 2.3034 | 0.8254 |
|
| 63 |
+
| 2.2004 | 2.15 | 15000 | 2.3068 | 0.8461 |
|
| 64 |
+
| 2.1774 | 2.29 | 16000 | 2.2799 | 0.8207 |
|
| 65 |
+
| 2.1684 | 2.44 | 17000 | 2.2746 | 0.8249 |
|
| 66 |
+
| 2.1637 | 2.58 | 18000 | 2.2540 | 0.7797 |
|
| 67 |
+
| 2.1418 | 2.72 | 19000 | 2.2595 | 0.7937 |
|
| 68 |
+
| 2.1309 | 2.87 | 20000 | 2.2771 | 0.8211 |
|
| 69 |
+
|
| 70 |
+
|
| 71 |
+
### Framework versions
|
| 72 |
+
|
| 73 |
+
- Transformers 4.17.0
|
| 74 |
+
- Pytorch 1.10.0+cu111
|
| 75 |
+
- Datasets 2.0.0
|
| 76 |
+
- Tokenizers 0.11.6
|