update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: tiny-mlm-imdb
|
| 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 |
+
# tiny-mlm-imdb
|
| 14 |
+
|
| 15 |
+
This model is a fine-tuned version of [google/bert_uncased_L-2_H-128_A-2](https://huggingface.co/google/bert_uncased_L-2_H-128_A-2) on the None dataset.
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 3.5540
|
| 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: 3e-05
|
| 37 |
+
- train_batch_size: 32
|
| 38 |
+
- eval_batch_size: 32
|
| 39 |
+
- seed: 42
|
| 40 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 41 |
+
- lr_scheduler_type: constant
|
| 42 |
+
- num_epochs: 200
|
| 43 |
+
|
| 44 |
+
### Training results
|
| 45 |
+
|
| 46 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 47 |
+
|:-------------:|:-----:|:-----:|:---------------:|
|
| 48 |
+
| 4.2358 | 0.16 | 500 | 3.8225 |
|
| 49 |
+
| 4.1206 | 0.32 | 1000 | 3.7793 |
|
| 50 |
+
| 4.0857 | 0.48 | 1500 | 3.7520 |
|
| 51 |
+
| 4.0699 | 0.64 | 2000 | 3.7277 |
|
| 52 |
+
| 4.0378 | 0.8 | 2500 | 3.7125 |
|
| 53 |
+
| 4.0191 | 0.96 | 3000 | 3.7019 |
|
| 54 |
+
| 3.9747 | 1.12 | 3500 | 3.6871 |
|
| 55 |
+
| 3.9647 | 1.28 | 4000 | 3.6735 |
|
| 56 |
+
| 3.956 | 1.44 | 4500 | 3.6773 |
|
| 57 |
+
| 3.9574 | 1.6 | 5000 | 3.6580 |
|
| 58 |
+
| 3.9408 | 1.76 | 5500 | 3.6435 |
|
| 59 |
+
| 3.9421 | 1.92 | 6000 | 3.6419 |
|
| 60 |
+
| 3.9265 | 2.08 | 6500 | 3.6343 |
|
| 61 |
+
| 3.9198 | 2.24 | 7000 | 3.6306 |
|
| 62 |
+
| 3.9205 | 2.4 | 7500 | 3.6198 |
|
| 63 |
+
| 3.8985 | 2.56 | 8000 | 3.6158 |
|
| 64 |
+
| 3.9167 | 2.72 | 8500 | 3.6091 |
|
| 65 |
+
| 3.9111 | 2.88 | 9000 | 3.6073 |
|
| 66 |
+
| 3.8882 | 3.04 | 9500 | 3.5922 |
|
| 67 |
+
| 3.8761 | 3.2 | 10000 | 3.5908 |
|
| 68 |
+
| 3.8603 | 3.36 | 10500 | 3.5841 |
|
| 69 |
+
| 3.8621 | 3.52 | 11000 | 3.5835 |
|
| 70 |
+
| 3.8332 | 3.68 | 11500 | 3.5883 |
|
| 71 |
+
| 3.8523 | 3.84 | 12000 | 3.5798 |
|
| 72 |
+
| 3.8449 | 4.0 | 12500 | 3.5771 |
|
| 73 |
+
| 3.8284 | 4.16 | 13000 | 3.5653 |
|
| 74 |
+
| 3.8253 | 4.32 | 13500 | 3.5701 |
|
| 75 |
+
| 3.8021 | 4.48 | 14000 | 3.5681 |
|
| 76 |
+
| 3.8316 | 4.64 | 14500 | 3.5537 |
|
| 77 |
+
| 3.8318 | 4.8 | 15000 | 3.5609 |
|
| 78 |
+
| 3.82 | 4.96 | 15500 | 3.5579 |
|
| 79 |
+
| 3.8094 | 5.12 | 16000 | 3.5540 |
|
| 80 |
+
|
| 81 |
+
|
| 82 |
+
### Framework versions
|
| 83 |
+
|
| 84 |
+
- Transformers 4.25.1
|
| 85 |
+
- Pytorch 1.12.1
|
| 86 |
+
- Datasets 2.7.1
|
| 87 |
+
- Tokenizers 0.13.2
|