Update README.md
Browse files
README.md
CHANGED
|
@@ -5,6 +5,9 @@ tags:
|
|
| 5 |
model-index:
|
| 6 |
- name: MUmairAB/bert-based-MaskedLM
|
| 7 |
results: []
|
|
|
|
|
|
|
|
|
|
| 8 |
---
|
| 9 |
|
| 10 |
<!-- This model card has been generated automatically according to the information Keras had access to. You should
|
|
@@ -12,23 +15,45 @@ probably proofread and complete it, then remove this comment. -->
|
|
| 12 |
|
| 13 |
# MUmairAB/bert-based-MaskedLM
|
| 14 |
|
| 15 |
-
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on
|
| 16 |
It achieves the following results on the evaluation set:
|
| 17 |
- Train Loss: 2.4360
|
| 18 |
- Validation Loss: 2.3284
|
| 19 |
-
- Epoch:
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
| 23 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 24 |
|
| 25 |
## Intended uses & limitations
|
| 26 |
|
| 27 |
-
|
| 28 |
|
| 29 |
## Training and evaluation data
|
| 30 |
|
| 31 |
-
|
| 32 |
|
| 33 |
## Training procedure
|
| 34 |
|
|
@@ -42,20 +67,31 @@ The following hyperparameters were used during training:
|
|
| 42 |
|
| 43 |
| Train Loss | Validation Loss | Epoch |
|
| 44 |
|:----------:|:---------------:|:-----:|
|
| 45 |
-
|
|
| 46 |
-
| 2.
|
| 47 |
-
| 2.
|
| 48 |
-
| 2.
|
| 49 |
-
| 2.
|
| 50 |
-
| 2.
|
| 51 |
-
| 2.
|
| 52 |
-
| 2.
|
| 53 |
-
| 2.
|
| 54 |
-
| 2.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 55 |
|
| 56 |
|
| 57 |
### Framework versions
|
| 58 |
|
| 59 |
- Transformers 4.30.2
|
| 60 |
- TensorFlow 2.12.0
|
| 61 |
-
- Tokenizers 0.13.3
|
|
|
|
| 5 |
model-index:
|
| 6 |
- name: MUmairAB/bert-based-MaskedLM
|
| 7 |
results: []
|
| 8 |
+
datasets:
|
| 9 |
+
- imdb
|
| 10 |
+
pipeline_tag: fill-mask
|
| 11 |
---
|
| 12 |
|
| 13 |
<!-- This model card has been generated automatically according to the information Keras had access to. You should
|
|
|
|
| 15 |
|
| 16 |
# MUmairAB/bert-based-MaskedLM
|
| 17 |
|
| 18 |
+
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on [IMDB Movies Review](https://huggingface.co/datasets/imdb) dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Train Loss: 2.4360
|
| 21 |
- Validation Loss: 2.3284
|
| 22 |
+
- Epoch: 20
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
| 26 |
+
[DistilBERT-base-uncased](https://huggingface.co/distilbert-base-uncased)
|
| 27 |
+
```
|
| 28 |
+
Model: "tf_distil_bert_for_masked_lm"
|
| 29 |
+
_________________________________________________________________
|
| 30 |
+
Layer (type) Output Shape Param #
|
| 31 |
+
=================================================================
|
| 32 |
+
distilbert (TFDistilBertMai multiple 66362880
|
| 33 |
+
nLayer)
|
| 34 |
+
|
| 35 |
+
vocab_transform (Dense) multiple 590592
|
| 36 |
+
|
| 37 |
+
vocab_layer_norm (LayerNorm multiple 1536
|
| 38 |
+
alization)
|
| 39 |
+
|
| 40 |
+
vocab_projector (TFDistilBe multiple 23866170
|
| 41 |
+
rtLMHead)
|
| 42 |
+
|
| 43 |
+
=================================================================
|
| 44 |
+
Total params: 66,985,530
|
| 45 |
+
Trainable params: 66,985,530
|
| 46 |
+
Non-trainable params: 0
|
| 47 |
+
_________________________________________________________________
|
| 48 |
+
```
|
| 49 |
|
| 50 |
## Intended uses & limitations
|
| 51 |
|
| 52 |
+
The model was trained on IMDB movies review dataset. So, it inherits the language biases from the dataset.
|
| 53 |
|
| 54 |
## Training and evaluation data
|
| 55 |
|
| 56 |
+
The model was trained on [IMDB Movies Review](https://huggingface.co/datasets/imdb) dataset.
|
| 57 |
|
| 58 |
## Training procedure
|
| 59 |
|
|
|
|
| 67 |
|
| 68 |
| Train Loss | Validation Loss | Epoch |
|
| 69 |
|:----------:|:---------------:|:-----:|
|
| 70 |
+
| 3.0754 | 2.7548 | 0 |
|
| 71 |
+
| 2.7969 | 2.6209 | 1 |
|
| 72 |
+
| 2.7214 | 2.5588 | 2 |
|
| 73 |
+
| 2.6626 | 2.5554 | 3 |
|
| 74 |
+
| 2.6466 | 2.4881 | 4 |
|
| 75 |
+
| 2.6238 | 2.4775 | 5 |
|
| 76 |
+
| 2.5696 | 2.4280 | 6 |
|
| 77 |
+
| 2.5504 | 2.3924 | 7 |
|
| 78 |
+
| 2.5171 | 2.3725 | 8 |
|
| 79 |
+
| 2.5180 | 2.3142 | 9 |
|
| 80 |
+
| 2.4443 | 2.2974 | 10 |
|
| 81 |
+
| 2.4497 | 2.3317 | 11 |
|
| 82 |
+
| 2.4371 | 2.3317 | 12 |
|
| 83 |
+
| 2.4377 | 2.3237 | 13 |
|
| 84 |
+
| 2.4369 | 2.3338 | 14 |
|
| 85 |
+
| 2.4350 | 2.3021 | 15 |
|
| 86 |
+
| 2.4267 | 2.3264 | 16 |
|
| 87 |
+
| 2.4557 | 2.3280 | 17 |
|
| 88 |
+
| 2.4461 | 2.3165 | 18 |
|
| 89 |
+
| 2.4360 | 2.3284 | 19 |
|
| 90 |
+
|
| 91 |
|
| 92 |
|
| 93 |
### Framework versions
|
| 94 |
|
| 95 |
- Transformers 4.30.2
|
| 96 |
- TensorFlow 2.12.0
|
| 97 |
+
- Tokenizers 0.13.3
|