Upload TFBertForSequenceClassification
Browse files- README.md +30 -4
- config.json +8 -2
- tf_model.h5 +2 -2
README.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
-
base_model: veriga/
|
| 4 |
tags:
|
| 5 |
- generated_from_keras_callback
|
| 6 |
model-index:
|
|
@@ -13,9 +13,13 @@ probably proofread and complete it, then remove this comment. -->
|
|
| 13 |
|
| 14 |
# tf_disilbert_binary
|
| 15 |
|
| 16 |
-
This model is a fine-tuned version of [veriga/
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
@@ -34,11 +38,33 @@ More information needed
|
|
| 34 |
### Training hyperparameters
|
| 35 |
|
| 36 |
The following hyperparameters were used during training:
|
| 37 |
-
- optimizer: {'name': 'Adam', 'learning_rate':
|
| 38 |
- training_precision: float32
|
| 39 |
|
| 40 |
### Training results
|
| 41 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 42 |
|
| 43 |
|
| 44 |
### Framework versions
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
+
base_model: veriga/tf_disilbert_binary
|
| 4 |
tags:
|
| 5 |
- generated_from_keras_callback
|
| 6 |
model-index:
|
|
|
|
| 13 |
|
| 14 |
# tf_disilbert_binary
|
| 15 |
|
| 16 |
+
This model is a fine-tuned version of [veriga/tf_disilbert_binary](https://huggingface.co/veriga/tf_disilbert_binary) on an unknown dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
+
- Train Loss: 0.6899
|
| 19 |
+
- Train Binary Accuracy: 0.5442
|
| 20 |
+
- Validation Loss: 0.6920
|
| 21 |
+
- Validation Binary Accuracy: 0.5503
|
| 22 |
+
- Epoch: 19
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 38 |
### Training hyperparameters
|
| 39 |
|
| 40 |
The following hyperparameters were used during training:
|
| 41 |
+
- optimizer: {'name': 'Adam', 'learning_rate': 0.0005, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
|
| 42 |
- training_precision: float32
|
| 43 |
|
| 44 |
### Training results
|
| 45 |
|
| 46 |
+
| Train Loss | Train Binary Accuracy | Validation Loss | Validation Binary Accuracy | Epoch |
|
| 47 |
+
|:----------:|:---------------------:|:---------------:|:--------------------------:|:-----:|
|
| 48 |
+
| 0.6901 | 0.5372 | 0.6922 | 0.5486 | 0 |
|
| 49 |
+
| 0.6903 | 0.5430 | 0.6883 | 0.5495 | 1 |
|
| 50 |
+
| 0.6903 | 0.5392 | 0.6887 | 0.5486 | 2 |
|
| 51 |
+
| 0.6905 | 0.5378 | 0.6916 | 0.5477 | 3 |
|
| 52 |
+
| 0.6910 | 0.5355 | 0.6883 | 0.5495 | 4 |
|
| 53 |
+
| 0.6903 | 0.5410 | 0.6890 | 0.5469 | 5 |
|
| 54 |
+
| 0.6904 | 0.5423 | 0.6885 | 0.5486 | 6 |
|
| 55 |
+
| 0.6903 | 0.5425 | 0.6888 | 0.5486 | 7 |
|
| 56 |
+
| 0.6900 | 0.5421 | 0.6890 | 0.5495 | 8 |
|
| 57 |
+
| 0.6899 | 0.5435 | 0.6882 | 0.5503 | 9 |
|
| 58 |
+
| 0.6899 | 0.5421 | 0.6885 | 0.5512 | 10 |
|
| 59 |
+
| 0.6902 | 0.5435 | 0.6877 | 0.5521 | 11 |
|
| 60 |
+
| 0.6902 | 0.5404 | 0.6883 | 0.5512 | 12 |
|
| 61 |
+
| 0.6900 | 0.5420 | 0.6893 | 0.5477 | 13 |
|
| 62 |
+
| 0.6900 | 0.5430 | 0.6885 | 0.5486 | 14 |
|
| 63 |
+
| 0.6899 | 0.5435 | 0.6901 | 0.5521 | 15 |
|
| 64 |
+
| 0.6907 | 0.5415 | 0.6890 | 0.5469 | 16 |
|
| 65 |
+
| 0.6902 | 0.5417 | 0.6906 | 0.5495 | 17 |
|
| 66 |
+
| 0.6903 | 0.5395 | 0.6898 | 0.5503 | 18 |
|
| 67 |
+
| 0.6899 | 0.5442 | 0.6920 | 0.5503 | 19 |
|
| 68 |
|
| 69 |
|
| 70 |
### Framework versions
|
config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "veriga/
|
| 3 |
"architectures": [
|
| 4 |
"BertForSequenceClassification"
|
| 5 |
],
|
|
@@ -9,8 +9,14 @@
|
|
| 9 |
"hidden_act": "gelu",
|
| 10 |
"hidden_dropout_prob": 0.1,
|
| 11 |
"hidden_size": 768,
|
|
|
|
|
|
|
|
|
|
| 12 |
"initializer_range": 0.02,
|
| 13 |
"intermediate_size": 3072,
|
|
|
|
|
|
|
|
|
|
| 14 |
"layer_norm_eps": 1e-12,
|
| 15 |
"max_position_embeddings": 1024,
|
| 16 |
"model_type": "bert",
|
|
@@ -19,7 +25,7 @@
|
|
| 19 |
"pad_token_id": 0,
|
| 20 |
"position_embedding_type": "absolute",
|
| 21 |
"problem_type": "multi_label_classification",
|
| 22 |
-
"torch_dtype": "
|
| 23 |
"transformers_version": "4.36.2",
|
| 24 |
"type_vocab_size": 2,
|
| 25 |
"use_cache": true,
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "veriga/tf_disilbert_binary",
|
| 3 |
"architectures": [
|
| 4 |
"BertForSequenceClassification"
|
| 5 |
],
|
|
|
|
| 9 |
"hidden_act": "gelu",
|
| 10 |
"hidden_dropout_prob": 0.1,
|
| 11 |
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "LABEL_0"
|
| 14 |
+
},
|
| 15 |
"initializer_range": 0.02,
|
| 16 |
"intermediate_size": 3072,
|
| 17 |
+
"label2id": {
|
| 18 |
+
"LABEL_0": 0
|
| 19 |
+
},
|
| 20 |
"layer_norm_eps": 1e-12,
|
| 21 |
"max_position_embeddings": 1024,
|
| 22 |
"model_type": "bert",
|
|
|
|
| 25 |
"pad_token_id": 0,
|
| 26 |
"position_embedding_type": "absolute",
|
| 27 |
"problem_type": "multi_label_classification",
|
| 28 |
+
"torch_dtype": "float64",
|
| 29 |
"transformers_version": "4.36.2",
|
| 30 |
"type_vocab_size": 2,
|
| 31 |
"use_cache": true,
|
tf_model.h5
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e8f8f13f2d0c69c1f25718b29ef2e24186f2043d0f9a7e8ffefa50327f43fbce
|
| 3 |
+
size 448326996
|