Hugging Face's logo

qunfengd
/
esm2_t30_150M_UR50D-finetuned-AMP_Classification

Text Classification
Transformers
TensorFlow
esm
generated_from_keras_callback
Inference Endpoints
Model card Files Community
qunfengd commited on
Commit
c53f8c6
•
1 Parent(s): f9e30a2

Upload TFEsmForSequenceClassification

Browse files
Files changed (3) hide show
  1. README.md +75 -0
  2. config.json +38 -0
  3. tf_model.h5 +3 -0
README.md ADDED
@@ -0,0 +1,75 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ base_model: facebook/esm2_t30_150M_UR50D
4
+ tags:
5
+ - generated_from_keras_callback
6
+ model-index:
7
+ - name: esm2_t30_150M_UR50D-finetuned-AMP_Classification
8
+ results: []
9
+ ---
10
+
11
+ <!-- This model card has been generated automatically according to the information Keras had access to. You should
12
+ probably proofread and complete it, then remove this comment. -->
13
+
14
+ # esm2_t30_150M_UR50D-finetuned-AMP_Classification
15
+
16
+ This model is a fine-tuned version of [facebook/esm2_t30_150M_UR50D](https://huggingface.co/facebook/esm2_t30_150M_UR50D) on an unknown dataset.
17
+ It achieves the following results on the evaluation set:
18
+ - Train Loss: 0.0433
19
+ - Train Accuracy: 0.9871
20
+ - Validation Loss: 0.7702
21
+ - Validation Accuracy: 0.8014
22
+ - Epoch: 19
23
+
24
+ ## Model description
25
+
26
+ More information needed
27
+
28
+ ## Intended uses & limitations
29
+
30
+ More information needed
31
+
32
+ ## Training and evaluation data
33
+
34
+ More information needed
35
+
36
+ ## Training procedure
37
+
38
+ ### Training hyperparameters
39
+
40
+ The following hyperparameters were used during training:
41
+ - optimizer: {'name': 'AdamWeightDecay', 'learning_rate': 2e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.0}
42
+ - training_precision: float32
43
+
44
+ ### Training results
45
+
46
+ | Train Loss | Train Accuracy | Validation Loss | Validation Accuracy | Epoch |
47
+ |:----------:|:--------------:|:---------------:|:-------------------:|:-----:|
48
+ | 0.6498 | 0.6047 | 0.6345 | 0.6288 | 0 |
49
+ | 0.5714 | 0.6877 | 0.5871 | 0.6882 | 1 |
50
+ | 0.3898 | 0.8198 | 0.5698 | 0.7242 | 2 |
51
+ | 0.2481 | 0.8921 | 0.5758 | 0.7696 | 3 |
52
+ | 0.1838 | 0.9248 | 0.6483 | 0.7730 | 4 |
53
+ | 0.1475 | 0.9390 | 0.6187 | 0.7904 | 5 |
54
+ | 0.1147 | 0.9541 | 0.6663 | 0.8007 | 6 |
55
+ | 0.0948 | 0.9618 | 0.7591 | 0.7819 | 7 |
56
+ | 0.0800 | 0.9701 | 0.7534 | 0.7959 | 8 |
57
+ | 0.0709 | 0.9739 | 0.8595 | 0.7810 | 9 |
58
+ | 0.0629 | 0.9767 | 0.8192 | 0.7907 | 10 |
59
+ | 0.0578 | 0.9792 | 0.8855 | 0.7946 | 11 |
60
+ | 0.0532 | 0.9814 | 0.9993 | 0.7762 | 12 |
61
+ | 0.0586 | 0.9801 | 0.9058 | 0.7761 | 13 |
62
+ | 0.0534 | 0.9816 | 0.8338 | 0.7786 | 14 |
63
+ | 0.0508 | 0.9824 | 0.7899 | 0.8033 | 15 |
64
+ | 0.0472 | 0.9840 | 0.9000 | 0.7800 | 16 |
65
+ | 0.0441 | 0.9851 | 0.8732 | 0.7911 | 17 |
66
+ | 0.0486 | 0.9846 | 0.8166 | 0.8088 | 18 |
67
+ | 0.0433 | 0.9871 | 0.7702 | 0.8014 | 19 |
68
+
69
+
70
+ ### Framework versions
71
+
72
+ - Transformers 4.40.1
73
+ - TensorFlow 2.15.0
74
+ - Datasets 2.19.1
75
+ - Tokenizers 0.19.1
config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/esm2_t30_150M_UR50D",
3
+ "architectures": [
4
+ "EsmForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.0,
7
+ "classifier_dropout": null,
8
+ "emb_layer_norm_before": false,
9
+ "esmfold_config": null,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.0,
12
+ "hidden_size": 640,
13
+ "id2label": {
14
+ "0": "negative",
15
+ "1": "positive"
16
+ },
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 2560,
19
+ "is_folding_model": false,
20
+ "label2id": {
21
+ "negative": 0,
22
+ "positive": 1
23
+ },
24
+ "layer_norm_eps": 1e-05,
25
+ "mask_token_id": 32,
26
+ "max_position_embeddings": 1026,
27
+ "model_type": "esm",
28
+ "num_attention_heads": 20,
29
+ "num_hidden_layers": 30,
30
+ "pad_token_id": 1,
31
+ "position_embedding_type": "rotary",
32
+ "token_dropout": true,
33
+ "torch_dtype": "float32",
34
+ "transformers_version": "4.40.1",
35
+ "use_cache": true,
36
+ "vocab_list": null,
37
+ "vocab_size": 33
38
+ }
tf_model.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:65ddcc9ae19b3ae6734ca99c5c30430583d71ff76020d26fb4c061b93bb1505c
3
+ size 595896088