spaly99 commited on
Commit
5bd93f6
1 Parent(s): 06ba6ae

Add SetFit model

Browse files
1_Pooling/config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 768,
3
+ "pooling_mode_cls_token": false,
4
+ "pooling_mode_mean_tokens": true,
5
+ "pooling_mode_max_tokens": false,
6
+ "pooling_mode_mean_sqrt_len_tokens": false,
7
+ "pooling_mode_weightedmean_tokens": false,
8
+ "pooling_mode_lasttoken": false
9
+ }
README.md ADDED
@@ -0,0 +1,217 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: setfit
3
+ tags:
4
+ - setfit
5
+ - sentence-transformers
6
+ - text-classification
7
+ - generated_from_setfit_trainer
8
+ metrics:
9
+ - accuracy
10
+ - precision
11
+ - recall
12
+ - f1
13
+ widget:
14
+ - text: 'mar | : { ‘ : Ui * | z : gg ‘ey bal _ . '' '
15
+ - text: www.statista.fr
16
+ - text: French Polynesia
17
+ - text: 'SHOP DRESS SHIRTS > '
18
+ - text: '’ \ Rather Go Naked a Wé FE '
19
+ pipeline_tag: text-classification
20
+ inference: true
21
+ base_model: sentence-transformers/paraphrase-mpnet-base-v2
22
+ model-index:
23
+ - name: SetFit with sentence-transformers/paraphrase-mpnet-base-v2
24
+ results:
25
+ - task:
26
+ type: text-classification
27
+ name: Text Classification
28
+ dataset:
29
+ name: Unknown
30
+ type: unknown
31
+ split: test
32
+ metrics:
33
+ - type: accuracy
34
+ value: 0.9
35
+ name: Accuracy
36
+ - type: precision
37
+ value: 0.8823529411764706
38
+ name: Precision
39
+ - type: recall
40
+ value: 0.8823529411764706
41
+ name: Recall
42
+ - type: f1
43
+ value: 0.8823529411764706
44
+ name: F1
45
+ ---
46
+
47
+ # SetFit with sentence-transformers/paraphrase-mpnet-base-v2
48
+
49
+ This is a [SetFit](https://github.com/huggingface/setfit) model that can be used for Text Classification. This SetFit model uses [sentence-transformers/paraphrase-mpnet-base-v2](https://huggingface.co/sentence-transformers/paraphrase-mpnet-base-v2) as the Sentence Transformer embedding model. A [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance is used for classification.
50
+
51
+ The model has been trained using an efficient few-shot learning technique that involves:
52
+
53
+ 1. Fine-tuning a [Sentence Transformer](https://www.sbert.net) with contrastive learning.
54
+ 2. Training a classification head with features from the fine-tuned Sentence Transformer.
55
+
56
+ ## Model Details
57
+
58
+ ### Model Description
59
+ - **Model Type:** SetFit
60
+ - **Sentence Transformer body:** [sentence-transformers/paraphrase-mpnet-base-v2](https://huggingface.co/sentence-transformers/paraphrase-mpnet-base-v2)
61
+ - **Classification head:** a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance
62
+ - **Maximum Sequence Length:** 512 tokens
63
+ - **Number of Classes:** 2 classes
64
+ <!-- - **Training Dataset:** [Unknown](https://huggingface.co/datasets/unknown) -->
65
+ <!-- - **Language:** Unknown -->
66
+ <!-- - **License:** Unknown -->
67
+
68
+ ### Model Sources
69
+
70
+ - **Repository:** [SetFit on GitHub](https://github.com/huggingface/setfit)
71
+ - **Paper:** [Efficient Few-Shot Learning Without Prompts](https://arxiv.org/abs/2209.11055)
72
+ - **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
73
+
74
+ ### Model Labels
75
+ | Label | Examples |
76
+ |:------|:------------------------------------------------------------------------------------------------------------------------------------------------------------------|
77
+ | True | <ul><li>'WHEN DID HE STOP TREATING YOU LIKE A PRINGESS? '</li><li>'Samsung” 65" Class Curved 4K Ultra HD '</li><li>'Finger Lickin’ Mini Fillet Burge: '</li></ul> |
78
+ | False | <ul><li>'speedtest.net'</li><li>'Dane z przeglądania i interakcji'</li><li>'Announcements'</li></ul> |
79
+
80
+ ## Evaluation
81
+
82
+ ### Metrics
83
+ | Label | Accuracy | Precision | Recall | F1 |
84
+ |:--------|:---------|:----------|:-------|:-------|
85
+ | **all** | 0.9 | 0.8824 | 0.8824 | 0.8824 |
86
+
87
+ ## Uses
88
+
89
+ ### Direct Use for Inference
90
+
91
+ First install the SetFit library:
92
+
93
+ ```bash
94
+ pip install setfit
95
+ ```
96
+
97
+ Then you can load this model and run inference.
98
+
99
+ ```python
100
+ from setfit import SetFitModel
101
+
102
+ # Download from the 🤗 Hub
103
+ model = SetFitModel.from_pretrained("setfit_model_id")
104
+ # Run inference
105
+ preds = model("www.statista.fr")
106
+ ```
107
+
108
+ <!--
109
+ ### Downstream Use
110
+
111
+ *List how someone could finetune this model on their own dataset.*
112
+ -->
113
+
114
+ <!--
115
+ ### Out-of-Scope Use
116
+
117
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
118
+ -->
119
+
120
+ <!--
121
+ ## Bias, Risks and Limitations
122
+
123
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
124
+ -->
125
+
126
+ <!--
127
+ ### Recommendations
128
+
129
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
130
+ -->
131
+
132
+ ## Training Details
133
+
134
+ ### Training Set Metrics
135
+ | Training set | Min | Median | Max |
136
+ |:-------------|:----|:-------|:----|
137
+ | Word count | 1 | 8.6188 | 267 |
138
+
139
+ | Label | Training Sample Count |
140
+ |:------|:----------------------|
141
+ | False | 77 |
142
+ | True | 83 |
143
+
144
+ ### Training Hyperparameters
145
+ - batch_size: (16, 2)
146
+ - num_epochs: (1, 16)
147
+ - max_steps: -1
148
+ - sampling_strategy: oversampling
149
+ - num_iterations: 20
150
+ - body_learning_rate: (2e-05, 1e-05)
151
+ - head_learning_rate: 0.01
152
+ - loss: CosineSimilarityLoss
153
+ - distance_metric: cosine_distance
154
+ - margin: 0.25
155
+ - end_to_end: False
156
+ - use_amp: False
157
+ - warmup_proportion: 0.1
158
+ - seed: 42
159
+ - run_name: PG-OCR-test-2
160
+ - eval_max_steps: -1
161
+ - load_best_model_at_end: False
162
+
163
+ ### Training Results
164
+ | Epoch | Step | Training Loss | Validation Loss |
165
+ |:------:|:----:|:-------------:|:---------------:|
166
+ | 0.0025 | 1 | 0.3647 | - |
167
+ | 0.125 | 50 | 0.2765 | - |
168
+ | 0.25 | 100 | 0.0519 | - |
169
+ | 0.375 | 150 | 0.0045 | - |
170
+ | 0.5 | 200 | 0.055 | - |
171
+ | 0.625 | 250 | 0.0033 | - |
172
+ | 0.75 | 300 | 0.1017 | - |
173
+ | 0.875 | 350 | 0.02 | - |
174
+ | 1.0 | 400 | 0.0015 | - |
175
+
176
+ ### Framework Versions
177
+ - Python: 3.11.0
178
+ - SetFit: 1.0.3
179
+ - Sentence Transformers: 2.3.0
180
+ - Transformers: 4.37.2
181
+ - PyTorch: 2.2.1+cu121
182
+ - Datasets: 2.16.1
183
+ - Tokenizers: 0.15.1
184
+
185
+ ## Citation
186
+
187
+ ### BibTeX
188
+ ```bibtex
189
+ @article{https://doi.org/10.48550/arxiv.2209.11055,
190
+ doi = {10.48550/ARXIV.2209.11055},
191
+ url = {https://arxiv.org/abs/2209.11055},
192
+ author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
193
+ keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
194
+ title = {Efficient Few-Shot Learning Without Prompts},
195
+ publisher = {arXiv},
196
+ year = {2022},
197
+ copyright = {Creative Commons Attribution 4.0 International}
198
+ }
199
+ ```
200
+
201
+ <!--
202
+ ## Glossary
203
+
204
+ *Clearly define terms in order to be accessible across audiences.*
205
+ -->
206
+
207
+ <!--
208
+ ## Model Card Authors
209
+
210
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
211
+ -->
212
+
213
+ <!--
214
+ ## Model Card Contact
215
+
216
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
217
+ -->
config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "sentence-transformers/paraphrase-mpnet-base-v2",
3
+ "architectures": [
4
+ "MPNetModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "eos_token_id": 2,
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-05,
15
+ "max_position_embeddings": 514,
16
+ "model_type": "mpnet",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 12,
19
+ "pad_token_id": 1,
20
+ "relative_attention_num_buckets": 32,
21
+ "torch_dtype": "float32",
22
+ "transformers_version": "4.37.2",
23
+ "vocab_size": 30527
24
+ }
config_sentence_transformers.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "sentence_transformers": "2.0.0",
4
+ "transformers": "4.7.0",
5
+ "pytorch": "1.9.0+cu102"
6
+ }
7
+ }
config_setfit.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "normalize_embeddings": false,
3
+ "labels": null
4
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:80bb1496041fcd14467a4f837dbe4fc62bb21d514e269c7ae434a07928da9e16
3
+ size 437967672
model_head.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bd7626ba6a3a59755deeaf71686b94e09eceb7dea2b743835558d6e7e079bf05
3
+ size 6991
modules.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "idx": 0,
4
+ "name": "0",
5
+ "path": "",
6
+ "type": "sentence_transformers.models.Transformer"
7
+ },
8
+ {
9
+ "idx": 1,
10
+ "name": "1",
11
+ "path": "1_Pooling",
12
+ "type": "sentence_transformers.models.Pooling"
13
+ }
14
+ ]
sentence_bert_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "max_seq_length": 512,
3
+ "do_lower_case": false
4
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "<mask>",
25
+ "lstrip": true,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "<pad>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "</s>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "[UNK]",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "104": {
28
+ "content": "[UNK]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "30526": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": true,
46
+ "cls_token": "<s>",
47
+ "do_basic_tokenize": true,
48
+ "do_lower_case": true,
49
+ "eos_token": "</s>",
50
+ "mask_token": "<mask>",
51
+ "model_max_length": 512,
52
+ "never_split": null,
53
+ "pad_token": "<pad>",
54
+ "sep_token": "</s>",
55
+ "strip_accents": null,
56
+ "tokenize_chinese_chars": true,
57
+ "tokenizer_class": "MPNetTokenizer",
58
+ "unk_token": "[UNK]"
59
+ }
vocab.txt ADDED
The diff for this file is too large to render. See raw diff