nazhan commited on
Commit
f9cd40e
1 Parent(s): 57afc74

Add SetFit model

Browse files
1_Pooling/config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 1024,
3
+ "pooling_mode_cls_token": true,
4
+ "pooling_mode_mean_tokens": false,
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
+ "include_prompt": true
10
+ }
README.md ADDED
@@ -0,0 +1,255 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: BAAI/bge-large-en-v1.5
3
+ library_name: setfit
4
+ metrics:
5
+ - accuracy
6
+ pipeline_tag: text-classification
7
+ tags:
8
+ - setfit
9
+ - sentence-transformers
10
+ - text-classification
11
+ - generated_from_setfit_trainer
12
+ widget:
13
+ - text: What’s the total number of orders placed by each customer?
14
+ - text: I like to read books and listen to music in my free time. How about you?
15
+ - text: Get company-wise intangible asset ratio.
16
+ - text: Show me data_asset_001_ta by product.
17
+ - text: Show me average asset value.
18
+ inference: true
19
+ model-index:
20
+ - name: SetFit with BAAI/bge-large-en-v1.5
21
+ results:
22
+ - task:
23
+ type: text-classification
24
+ name: Text Classification
25
+ dataset:
26
+ name: Unknown
27
+ type: unknown
28
+ split: test
29
+ metrics:
30
+ - type: accuracy
31
+ value: 0.9915254237288136
32
+ name: Accuracy
33
+ ---
34
+
35
+ # SetFit with BAAI/bge-large-en-v1.5
36
+
37
+ This is a [SetFit](https://github.com/huggingface/setfit) model that can be used for Text Classification. This SetFit model uses [BAAI/bge-large-en-v1.5](https://huggingface.co/BAAI/bge-large-en-v1.5) 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.
38
+
39
+ The model has been trained using an efficient few-shot learning technique that involves:
40
+
41
+ 1. Fine-tuning a [Sentence Transformer](https://www.sbert.net) with contrastive learning.
42
+ 2. Training a classification head with features from the fine-tuned Sentence Transformer.
43
+
44
+ ## Model Details
45
+
46
+ ### Model Description
47
+ - **Model Type:** SetFit
48
+ - **Sentence Transformer body:** [BAAI/bge-large-en-v1.5](https://huggingface.co/BAAI/bge-large-en-v1.5)
49
+ - **Classification head:** a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance
50
+ - **Maximum Sequence Length:** 512 tokens
51
+ - **Number of Classes:** 7 classes
52
+ <!-- - **Training Dataset:** [Unknown](https://huggingface.co/datasets/unknown) -->
53
+ <!-- - **Language:** Unknown -->
54
+ <!-- - **License:** Unknown -->
55
+
56
+ ### Model Sources
57
+
58
+ - **Repository:** [SetFit on GitHub](https://github.com/huggingface/setfit)
59
+ - **Paper:** [Efficient Few-Shot Learning Without Prompts](https://arxiv.org/abs/2209.11055)
60
+ - **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
61
+
62
+ ### Model Labels
63
+ | Label | Examples |
64
+ |:-------------|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
65
+ | Aggregation | <ul><li>'Please show med CostVariance_Actual_vs_Forecast.'</li><li>'Get me data_asset_001_kpm group by metrics.'</li><li>'Provide data_asset_kpi_cf group by quarter.'</li></ul> |
66
+ | Tablejoin | <ul><li>'Join data_asset_kpi_cf with data_asset_001_kpm tables.'</li><li>'Could you link the Products and Orders tables to track sales trends for different product categories?'</li><li>'Can I have a merge of income statement and key performance metrics tables?'</li></ul> |
67
+ | Lookup | <ul><li>"Filter by the 'Sales' department and show me the employees."</li><li>"Filter by the 'Toys' category and get me the product names."</li><li>'Can you get me the products with a price above 100?'</li></ul> |
68
+ | Rejection | <ul><li>"Let's avoid generating additional reports."</li><li>"I'd rather not filter this dataset."</li><li>"I'd prefer not to apply any filters."</li></ul> |
69
+ | Lookup_1 | <ul><li>'Show me key income statement metrics.'</li><li>'can I have kpm table'</li><li>'Retrieve data_asset_kpi_ma_product records.'</li></ul> |
70
+ | Generalreply | <ul><li>"Hey! It's going pretty well, thanks for asking. How about yours?"</li><li>'Not much, just taking it one day at a time. How about you?'</li><li>"'What is your favorite quote?'"</li></ul> |
71
+ | Viewtables | <ul><li>'What are the table names that relate to customer service in the starhub_data_asset database?'</li><li>'What tables are available in the starhub_data_asset database that can be joined to track user behavior?'</li><li>'What are the tables that are available for analysis in the starhub_data_asset database?'</li></ul> |
72
+
73
+ ## Evaluation
74
+
75
+ ### Metrics
76
+ | Label | Accuracy |
77
+ |:--------|:---------|
78
+ | **all** | 0.9915 |
79
+
80
+ ## Uses
81
+
82
+ ### Direct Use for Inference
83
+
84
+ First install the SetFit library:
85
+
86
+ ```bash
87
+ pip install setfit
88
+ ```
89
+
90
+ Then you can load this model and run inference.
91
+
92
+ ```python
93
+ from setfit import SetFitModel
94
+
95
+ # Download from the 🤗 Hub
96
+ model = SetFitModel.from_pretrained("nazhan/bge-large-en-v1.5-brahmaputra-iter-10-3rd")
97
+ # Run inference
98
+ preds = model("Show me average asset value.")
99
+ ```
100
+
101
+ <!--
102
+ ### Downstream Use
103
+
104
+ *List how someone could finetune this model on their own dataset.*
105
+ -->
106
+
107
+ <!--
108
+ ### Out-of-Scope Use
109
+
110
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
111
+ -->
112
+
113
+ <!--
114
+ ## Bias, Risks and Limitations
115
+
116
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
117
+ -->
118
+
119
+ <!--
120
+ ### Recommendations
121
+
122
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
123
+ -->
124
+
125
+ ## Training Details
126
+
127
+ ### Training Set Metrics
128
+ | Training set | Min | Median | Max |
129
+ |:-------------|:----|:-------|:----|
130
+ | Word count | 1 | 8.7839 | 62 |
131
+
132
+ | Label | Training Sample Count |
133
+ |:-------------|:----------------------|
134
+ | Tablejoin | 127 |
135
+ | Rejection | 76 |
136
+ | Aggregation | 281 |
137
+ | Lookup | 59 |
138
+ | Generalreply | 71 |
139
+ | Viewtables | 75 |
140
+ | Lookup_1 | 158 |
141
+
142
+ ### Training Hyperparameters
143
+ - batch_size: (16, 16)
144
+ - num_epochs: (1, 1)
145
+ - max_steps: 2450
146
+ - sampling_strategy: oversampling
147
+ - body_learning_rate: (2e-05, 1e-05)
148
+ - head_learning_rate: 0.01
149
+ - loss: CosineSimilarityLoss
150
+ - distance_metric: cosine_distance
151
+ - margin: 0.25
152
+ - end_to_end: False
153
+ - use_amp: False
154
+ - warmup_proportion: 0.1
155
+ - seed: 42
156
+ - eval_max_steps: -1
157
+ - load_best_model_at_end: True
158
+
159
+ ### Training Results
160
+ | Epoch | Step | Training Loss | Validation Loss |
161
+ |:----------:|:--------:|:-------------:|:---------------:|
162
+ | 0.0000 | 1 | 0.2291 | - |
163
+ | 0.0025 | 50 | 0.2181 | - |
164
+ | 0.0050 | 100 | 0.127 | - |
165
+ | 0.0075 | 150 | 0.015 | - |
166
+ | 0.0100 | 200 | 0.0072 | - |
167
+ | 0.0125 | 250 | 0.0034 | - |
168
+ | 0.0149 | 300 | 0.0032 | - |
169
+ | 0.0174 | 350 | 0.0032 | - |
170
+ | 0.0199 | 400 | 0.0019 | - |
171
+ | 0.0224 | 450 | 0.0014 | - |
172
+ | 0.0249 | 500 | 0.0012 | - |
173
+ | 0.0274 | 550 | 0.0011 | - |
174
+ | 0.0299 | 600 | 0.0018 | - |
175
+ | 0.0324 | 650 | 0.0013 | - |
176
+ | 0.0349 | 700 | 0.0015 | - |
177
+ | 0.0374 | 750 | 0.0009 | - |
178
+ | 0.0399 | 800 | 0.0012 | - |
179
+ | 0.0423 | 850 | 0.0008 | - |
180
+ | 0.0448 | 900 | 0.001 | - |
181
+ | 0.0473 | 950 | 0.0009 | - |
182
+ | 0.0498 | 1000 | 0.0007 | - |
183
+ | 0.0523 | 1050 | 0.0009 | - |
184
+ | 0.0548 | 1100 | 0.001 | - |
185
+ | 0.0573 | 1150 | 0.0008 | - |
186
+ | 0.0598 | 1200 | 0.0006 | - |
187
+ | 0.0623 | 1250 | 0.0007 | - |
188
+ | 0.0648 | 1300 | 0.0006 | - |
189
+ | 0.0673 | 1350 | 0.0007 | - |
190
+ | 0.0697 | 1400 | 0.0007 | - |
191
+ | 0.0722 | 1450 | 0.0008 | - |
192
+ | 0.0747 | 1500 | 0.0006 | - |
193
+ | 0.0772 | 1550 | 0.0008 | - |
194
+ | 0.0797 | 1600 | 0.0005 | - |
195
+ | 0.0822 | 1650 | 0.0009 | - |
196
+ | 0.0847 | 1700 | 0.0006 | - |
197
+ | 0.0872 | 1750 | 0.0007 | - |
198
+ | 0.0897 | 1800 | 0.0007 | - |
199
+ | 0.0922 | 1850 | 0.0006 | - |
200
+ | 0.0947 | 1900 | 0.0006 | - |
201
+ | 0.0971 | 1950 | 0.0007 | - |
202
+ | 0.0996 | 2000 | 0.0005 | - |
203
+ | 0.1021 | 2050 | 0.0005 | - |
204
+ | 0.1046 | 2100 | 0.0004 | - |
205
+ | 0.1071 | 2150 | 0.0006 | - |
206
+ | 0.1096 | 2200 | 0.0007 | - |
207
+ | 0.1121 | 2250 | 0.0004 | - |
208
+ | 0.1146 | 2300 | 0.0006 | - |
209
+ | 0.1171 | 2350 | 0.0008 | - |
210
+ | 0.1196 | 2400 | 0.0007 | - |
211
+ | **0.1221** | **2450** | **0.0004** | **0.013** |
212
+
213
+ * The bold row denotes the saved checkpoint.
214
+ ### Framework Versions
215
+ - Python: 3.11.9
216
+ - SetFit: 1.0.3
217
+ - Sentence Transformers: 2.7.0
218
+ - Transformers: 4.42.4
219
+ - PyTorch: 2.4.0+cu121
220
+ - Datasets: 2.21.0
221
+ - Tokenizers: 0.19.1
222
+
223
+ ## Citation
224
+
225
+ ### BibTeX
226
+ ```bibtex
227
+ @article{https://doi.org/10.48550/arxiv.2209.11055,
228
+ doi = {10.48550/ARXIV.2209.11055},
229
+ url = {https://arxiv.org/abs/2209.11055},
230
+ author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
231
+ keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
232
+ title = {Efficient Few-Shot Learning Without Prompts},
233
+ publisher = {arXiv},
234
+ year = {2022},
235
+ copyright = {Creative Commons Attribution 4.0 International}
236
+ }
237
+ ```
238
+
239
+ <!--
240
+ ## Glossary
241
+
242
+ *Clearly define terms in order to be accessible across audiences.*
243
+ -->
244
+
245
+ <!--
246
+ ## Model Card Authors
247
+
248
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
249
+ -->
250
+
251
+ <!--
252
+ ## Model Card Contact
253
+
254
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
255
+ -->
config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "checkpoints/step_2450",
3
+ "architectures": [
4
+ "BertModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "gradient_checkpointing": false,
9
+ "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 1024,
12
+ "id2label": {
13
+ "0": "LABEL_0"
14
+ },
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 4096,
17
+ "label2id": {
18
+ "LABEL_0": 0
19
+ },
20
+ "layer_norm_eps": 1e-12,
21
+ "max_position_embeddings": 512,
22
+ "model_type": "bert",
23
+ "num_attention_heads": 16,
24
+ "num_hidden_layers": 24,
25
+ "pad_token_id": 0,
26
+ "position_embedding_type": "absolute",
27
+ "torch_dtype": "float32",
28
+ "transformers_version": "4.42.4",
29
+ "type_vocab_size": 2,
30
+ "use_cache": true,
31
+ "vocab_size": 30522
32
+ }
config_sentence_transformers.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "sentence_transformers": "2.2.2",
4
+ "transformers": "4.28.1",
5
+ "pytorch": "1.13.0+cu117"
6
+ },
7
+ "prompts": {},
8
+ "default_prompt_name": null
9
+ }
config_setfit.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "normalize_embeddings": false,
3
+ "labels": [
4
+ "Tablejoin",
5
+ "Rejection",
6
+ "Aggregation",
7
+ "Lookup",
8
+ "Generalreply",
9
+ "Viewtables",
10
+ "Lookup_1"
11
+ ]
12
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4003dfe8a0709586fb1f3eaf59f3bed9d7d75b6a84620103c069c3ce7dac996
3
+ size 1340612432
model_head.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62d239cca6b566d2b6262616079d612ed74ced47cc4c5a8d5f2e6fa8442b62b4
3
+ size 58575
modules.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ {
15
+ "idx": 2,
16
+ "name": "2",
17
+ "path": "2_Normalize",
18
+ "type": "sentence_transformers.models.Normalize"
19
+ }
20
+ ]
sentence_bert_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "max_seq_length": 512,
3
+ "do_lower_case": true
4
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "mask_token": "[MASK]",
49
+ "max_length": 512,
50
+ "model_max_length": 512,
51
+ "never_split": null,
52
+ "pad_to_multiple_of": null,
53
+ "pad_token": "[PAD]",
54
+ "pad_token_type_id": 0,
55
+ "padding_side": "right",
56
+ "sep_token": "[SEP]",
57
+ "stride": 0,
58
+ "strip_accents": null,
59
+ "tokenize_chinese_chars": true,
60
+ "tokenizer_class": "BertTokenizer",
61
+ "truncation_side": "right",
62
+ "truncation_strategy": "longest_first",
63
+ "unk_token": "[UNK]"
64
+ }
vocab.txt ADDED
The diff for this file is too large to render. See raw diff