ConstantinosVidiniotis commited on
Commit
ee97f41
·
verified ·
1 Parent(s): 469cf30

Push model using huggingface_hub.

Browse files
Files changed (4) hide show
  1. README.md +80 -41
  2. config_setfit.json +2 -2
  3. model.safetensors +1 -1
  4. model_head.pkl +1 -1
README.md CHANGED
@@ -5,31 +5,17 @@ tags:
5
  - text-classification
6
  - generated_from_setfit_trainer
7
  widget:
8
- - text: Why is the policy slow?
9
- - text: I would like to better understand why my pass-code is not working.
10
- - text: the compliance posture appears to be wrong.
11
- - text: Can you verify the balance?
12
- - text: Are there limits as to where my card will be accepted?
13
  metrics:
14
  - accuracy
15
  pipeline_tag: text-classification
16
  library_name: setfit
17
  inference: true
18
  base_model: sentence-transformers/all-MiniLM-L6-v2
19
- model-index:
20
- - name: SetFit with sentence-transformers/all-MiniLM-L6-v2
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: 0.695
32
- name: Accuracy
33
  ---
34
 
35
  # SetFit with sentence-transformers/all-MiniLM-L6-v2
@@ -60,19 +46,12 @@ The model has been trained using an efficient few-shot learning technique that i
60
  - **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
61
 
62
  ### Model Labels
63
- | Label | Examples |
64
- |:-----------|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
65
- | assertion | <ul><li>'It seems that the server compare.'</li><li>'I think the port is correct.'</li><li>'I believe the port is missing.'</li></ul> |
66
- | question | <ul><li>"I am suppose to be getting a refund from something I purchased but it hasn't showed up in my account."</li><li>'When should I schedule the report?'</li><li>'How do I change my account?'</li></ul> |
67
- | request | <ul><li>'update the host for me.'</li><li>'The ATM keeps declining my card! I tried two different ATMs already can you please check if everything is alright with my account??'</li><li>"I haven't gotten my new card."</li></ul> |
68
- | expressive | <ul><li>'Great, the balance is working!'</li><li>'I feel missing about the port.'</li><li>'Amazing, the invoice is fixed!'</li></ul> |
69
-
70
- ## Evaluation
71
-
72
- ### Metrics
73
- | Label | Accuracy |
74
- |:--------|:---------|
75
- | **all** | 0.695 |
76
 
77
  ## Uses
78
 
@@ -92,7 +71,7 @@ from setfit import SetFitModel
92
  # Download from the 🤗 Hub
93
  model = SetFitModel.from_pretrained("AutoCyberAI/crp-intent-setfit")
94
  # Run inference
95
- preds = model("Why is the policy slow?")
96
  ```
97
 
98
  <!--
@@ -124,14 +103,14 @@ preds = model("Why is the policy slow?")
124
  ### Training Set Metrics
125
  | Training set | Min | Median | Max |
126
  |:-------------|:----|:-------|:----|
127
- | Word count | 4 | 7.2188 | 23 |
128
 
129
  | Label | Training Sample Count |
130
  |:-----------|:----------------------|
131
- | assertion | 8 |
132
- | expressive | 8 |
133
- | question | 8 |
134
- | request | 8 |
135
 
136
  ### Training Hyperparameters
137
  - batch_size: (16, 16)
@@ -149,13 +128,73 @@ preds = model("Why is the policy slow?")
149
  - l2_weight: 0.01
150
  - seed: 42
151
  - eval_max_steps: -1
152
- - load_best_model_at_end: True
153
 
154
  ### Training Results
155
  | Epoch | Step | Training Loss | Validation Loss |
156
  |:------:|:----:|:-------------:|:---------------:|
157
- | 0.0208 | 1 | 0.3138 | - |
158
- | 1.0 | 48 | - | 0.2348 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
159
 
160
  ### Framework Versions
161
  - Python: 3.11.6
 
5
  - text-classification
6
  - generated_from_setfit_trainer
7
  widget:
8
+ - text: Why is a transfer I did recently still processing?
9
+ - text: Can you find the port?
10
+ - text: It seems that the balance find.
11
+ - text: the compliance posture appears to be updated.
12
+ - text: I believe the report is correct.
13
  metrics:
14
  - accuracy
15
  pipeline_tag: text-classification
16
  library_name: setfit
17
  inference: true
18
  base_model: sentence-transformers/all-MiniLM-L6-v2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
19
  ---
20
 
21
  # SetFit with sentence-transformers/all-MiniLM-L6-v2
 
46
  - **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
47
 
48
  ### Model Labels
49
+ | Label | Examples |
50
+ |:-----------|:---------------------------------------------------------------------------------------------------------------------------------------------|
51
+ | question | <ul><li>'Cash withdrawal was incorrect at ATM.'</li><li>'What is the server?'</li><li>'What is the invoice?'</li></ul> |
52
+ | request | <ul><li>'I need you to show the policy.'</li><li>'Can you summarise the compliance posture?'</li><li>'Can you find the port?'</li></ul> |
53
+ | assertion | <ul><li>'It seems that the report schedule.'</li><li>'I believe the balance is secure.'</li><li>'This is the compliance posture.'</li></ul> |
54
+ | expressive | <ul><li>'Great, the host is working!'</li><li>'I am frustrated with the port.'</li><li>'Amazing, the compliance posture is fixed!'</li></ul> |
 
 
 
 
 
 
 
55
 
56
  ## Uses
57
 
 
71
  # Download from the 🤗 Hub
72
  model = SetFitModel.from_pretrained("AutoCyberAI/crp-intent-setfit")
73
  # Run inference
74
+ preds = model("Can you find the port?")
75
  ```
76
 
77
  <!--
 
103
  ### Training Set Metrics
104
  | Training set | Min | Median | Max |
105
  |:-------------|:----|:-------|:----|
106
+ | Word count | 4 | 7.4883 | 53 |
107
 
108
  | Label | Training Sample Count |
109
  |:-----------|:----------------------|
110
+ | assertion | 64 |
111
+ | expressive | 64 |
112
+ | question | 64 |
113
+ | request | 64 |
114
 
115
  ### Training Hyperparameters
116
  - batch_size: (16, 16)
 
128
  - l2_weight: 0.01
129
  - seed: 42
130
  - eval_max_steps: -1
131
+ - load_best_model_at_end: False
132
 
133
  ### Training Results
134
  | Epoch | Step | Training Loss | Validation Loss |
135
  |:------:|:----:|:-------------:|:---------------:|
136
+ | 0.0003 | 1 | 0.4095 | - |
137
+ | 0.0163 | 50 | 0.3272 | - |
138
+ | 0.0326 | 100 | 0.2618 | - |
139
+ | 0.0488 | 150 | 0.1939 | - |
140
+ | 0.0651 | 200 | 0.1145 | - |
141
+ | 0.0814 | 250 | 0.0689 | - |
142
+ | 0.0977 | 300 | 0.0714 | - |
143
+ | 0.1139 | 350 | 0.0542 | - |
144
+ | 0.1302 | 400 | 0.0306 | - |
145
+ | 0.1465 | 450 | 0.0145 | - |
146
+ | 0.1628 | 500 | 0.0044 | - |
147
+ | 0.1790 | 550 | 0.0026 | - |
148
+ | 0.1953 | 600 | 0.0022 | - |
149
+ | 0.2116 | 650 | 0.0015 | - |
150
+ | 0.2279 | 700 | 0.0013 | - |
151
+ | 0.2441 | 750 | 0.0012 | - |
152
+ | 0.2604 | 800 | 0.001 | - |
153
+ | 0.2767 | 850 | 0.0008 | - |
154
+ | 0.2930 | 900 | 0.0009 | - |
155
+ | 0.3092 | 950 | 0.0008 | - |
156
+ | 0.3255 | 1000 | 0.0006 | - |
157
+ | 0.3418 | 1050 | 0.0006 | - |
158
+ | 0.3581 | 1100 | 0.0006 | - |
159
+ | 0.3743 | 1150 | 0.0006 | - |
160
+ | 0.3906 | 1200 | 0.0005 | - |
161
+ | 0.4069 | 1250 | 0.0005 | - |
162
+ | 0.4232 | 1300 | 0.0005 | - |
163
+ | 0.4395 | 1350 | 0.0005 | - |
164
+ | 0.4557 | 1400 | 0.0004 | - |
165
+ | 0.4720 | 1450 | 0.0004 | - |
166
+ | 0.4883 | 1500 | 0.0004 | - |
167
+ | 0.5046 | 1550 | 0.0004 | - |
168
+ | 0.5208 | 1600 | 0.0004 | - |
169
+ | 0.5371 | 1650 | 0.0004 | - |
170
+ | 0.5534 | 1700 | 0.0004 | - |
171
+ | 0.5697 | 1750 | 0.0003 | - |
172
+ | 0.5859 | 1800 | 0.0003 | - |
173
+ | 0.6022 | 1850 | 0.0003 | - |
174
+ | 0.6185 | 1900 | 0.0003 | - |
175
+ | 0.6348 | 1950 | 0.0003 | - |
176
+ | 0.6510 | 2000 | 0.0003 | - |
177
+ | 0.6673 | 2050 | 0.0003 | - |
178
+ | 0.6836 | 2100 | 0.0003 | - |
179
+ | 0.6999 | 2150 | 0.0003 | - |
180
+ | 0.7161 | 2200 | 0.0003 | - |
181
+ | 0.7324 | 2250 | 0.0003 | - |
182
+ | 0.7487 | 2300 | 0.0003 | - |
183
+ | 0.7650 | 2350 | 0.0003 | - |
184
+ | 0.7812 | 2400 | 0.0003 | - |
185
+ | 0.7975 | 2450 | 0.0003 | - |
186
+ | 0.8138 | 2500 | 0.0002 | - |
187
+ | 0.8301 | 2550 | 0.0002 | - |
188
+ | 0.8464 | 2600 | 0.0003 | - |
189
+ | 0.8626 | 2650 | 0.0002 | - |
190
+ | 0.8789 | 2700 | 0.0002 | - |
191
+ | 0.8952 | 2750 | 0.0002 | - |
192
+ | 0.9115 | 2800 | 0.0003 | - |
193
+ | 0.9277 | 2850 | 0.0002 | - |
194
+ | 0.9440 | 2900 | 0.0002 | - |
195
+ | 0.9603 | 2950 | 0.0002 | - |
196
+ | 0.9766 | 3000 | 0.0002 | - |
197
+ | 0.9928 | 3050 | 0.0002 | - |
198
 
199
  ### Framework Versions
200
  - Python: 3.11.6
config_setfit.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "normalize_embeddings": false,
3
  "labels": [
4
  "assertion",
5
  "expressive",
6
  "question",
7
  "request"
8
- ]
 
9
  }
 
1
  {
 
2
  "labels": [
3
  "assertion",
4
  "expressive",
5
  "question",
6
  "request"
7
+ ],
8
+ "normalize_embeddings": false
9
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ebdd01ad74d0ff9b7329ecdd47b13c4770a68ef05e3e04a474fe3e029c67aeb7
3
  size 90864192
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:39a4e71b3ba068c43a85886c12db3b6694290671f278529ee2b9314d6e1fddcd
3
  size 90864192
model_head.pkl CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9b8c2c52e3123a1b7907023d6de7549ca37121fedd2e2119f49ff9f17700835c
3
  size 7159
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f801a60e79115c275b21fc1d116c853079e09f3ba8405ccd74e5b3ad4ddce93
3
  size 7159