vipinbansal179
commited on
Commit
•
f70f6ac
1
Parent(s):
6b599de
Add SetFit model
Browse files- README.md +20 -45
- config.json +1 -1
- model.safetensors +1 -1
- model_head.pkl +1 -1
README.md
CHANGED
@@ -32,7 +32,7 @@ model-index:
|
|
32 |
split: test
|
33 |
metrics:
|
34 |
- type: accuracy
|
35 |
-
value: 0.
|
36 |
name: Accuracy
|
37 |
---
|
38 |
|
@@ -64,18 +64,18 @@ The model has been trained using an efficient few-shot learning technique that i
|
|
64 |
- **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
|
65 |
|
66 |
### Model Labels
|
67 |
-
| Label | Examples
|
68 |
-
|
69 |
-
| 2 | <ul><li>'
|
70 |
-
| 0 | <ul><li>'rs 260.00 debit a/c xxxxxx7783 credit krjngm @ oksbi upi ref:325154274303. ? call 18005700 -bob'</li><li>'
|
71 |
-
| 1 | <ul><li>'
|
72 |
|
73 |
## Evaluation
|
74 |
|
75 |
### Metrics
|
76 |
| Label | Accuracy |
|
77 |
|:--------|:---------|
|
78 |
-
| **all** | 0.
|
79 |
|
80 |
## Uses
|
81 |
|
@@ -127,17 +127,17 @@ preds = model("< # > use otp : 8233 login turtlemintpro zck+rfoaqnm")
|
|
127 |
### Training Set Metrics
|
128 |
| Training set | Min | Median | Max |
|
129 |
|:-------------|:----|:--------|:----|
|
130 |
-
| Word count |
|
131 |
|
132 |
| Label | Training Sample Count |
|
133 |
|:------|:----------------------|
|
134 |
-
| 0 |
|
135 |
-
| 1 |
|
136 |
-
| 2 |
|
137 |
|
138 |
### Training Hyperparameters
|
139 |
- batch_size: (16, 16)
|
140 |
-
- num_epochs: (
|
141 |
- max_steps: -1
|
142 |
- sampling_strategy: oversampling
|
143 |
- body_learning_rate: (2e-05, 1e-05)
|
@@ -153,39 +153,14 @@ preds = model("< # > use otp : 8233 login turtlemintpro zck+rfoaqnm")
|
|
153 |
- load_best_model_at_end: True
|
154 |
|
155 |
### Training Results
|
156 |
-
| Epoch | Step
|
157 |
-
|
158 |
-
| 0.
|
159 |
-
| 0
|
160 |
-
| 0
|
161 |
-
|
|
162 |
-
| 0
|
163 |
-
| 0
|
164 |
-
| 0.4249 | 300 | 0.0003 | - |
|
165 |
-
| 0.4958 | 350 | 0.0001 | - |
|
166 |
-
| 0.5666 | 400 | 0.0001 | - |
|
167 |
-
| 0.6374 | 450 | 0.0001 | - |
|
168 |
-
| 0.7082 | 500 | 0.0001 | - |
|
169 |
-
| 0.7790 | 550 | 0.0001 | - |
|
170 |
-
| 0.8499 | 600 | 0.0002 | - |
|
171 |
-
| 0.9207 | 650 | 0.0001 | - |
|
172 |
-
| 0.9915 | 700 | 0.0001 | - |
|
173 |
-
| **1.0** | **706** | **-** | **0.0168** |
|
174 |
-
| 1.0623 | 750 | 0.0 | - |
|
175 |
-
| 1.1331 | 800 | 0.0001 | - |
|
176 |
-
| 1.2040 | 850 | 0.0001 | - |
|
177 |
-
| 1.2748 | 900 | 0.0 | - |
|
178 |
-
| 1.3456 | 950 | 0.0001 | - |
|
179 |
-
| 1.4164 | 1000 | 0.0 | - |
|
180 |
-
| 1.4873 | 1050 | 0.0 | - |
|
181 |
-
| 1.5581 | 1100 | 0.0 | - |
|
182 |
-
| 1.6289 | 1150 | 0.0 | - |
|
183 |
-
| 1.6997 | 1200 | 0.0 | - |
|
184 |
-
| 1.7705 | 1250 | 0.0 | - |
|
185 |
-
| 1.8414 | 1300 | 0.0002 | - |
|
186 |
-
| 1.9122 | 1350 | 0.0 | - |
|
187 |
-
| 1.9830 | 1400 | 0.0 | - |
|
188 |
-
| 2.0 | 1412 | - | 0.0183 |
|
189 |
|
190 |
* The bold row denotes the saved checkpoint.
|
191 |
### Framework Versions
|
|
|
32 |
split: test
|
33 |
metrics:
|
34 |
- type: accuracy
|
35 |
+
value: 0.6944444444444444
|
36 |
name: Accuracy
|
37 |
---
|
38 |
|
|
|
64 |
- **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
|
65 |
|
66 |
### Model Labels
|
67 |
+
| Label | Examples |
|
68 |
+
|:------|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
|
69 |
+
| 2 | <ul><li>'hi ashwin , credit score qualify personal loan rs.10 lacs . apply get 10 mins https : //tneu.in/qilih6y4 & c apply - tataneu'</li><li>'592104 otp transaction inr 589.00 kotak bank card 6343 airtel , valid 3 mins . share otp anyone .'</li><li>'dear customer , get axis vistara credit card , complimentary economy ticket join & upto 4 ticket yearly , apply : https : //1kx.in/xrb8ksi6tkf loanmandi'</li></ul> |
|
70 |
+
| 0 | <ul><li>"'ve spend rs.17220.00 thru kotak bank debit card xx2169 inf * www.vfsglobal.com 12/10/2023 avl bal 196168.62 ? visit kotak.com/fraud"</li><li>'rs 260.00 debit a/c xxxxxx7783 credit krjngm @ oksbi upi ref:325154274303. ? call 18005700 -bob'</li><li>'rs.381.56 transfer a/c ... 7783 : upi/327734499842 . total bal : rs.27615.37cr . avlbl amt : rs.27615.37 ( 04-10-2023 19:07:22 ) - bank baroda'</li></ul> |
|
71 |
+
| 1 | <ul><li>'update ! inr5.66 credit federal bank account xxxx9374 jupiter app . happy bank !'</li><li>'dear customer , inr 619.00 credit a/c xx7556 31/08/2023 neft utr cms3535305583 fintech blue solutions private limit , info : fintech blue solutions private limi ted-sbi'</li></ul> |
|
72 |
|
73 |
## Evaluation
|
74 |
|
75 |
### Metrics
|
76 |
| Label | Accuracy |
|
77 |
|:--------|:---------|
|
78 |
+
| **all** | 0.6944 |
|
79 |
|
80 |
## Uses
|
81 |
|
|
|
127 |
### Training Set Metrics
|
128 |
| Training set | Min | Median | Max |
|
129 |
|:-------------|:----|:--------|:----|
|
130 |
+
| Word count | 14 | 22.0417 | 30 |
|
131 |
|
132 |
| Label | Training Sample Count |
|
133 |
|:------|:----------------------|
|
134 |
+
| 0 | 6 |
|
135 |
+
| 1 | 2 |
|
136 |
+
| 2 | 16 |
|
137 |
|
138 |
### Training Hyperparameters
|
139 |
- batch_size: (16, 16)
|
140 |
+
- num_epochs: (4, 4)
|
141 |
- max_steps: -1
|
142 |
- sampling_strategy: oversampling
|
143 |
- body_learning_rate: (2e-05, 1e-05)
|
|
|
153 |
- load_best_model_at_end: True
|
154 |
|
155 |
### Training Results
|
156 |
+
| Epoch | Step | Training Loss | Validation Loss |
|
157 |
+
|:-------:|:------:|:-------------:|:---------------:|
|
158 |
+
| 0.05 | 1 | 0.279 | - |
|
159 |
+
| **1.0** | **20** | **-** | **0.1602** |
|
160 |
+
| 2.0 | 40 | - | 0.1603 |
|
161 |
+
| 2.5 | 50 | 0.0013 | - |
|
162 |
+
| 3.0 | 60 | - | 0.1661 |
|
163 |
+
| 4.0 | 80 | - | 0.1706 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
164 |
|
165 |
* The bold row denotes the saved checkpoint.
|
166 |
### Framework Versions
|
config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "checkpoints/
|
3 |
"architectures": [
|
4 |
"MPNetModel"
|
5 |
],
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "checkpoints/step_20/",
|
3 |
"architectures": [
|
4 |
"MPNetModel"
|
5 |
],
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 437967672
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fc686d4ee9727ab20a11e61fec42072f6a2026a08427a68d92658f5c1ae2e37d
|
3 |
size 437967672
|
model_head.pkl
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 19311
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:84d61b69473b85170ab65e6b7e07704d6a59f5e03637221cab5346fa22a834fe
|
3 |
size 19311
|