Update README.md
Browse files
README.md
CHANGED
@@ -45,19 +45,14 @@ This model was trained within the context of a larger system for ABSA, which loo
|
|
45 |
- **Sentence Transformer body:** [sentence-transformers/all-mpnet-base-v2](https://huggingface.co/sentence-transformers/all-mpnet-base-v2)
|
46 |
- **Classification head:** a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance
|
47 |
- **spaCy Model:** en_core_web_sm
|
48 |
-
- **SetFitABSA Aspect Model:** [/
|
49 |
-
- **SetFitABSA Polarity Model:** [/
|
50 |
- **Maximum Sequence Length:** 384 tokens
|
51 |
- **Number of Classes:** 3 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 |
|
@@ -83,11 +78,11 @@ from setfit import AbsaModel
|
|
83 |
|
84 |
# Download from the 🤗 Hub
|
85 |
model = AbsaModel.from_pretrained(
|
86 |
-
"/
|
87 |
-
"/
|
88 |
)
|
89 |
# Run inference
|
90 |
-
preds = model("
|
91 |
```
|
92 |
|
93 |
<!--
|
@@ -128,11 +123,11 @@ preds = model("The food was great, but the venue is just way too busy.")
|
|
128 |
| positive | 703 |
|
129 |
|
130 |
### Training Hyperparameters
|
131 |
-
- batch_size:
|
132 |
-
- num_epochs:
|
133 |
- max_steps: -1
|
134 |
- sampling_strategy: oversampling
|
135 |
-
- body_learning_rate:
|
136 |
- head_learning_rate: 0.01
|
137 |
- loss: CosineSimilarityLoss
|
138 |
- distance_metric: cosine_distance
|
@@ -145,28 +140,6 @@ preds = model("The food was great, but the venue is just way too busy.")
|
|
145 |
- eval_max_steps: -1
|
146 |
- load_best_model_at_end: True
|
147 |
|
148 |
-
### Training Results
|
149 |
-
| Epoch | Step | Training Loss | Validation Loss |
|
150 |
-
|:------:|:----:|:-------------:|:---------------:|
|
151 |
-
| 0.0000 | 1 | 0.2969 | - |
|
152 |
-
| 0.0025 | 50 | 0.3201 | - |
|
153 |
-
| 0.0049 | 100 | 0.2933 | 0.2839 |
|
154 |
-
| 0.0074 | 150 | 0.262 | - |
|
155 |
-
| 0.0098 | 200 | 0.2523 | 0.2480 |
|
156 |
-
| 0.0123 | 250 | 0.2403 | - |
|
157 |
-
| 0.0147 | 300 | 0.2185 | 0.2199 |
|
158 |
-
| 0.0172 | 350 | 0.1983 | - |
|
159 |
-
| 0.0196 | 400 | 0.1874 | 0.2003 |
|
160 |
-
| 0.0221 | 450 | 0.1727 | - |
|
161 |
-
| 0.0245 | 500 | 0.1568 | 0.1882 |
|
162 |
-
| 0.0270 | 550 | 0.1386 | - |
|
163 |
-
| 0.0294 | 600 | 0.1181 | 0.1742 |
|
164 |
-
| 0.0319 | 650 | 0.1023 | - |
|
165 |
-
| 0.0343 | 700 | 0.0877 | 0.1766 |
|
166 |
-
| 0.0368 | 750 | 0.0717 | - |
|
167 |
-
| 0.0392 | 800 | 0.0555 | 0.1854 |
|
168 |
-
| 0.0417 | 850 | 0.0447 | - |
|
169 |
-
| 0.0441 | 900 | 0.0343 | 0.1841 |
|
170 |
|
171 |
### Framework Versions
|
172 |
- Python: 3.11.11
|
|
|
45 |
- **Sentence Transformer body:** [sentence-transformers/all-mpnet-base-v2](https://huggingface.co/sentence-transformers/all-mpnet-base-v2)
|
46 |
- **Classification head:** a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance
|
47 |
- **spaCy Model:** en_core_web_sm
|
48 |
+
- **SetFitABSA Aspect Model:** [/Askinkaty/setfit-finance-aspect](https://huggingface.co/Askinkaty/setfit-finance-aspect)
|
49 |
+
- **SetFitABSA Polarity Model:** [/Askinkaty/setfit-finance-polarity](https://huggingface.co/Askinkaty/setfit-finance-polarity)
|
50 |
- **Maximum Sequence Length:** 384 tokens
|
51 |
- **Number of Classes:** 3 classes
|
52 |
<!-- - **Training Dataset:** [Unknown](https://huggingface.co/datasets/unknown) -->
|
53 |
<!-- - **Language:** Unknown -->
|
54 |
<!-- - **License:** Unknown -->
|
55 |
|
|
|
|
|
|
|
|
|
|
|
56 |
|
57 |
### Model Labels
|
58 |
| Label | Examples |
|
|
|
78 |
|
79 |
# Download from the 🤗 Hub
|
80 |
model = AbsaModel.from_pretrained(
|
81 |
+
"Askinkaty/setfit-finance-aspect",
|
82 |
+
"Askinkaty/setfit-finance-polarity",
|
83 |
)
|
84 |
# Run inference
|
85 |
+
preds = model("Banking stocks to see lot of traction: Mitesh Thacker.")
|
86 |
```
|
87 |
|
88 |
<!--
|
|
|
123 |
| positive | 703 |
|
124 |
|
125 |
### Training Hyperparameters
|
126 |
+
- batch_size: 64
|
127 |
+
- num_epochs: 2
|
128 |
- max_steps: -1
|
129 |
- sampling_strategy: oversampling
|
130 |
+
- body_learning_rate: 1e-05
|
131 |
- head_learning_rate: 0.01
|
132 |
- loss: CosineSimilarityLoss
|
133 |
- distance_metric: cosine_distance
|
|
|
140 |
- eval_max_steps: -1
|
141 |
- load_best_model_at_end: True
|
142 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
143 |
|
144 |
### Framework Versions
|
145 |
- Python: 3.11.11
|