patrickfleith
commited on
Commit
•
d0c7ebf
1
Parent(s):
df401c4
Push model using huggingface_hub.
Browse files- README.md +2 -81
- config.json +1 -1
- tokenizer_config.json +7 -0
README.md
CHANGED
@@ -9,36 +9,8 @@ tags:
|
|
9 |
- sentence-transformers
|
10 |
- text-classification
|
11 |
- generated_from_setfit_trainer
|
12 |
-
widget:
|
13 |
-
- text: Deep throttling capability, which allows a rocket engine to vary its thrust
|
14 |
-
over a wide range, is essential for applications requiring precise landing maneuvers,
|
15 |
-
such as lunar landers.
|
16 |
-
- text: The use of staged combustion cycles, such as the full-flow staged combustion
|
17 |
-
cycle, can enhance the performance of liquid rocket engines by utilizing propellants
|
18 |
-
more efficiently.
|
19 |
-
- text: The satellite's power budget, which balances generation, storage, and consumption,
|
20 |
-
is meticulously planned to ensure that all systems remain operational throughout
|
21 |
-
the mission duration.
|
22 |
-
- text: The thrust chamber is a critical component where the combustion of propellants
|
23 |
-
occurs, generating high-pressure and high-temperature exhaust gases.
|
24 |
-
- text: The electrical generation capability of a satellite is primarily determined
|
25 |
-
by the efficiency and surface area of its photovoltaic cells, which convert incident
|
26 |
-
solar radiation into electrical energy.
|
27 |
inference: true
|
28 |
-
model-index:
|
29 |
-
- name: SetFit with BAAI/bge-small-en-v1.5
|
30 |
-
results:
|
31 |
-
- task:
|
32 |
-
type: text-classification
|
33 |
-
name: Text Classification
|
34 |
-
dataset:
|
35 |
-
name: Unknown
|
36 |
-
type: unknown
|
37 |
-
split: test
|
38 |
-
metrics:
|
39 |
-
- type: accuracy
|
40 |
-
value: 1.0
|
41 |
-
name: Accuracy
|
42 |
---
|
43 |
|
44 |
# SetFit with BAAI/bge-small-en-v1.5
|
@@ -68,20 +40,6 @@ The model has been trained using an efficient few-shot learning technique that i
|
|
68 |
- **Paper:** [Efficient Few-Shot Learning Without Prompts](https://arxiv.org/abs/2209.11055)
|
69 |
- **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
|
70 |
|
71 |
-
### Model Labels
|
72 |
-
| Label | Examples |
|
73 |
-
|:----------------|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
|
74 |
-
| Thermal Control | <ul><li>'The satellite thermal control subsystem (TCS) is crucial for maintaining operational temperatures of all onboard instruments and systems within their specified limits.'</li><li>'Phase change materials (PCMs) are employed in some satellite TCS designs to absorb and release thermal energy, stabilizing temperature fluctuations during orbital transitions.'</li><li>'Describe the process and importance of using computational fluid dynamics (CFD) in analyzing spacecraft thermal environments.'</li></ul> |
|
75 |
-
| Power Subsystem | <ul><li>'Electromagnetic interference (EMI) shielding and grounding techniques are essential in satellite design to prevent power system noise from affecting sensitive communication and navigation subsystems.'</li><li>'The structural integrity of solar panels must be robust enough to withstand the mechanical stresses of launch and the harsh thermal cycles of the space environment.'</li><li>'The integration of maximum power point tracking (MPPT) technology enhances the efficiency of solar arrays by dynamically adjusting the load to match the optimal power output of the photovoltaic cells.'</li></ul> |
|
76 |
-
| Propulsion | <ul><li>'The efficiency of a rocket engine is primarily determined by its specific impulse (Isp), which measures the thrust produced per unit of propellant consumed.'</li><li>'Liquid rocket engines utilize cryogenic fuels and oxidizers, such as liquid hydrogen and liquid oxygen, which require complex storage and handling systems to maintain their extremely low temperatures.'</li><li>"Rocket engines operate on the principle of Newton's Third Law of Motion, where the expulsion of high-speed exhaust gases produces a reaction force that propels the rocket forward."</li></ul> |
|
77 |
-
|
78 |
-
## Evaluation
|
79 |
-
|
80 |
-
### Metrics
|
81 |
-
| Label | Accuracy |
|
82 |
-
|:--------|:---------|
|
83 |
-
| **all** | 1.0 |
|
84 |
-
|
85 |
## Uses
|
86 |
|
87 |
### Direct Use for Inference
|
@@ -100,7 +58,7 @@ from setfit import SetFitModel
|
|
100 |
# Download from the 🤗 Hub
|
101 |
model = SetFitModel.from_pretrained("patrickfleith/my-awesome-astro-text-classifier")
|
102 |
# Run inference
|
103 |
-
preds = model("
|
104 |
```
|
105 |
|
106 |
<!--
|
@@ -129,43 +87,6 @@ preds = model("The thrust chamber is a critical component where the combustion o
|
|
129 |
|
130 |
## Training Details
|
131 |
|
132 |
-
### Training Set Metrics
|
133 |
-
| Training set | Min | Median | Max |
|
134 |
-
|:-------------|:----|:--------|:----|
|
135 |
-
| Word count | 11 | 23.6176 | 30 |
|
136 |
-
|
137 |
-
| Label | Training Sample Count |
|
138 |
-
|:----------------|:----------------------|
|
139 |
-
| Propulsion | 13 |
|
140 |
-
| Thermal Control | 10 |
|
141 |
-
| Power Subsystem | 11 |
|
142 |
-
|
143 |
-
### Training Hyperparameters
|
144 |
-
- batch_size: (32, 32)
|
145 |
-
- num_epochs: (10, 10)
|
146 |
-
- max_steps: -1
|
147 |
-
- sampling_strategy: oversampling
|
148 |
-
- body_learning_rate: (2e-05, 1e-05)
|
149 |
-
- head_learning_rate: 0.01
|
150 |
-
- loss: CosineSimilarityLoss
|
151 |
-
- distance_metric: cosine_distance
|
152 |
-
- margin: 0.25
|
153 |
-
- end_to_end: False
|
154 |
-
- use_amp: False
|
155 |
-
- warmup_proportion: 0.1
|
156 |
-
- seed: 42
|
157 |
-
- eval_max_steps: -1
|
158 |
-
- load_best_model_at_end: False
|
159 |
-
|
160 |
-
### Training Results
|
161 |
-
| Epoch | Step | Training Loss | Validation Loss |
|
162 |
-
|:------:|:----:|:-------------:|:---------------:|
|
163 |
-
| 0.0417 | 1 | 0.2181 | - |
|
164 |
-
| 2.0833 | 50 | 0.0145 | - |
|
165 |
-
| 4.1667 | 100 | 0.0047 | - |
|
166 |
-
| 6.25 | 150 | 0.0035 | - |
|
167 |
-
| 8.3333 | 200 | 0.0028 | - |
|
168 |
-
|
169 |
### Framework Versions
|
170 |
- Python: 3.10.12
|
171 |
- SetFit: 1.0.3
|
|
|
9 |
- sentence-transformers
|
10 |
- text-classification
|
11 |
- generated_from_setfit_trainer
|
12 |
+
widget: []
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
13 |
inference: true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
14 |
---
|
15 |
|
16 |
# SetFit with BAAI/bge-small-en-v1.5
|
|
|
40 |
- **Paper:** [Efficient Few-Shot Learning Without Prompts](https://arxiv.org/abs/2209.11055)
|
41 |
- **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
|
42 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
43 |
## Uses
|
44 |
|
45 |
### Direct Use for Inference
|
|
|
58 |
# Download from the 🤗 Hub
|
59 |
model = SetFitModel.from_pretrained("patrickfleith/my-awesome-astro-text-classifier")
|
60 |
# Run inference
|
61 |
+
preds = model("I loved the spiderman movie!")
|
62 |
```
|
63 |
|
64 |
<!--
|
|
|
87 |
|
88 |
## Training Details
|
89 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
90 |
### Framework Versions
|
91 |
- Python: 3.10.12
|
92 |
- SetFit: 1.0.3
|
config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "
|
3 |
"architectures": [
|
4 |
"BertModel"
|
5 |
],
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "patrickfleith/my-awesome-astro-text-classifier",
|
3 |
"architectures": [
|
4 |
"BertModel"
|
5 |
],
|
tokenizer_config.json
CHANGED
@@ -46,12 +46,19 @@
|
|
46 |
"do_basic_tokenize": true,
|
47 |
"do_lower_case": true,
|
48 |
"mask_token": "[MASK]",
|
|
|
49 |
"model_max_length": 512,
|
50 |
"never_split": null,
|
|
|
51 |
"pad_token": "[PAD]",
|
|
|
|
|
52 |
"sep_token": "[SEP]",
|
|
|
53 |
"strip_accents": null,
|
54 |
"tokenize_chinese_chars": true,
|
55 |
"tokenizer_class": "BertTokenizer",
|
|
|
|
|
56 |
"unk_token": "[UNK]"
|
57 |
}
|
|
|
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 |
}
|