File size: 3,841 Bytes
d8f974d 30fe23f d8f974d 30fe23f d8f974d 7fa1729 3783211 7fa1729 30fe23f 7fa1729 d8f974d 30fe23f d8f974d |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 |
---
base_model: DCAMA
language: en
license: mit
tags:
- few-shot segmentation
- distillation
- image-segmentation
name: DistillFSS-DCAMA
library: pytorch
ArXiv: '2512.05613'
repo_url: https://github.com/pasqualedem/DistillFSS
paper_url: https://arxiv.org/abs/2512.05613
parameters: "dataloader:\n num_workers: 0\ndataset:\n datasets:\n test_weedmap:\n\
\ prompt_images: 5\n test_root: data/weedmap/0_rotations_processed_003_test/RedEdge/003\n\
\ train_root: data/weedmap/0_rotations_processed_003_test/RedEdge/000\n preprocess:\n\
\ image_size: 384\n mean:\n - 0.485\n - 0.456\n - 0.406\n std:\n\
\ - 0.229\n - 0.224\n - 0.225\nmodel:\n name: distillator\n params:\n\
\ student:\n name: conv_distillator\n num_classes: 2\n teacher:\n\
\ backbone: swin\n backbone_checkpoint: checkpoints/swin_base_patch4_window12_384.pth\n\
\ concat_support: false\n image_size: 384\n model_checkpoint: checkpoints/swin_fold0_pascal_modcross_soft.pt\n\
\ name: dcama\npush_to_hub:\n repo_name: pasqualedem/DistillFSS_WeedMap_DCAMA_5shot\n\
refinement:\n hot_parameters:\n - model.conv1\n - model.conv2\n - model.conv3\n\
\ - model.mixer1\n - model.mixer2\n - model.mixer3\n - student\n iterations_is_num_classes:\
\ false\n loss:\n name: refine_distill\n lr: 0.001\n max_iterations: 500\n\
\ subsample: 1\n substitutor: paired\ntest:\n prompt_to_use: null\ntracker:\n\
\ cache_dir: tmp\n group: WeedMap\n log_frequency: 1\n project: FSSWeed\n tags:\n\
\ - WeedMap\n - Distill\n test_image_log_frequency: 10\n tmp_dir: tmp\n train_image_log_frequency:\
\ 25\n"
repo_id: pasqualedem/DistillFSS_WeedMap_DCAMA_5shot
---
DistillFSS-DCAMA is a distilled version of the DCAMA model for a specific downstream segmentation task. The DistillFSS framework allows to distill large few-shot segmentation models into smaller and more efficient ones, while improving or maintaining their performance on the target task.
- Code: https://github.com/pasqualedem/DistillFSS
- Paper: https://arxiv.org/abs/2512.05613
How to use this model:
Clone the repository:
```bash
git clone https://github.com/pasqualedem/DistillFSS.git
```
Install the required dependencies as specified in the repository.
Load the model using the following code snippet:
```python
from distillfss.models.dcama.distillator import DistilledDCAMA
model = DistilledDCAMA.from_pretrained("pasqualedem/DistillFSS_WeedMap_DCAMA_5shot")
```
YAML configuration:
```yaml
dataloader:
num_workers: 0
dataset:
datasets:
test_weedmap:
prompt_images: 5
test_root: data/weedmap/0_rotations_processed_003_test/RedEdge/003
train_root: data/weedmap/0_rotations_processed_003_test/RedEdge/000
preprocess:
image_size: 384
mean:
- 0.485
- 0.456
- 0.406
std:
- 0.229
- 0.224
- 0.225
model:
name: distillator
params:
student:
name: conv_distillator
num_classes: 2
teacher:
backbone: swin
backbone_checkpoint: checkpoints/swin_base_patch4_window12_384.pth
concat_support: false
image_size: 384
model_checkpoint: checkpoints/swin_fold0_pascal_modcross_soft.pt
name: dcama
push_to_hub:
repo_name: pasqualedem/DistillFSS_WeedMap_DCAMA_5shot
refinement:
hot_parameters:
- model.conv1
- model.conv2
- model.conv3
- model.mixer1
- model.mixer2
- model.mixer3
- student
iterations_is_num_classes: false
loss:
name: refine_distill
lr: 0.001
max_iterations: 500
subsample: 1
substitutor: paired
test:
prompt_to_use: null
tracker:
cache_dir: tmp
group: WeedMap
log_frequency: 1
project: FSSWeed
tags:
- WeedMap
- Distill
test_image_log_frequency: 10
tmp_dir: tmp
train_image_log_frequency: 25
``` |