pasqualedem commited on
Commit
30fe23f
·
verified ·
1 Parent(s): 68de633

Push model using huggingface_hub.

Browse files
Files changed (2) hide show
  1. README.md +4 -4
  2. model.safetensors +1 -1
README.md CHANGED
@@ -19,7 +19,7 @@ parameters: "dataloader:\n num_workers: 0\ndataset:\n datasets:\n test_weed
19
  \ student:\n name: conv_distillator\n num_classes: 2\n teacher:\n\
20
  \ backbone: swin\n backbone_checkpoint: checkpoints/swin_base_patch4_window12_384.pth\n\
21
  \ concat_support: false\n image_size: 384\n model_checkpoint: checkpoints/swin_fold0_pascal_modcross_soft.pt\n\
22
- \ name: dcama\npush_to_hub:\n repo_name: DistillFSS_WeedMap_DCAMA_5shot\n\
23
  refinement:\n hot_parameters:\n - model.conv1\n - model.conv2\n - model.conv3\n\
24
  \ - model.mixer1\n - model.mixer2\n - model.mixer3\n - student\n iterations_is_num_classes:\
25
  \ false\n loss:\n name: refine_distill\n lr: 0.001\n max_iterations: 500\n\
@@ -27,7 +27,7 @@ parameters: "dataloader:\n num_workers: 0\ndataset:\n datasets:\n test_weed
27
  \ cache_dir: tmp\n group: WeedMap\n log_frequency: 1\n project: FSSWeed\n tags:\n\
28
  \ - WeedMap\n - Distill\n test_image_log_frequency: 10\n tmp_dir: tmp\n train_image_log_frequency:\
29
  \ 25\n"
30
- repo_id: DistillFSS_WeedMap_DCAMA_5shot
31
  ---
32
 
33
  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.
@@ -45,7 +45,7 @@ Install the required dependencies as specified in the repository.
45
  Load the model using the following code snippet:
46
  ```python
47
  from distillfss.models.dcama.distillator import DistilledDCAMA
48
- model = DistilledDCAMA.from_pretrained("DistillFSS_WeedMap_DCAMA_5shot")
49
  ```
50
 
51
  YAML configuration:
@@ -82,7 +82,7 @@ model:
82
  model_checkpoint: checkpoints/swin_fold0_pascal_modcross_soft.pt
83
  name: dcama
84
  push_to_hub:
85
- repo_name: DistillFSS_WeedMap_DCAMA_5shot
86
  refinement:
87
  hot_parameters:
88
  - model.conv1
 
19
  \ student:\n name: conv_distillator\n num_classes: 2\n teacher:\n\
20
  \ backbone: swin\n backbone_checkpoint: checkpoints/swin_base_patch4_window12_384.pth\n\
21
  \ concat_support: false\n image_size: 384\n model_checkpoint: checkpoints/swin_fold0_pascal_modcross_soft.pt\n\
22
+ \ name: dcama\npush_to_hub:\n repo_name: pasqualedem/DistillFSS_WeedMap_DCAMA_5shot\n\
23
  refinement:\n hot_parameters:\n - model.conv1\n - model.conv2\n - model.conv3\n\
24
  \ - model.mixer1\n - model.mixer2\n - model.mixer3\n - student\n iterations_is_num_classes:\
25
  \ false\n loss:\n name: refine_distill\n lr: 0.001\n max_iterations: 500\n\
 
27
  \ cache_dir: tmp\n group: WeedMap\n log_frequency: 1\n project: FSSWeed\n tags:\n\
28
  \ - WeedMap\n - Distill\n test_image_log_frequency: 10\n tmp_dir: tmp\n train_image_log_frequency:\
29
  \ 25\n"
30
+ repo_id: pasqualedem/DistillFSS_WeedMap_DCAMA_5shot
31
  ---
32
 
33
  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.
 
45
  Load the model using the following code snippet:
46
  ```python
47
  from distillfss.models.dcama.distillator import DistilledDCAMA
48
+ model = DistilledDCAMA.from_pretrained("pasqualedem/DistillFSS_WeedMap_DCAMA_5shot")
49
  ```
50
 
51
  YAML configuration:
 
82
  model_checkpoint: checkpoints/swin_fold0_pascal_modcross_soft.pt
83
  name: dcama
84
  push_to_hub:
85
+ repo_name: pasqualedem/DistillFSS_WeedMap_DCAMA_5shot
86
  refinement:
87
  hot_parameters:
88
  - model.conv1
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b99613d1a918113c8c4e6bc80da19d80dc56e90f4ac74283d1773ad650c0192e
3
  size 554599424
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4d83932e82dd8ab0fd8d808efc9a29f7c89ec96d0c9803e4e3659f565719e89
3
  size 554599424