Add model cards for test_holdout (batch 3)
Browse files- test_holdout/model_idx_4599/README.md +40 -0
- test_holdout/model_idx_4600/README.md +40 -0
- test_holdout/model_idx_4601/README.md +40 -0
- test_holdout/model_idx_4602/README.md +40 -0
- test_holdout/model_idx_4603/README.md +40 -0
- test_holdout/model_idx_4604/README.md +40 -0
- test_holdout/model_idx_4605/README.md +40 -0
- test_holdout/model_idx_4606/README.md +40 -0
- test_holdout/model_idx_4607/README.md +40 -0
- test_holdout/model_idx_4608/README.md +40 -0
- test_holdout/model_idx_4609/README.md +40 -0
- test_holdout/model_idx_4610/README.md +40 -0
- test_holdout/model_idx_4611/README.md +40 -0
- test_holdout/model_idx_4612/README.md +40 -0
- test_holdout/model_idx_4613/README.md +40 -0
- test_holdout/model_idx_4614/README.md +40 -0
- test_holdout/model_idx_4615/README.md +40 -0
- test_holdout/model_idx_4616/README.md +40 -0
- test_holdout/model_idx_4617/README.md +40 -0
- test_holdout/model_idx_4618/README.md +40 -0
- test_holdout/model_idx_4619/README.md +40 -0
- test_holdout/model_idx_4620/README.md +40 -0
- test_holdout/model_idx_4621/README.md +40 -0
- test_holdout/model_idx_4622/README.md +40 -0
- test_holdout/model_idx_4623/README.md +40 -0
- test_holdout/model_idx_4624/README.md +40 -0
- test_holdout/model_idx_4625/README.md +40 -0
- test_holdout/model_idx_4626/README.md +40 -0
- test_holdout/model_idx_4627/README.md +40 -0
- test_holdout/model_idx_4628/README.md +40 -0
- test_holdout/model_idx_4629/README.md +40 -0
- test_holdout/model_idx_4630/README.md +40 -0
- test_holdout/model_idx_4631/README.md +40 -0
- test_holdout/model_idx_4632/README.md +40 -0
- test_holdout/model_idx_4633/README.md +40 -0
- test_holdout/model_idx_4634/README.md +40 -0
- test_holdout/model_idx_4635/README.md +40 -0
- test_holdout/model_idx_4636/README.md +40 -0
- test_holdout/model_idx_4637/README.md +40 -0
- test_holdout/model_idx_4638/README.md +40 -0
- test_holdout/model_idx_4639/README.md +40 -0
- test_holdout/model_idx_4640/README.md +40 -0
- test_holdout/model_idx_4641/README.md +40 -0
- test_holdout/model_idx_4642/README.md +40 -0
- test_holdout/model_idx_4643/README.md +40 -0
- test_holdout/model_idx_4644/README.md +40 -0
- test_holdout/model_idx_4645/README.md +40 -0
- test_holdout/model_idx_4646/README.md +40 -0
- test_holdout/model_idx_4647/README.md +40 -0
- test_holdout/model_idx_4648/README.md +40 -0
test_holdout/model_idx_4599/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-4
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4599
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-4` |
|
| 30 |
+
| **ImageNet Class** | palace (n03877845) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 9e-05 |
|
| 37 |
+
| Max Train Steps | 700 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 18 |
|
| 40 |
+
| N Training Samples | 7 |
|
test_holdout/model_idx_4600/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: stable-diffusion-v1-5/stable-diffusion-v1-5
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4600
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `stable-diffusion-v1-5/stable-diffusion-v1-5` |
|
| 30 |
+
| **ImageNet Class** | black-footed_ferret (n02443484) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0005 |
|
| 37 |
+
| Max Train Steps | 700 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 17 |
|
| 40 |
+
| N Training Samples | 8 |
|
test_holdout/model_idx_4601/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-4
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4601
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-4` |
|
| 30 |
+
| **ImageNet Class** | chainlink_fence (n03000134) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 5e-05 |
|
| 37 |
+
| Max Train Steps | 650 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 33 |
|
| 40 |
+
| N Training Samples | 5 |
|
test_holdout/model_idx_4602/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-3
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4602
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-3` |
|
| 30 |
+
| **ImageNet Class** | Yorkshire_terrier (n02094433) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0005 |
|
| 37 |
+
| Max Train Steps | 500 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 4 |
|
| 40 |
+
| N Training Samples | 10 |
|
test_holdout/model_idx_4603/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-4
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4603
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-4` |
|
| 30 |
+
| **ImageNet Class** | black_and_gold_garden_spider (n01773157) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 3e-05 |
|
| 37 |
+
| Max Train Steps | 700 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 52 |
|
| 40 |
+
| N Training Samples | 7 |
|
test_holdout/model_idx_4604/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: stable-diffusion-v1-5/stable-diffusion-v1-5
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4604
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `stable-diffusion-v1-5/stable-diffusion-v1-5` |
|
| 30 |
+
| **ImageNet Class** | Dungeness_crab (n01978287) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0001 |
|
| 37 |
+
| Max Train Steps | 600 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 22 |
|
| 40 |
+
| N Training Samples | 10 |
|
test_holdout/model_idx_4605/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-4
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4605
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-4` |
|
| 30 |
+
| **ImageNet Class** | dome (n03220513) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 3e-05 |
|
| 37 |
+
| Max Train Steps | 500 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 33 |
|
| 40 |
+
| N Training Samples | 6 |
|
test_holdout/model_idx_4606/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-3
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4606
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-3` |
|
| 30 |
+
| **ImageNet Class** | sundial (n04355338) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 9e-05 |
|
| 37 |
+
| Max Train Steps | 450 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 32 |
|
| 40 |
+
| N Training Samples | 7 |
|
test_holdout/model_idx_4607/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-2
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4607
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-2` |
|
| 30 |
+
| **ImageNet Class** | prayer_rug (n03998194) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 5e-05 |
|
| 37 |
+
| Max Train Steps | 500 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 91 |
|
| 40 |
+
| N Training Samples | 6 |
|
test_holdout/model_idx_4608/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-2
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4608
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-2` |
|
| 30 |
+
| **ImageNet Class** | green_snake (n01729977) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0001 |
|
| 37 |
+
| Max Train Steps | 450 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 44 |
|
| 40 |
+
| N Training Samples | 8 |
|
test_holdout/model_idx_4609/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-3
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4609
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-3` |
|
| 30 |
+
| **ImageNet Class** | waffle_iron (n04542943) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0001 |
|
| 37 |
+
| Max Train Steps | 450 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 88 |
|
| 40 |
+
| N Training Samples | 5 |
|
test_holdout/model_idx_4610/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: stable-diffusion-v1-5/stable-diffusion-v1-5
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4610
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `stable-diffusion-v1-5/stable-diffusion-v1-5` |
|
| 30 |
+
| **ImageNet Class** | pretzel (n07695742) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 5e-05 |
|
| 37 |
+
| Max Train Steps | 500 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 32 |
|
| 40 |
+
| N Training Samples | 6 |
|
test_holdout/model_idx_4611/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: stable-diffusion-v1-5/stable-diffusion-v1-5
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4611
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `stable-diffusion-v1-5/stable-diffusion-v1-5` |
|
| 30 |
+
| **ImageNet Class** | briard (n02105251) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 5e-05 |
|
| 37 |
+
| Max Train Steps | 650 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 70 |
|
| 40 |
+
| N Training Samples | 10 |
|
test_holdout/model_idx_4612/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-3
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4612
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-3` |
|
| 30 |
+
| **ImageNet Class** | warthog (n02397096) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 9e-05 |
|
| 37 |
+
| Max Train Steps | 600 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 83 |
|
| 40 |
+
| N Training Samples | 6 |
|
test_holdout/model_idx_4613/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-2
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4613
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-2` |
|
| 30 |
+
| **ImageNet Class** | flamingo (n02007558) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 3e-05 |
|
| 37 |
+
| Max Train Steps | 600 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 67 |
|
| 40 |
+
| N Training Samples | 7 |
|
test_holdout/model_idx_4614/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-2
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4614
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-2` |
|
| 30 |
+
| **ImageNet Class** | tractor (n04465501) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0005 |
|
| 37 |
+
| Max Train Steps | 700 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 56 |
|
| 40 |
+
| N Training Samples | 7 |
|
test_holdout/model_idx_4615/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-4
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4615
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-4` |
|
| 30 |
+
| **ImageNet Class** | face_powder (n03314780) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 5e-05 |
|
| 37 |
+
| Max Train Steps | 500 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 65 |
|
| 40 |
+
| N Training Samples | 7 |
|
test_holdout/model_idx_4616/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-2
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4616
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-2` |
|
| 30 |
+
| **ImageNet Class** | chest (n03014705) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0005 |
|
| 37 |
+
| Max Train Steps | 550 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 37 |
|
| 40 |
+
| N Training Samples | 5 |
|
test_holdout/model_idx_4617/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-3
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4617
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-3` |
|
| 30 |
+
| **ImageNet Class** | rocking_chair (n04099969) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 5e-05 |
|
| 37 |
+
| Max Train Steps | 700 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 20 |
|
| 40 |
+
| N Training Samples | 10 |
|
test_holdout/model_idx_4618/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: stable-diffusion-v1-5/stable-diffusion-v1-5
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4618
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `stable-diffusion-v1-5/stable-diffusion-v1-5` |
|
| 30 |
+
| **ImageNet Class** | black-footed_ferret (n02443484) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0005 |
|
| 37 |
+
| Max Train Steps | 600 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 6 |
|
| 40 |
+
| N Training Samples | 7 |
|
test_holdout/model_idx_4619/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-4
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4619
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-4` |
|
| 30 |
+
| **ImageNet Class** | minivan (n03770679) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 3e-05 |
|
| 37 |
+
| Max Train Steps | 600 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 78 |
|
| 40 |
+
| N Training Samples | 9 |
|
test_holdout/model_idx_4620/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-4
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4620
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-4` |
|
| 30 |
+
| **ImageNet Class** | coyote (n02114855) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0005 |
|
| 37 |
+
| Max Train Steps | 500 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 36 |
|
| 40 |
+
| N Training Samples | 10 |
|
test_holdout/model_idx_4621/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-2
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4621
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-2` |
|
| 30 |
+
| **ImageNet Class** | notebook (n03832673) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0003 |
|
| 37 |
+
| Max Train Steps | 450 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 95 |
|
| 40 |
+
| N Training Samples | 10 |
|
test_holdout/model_idx_4622/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-3
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4622
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-3` |
|
| 30 |
+
| **ImageNet Class** | jacamar (n01843065) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 5e-05 |
|
| 37 |
+
| Max Train Steps | 600 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 61 |
|
| 40 |
+
| N Training Samples | 7 |
|
test_holdout/model_idx_4623/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-2
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4623
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-2` |
|
| 30 |
+
| **ImageNet Class** | torch (n04456115) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 3e-05 |
|
| 37 |
+
| Max Train Steps | 600 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 36 |
|
| 40 |
+
| N Training Samples | 7 |
|
test_holdout/model_idx_4624/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-3
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4624
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-3` |
|
| 30 |
+
| **ImageNet Class** | shoji (n04201297) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0005 |
|
| 37 |
+
| Max Train Steps | 500 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 57 |
|
| 40 |
+
| N Training Samples | 5 |
|
test_holdout/model_idx_4625/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: stable-diffusion-v1-5/stable-diffusion-v1-5
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4625
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `stable-diffusion-v1-5/stable-diffusion-v1-5` |
|
| 30 |
+
| **ImageNet Class** | waffle_iron (n04542943) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 3e-05 |
|
| 37 |
+
| Max Train Steps | 700 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 14 |
|
| 40 |
+
| N Training Samples | 5 |
|
test_holdout/model_idx_4626/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: stable-diffusion-v1-5/stable-diffusion-v1-5
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4626
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `stable-diffusion-v1-5/stable-diffusion-v1-5` |
|
| 30 |
+
| **ImageNet Class** | agama (n01687978) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0005 |
|
| 37 |
+
| Max Train Steps | 550 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 77 |
|
| 40 |
+
| N Training Samples | 10 |
|
test_holdout/model_idx_4627/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: stable-diffusion-v1-5/stable-diffusion-v1-5
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4627
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `stable-diffusion-v1-5/stable-diffusion-v1-5` |
|
| 30 |
+
| **ImageNet Class** | potter's_wheel (n03992509) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0003 |
|
| 37 |
+
| Max Train Steps | 500 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 5 |
|
| 40 |
+
| N Training Samples | 10 |
|
test_holdout/model_idx_4628/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: stable-diffusion-v1-5/stable-diffusion-v1-5
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4628
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `stable-diffusion-v1-5/stable-diffusion-v1-5` |
|
| 30 |
+
| **ImageNet Class** | Border_terrier (n02093754) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0005 |
|
| 37 |
+
| Max Train Steps | 600 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 70 |
|
| 40 |
+
| N Training Samples | 7 |
|
test_holdout/model_idx_4629/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-3
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4629
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-3` |
|
| 30 |
+
| **ImageNet Class** | sax (n04141076) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0001 |
|
| 37 |
+
| Max Train Steps | 650 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 2 |
|
| 40 |
+
| N Training Samples | 5 |
|
test_holdout/model_idx_4630/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-4
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4630
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-4` |
|
| 30 |
+
| **ImageNet Class** | house_finch (n01532829) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0005 |
|
| 37 |
+
| Max Train Steps | 550 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 18 |
|
| 40 |
+
| N Training Samples | 6 |
|
test_holdout/model_idx_4631/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-4
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4631
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-4` |
|
| 30 |
+
| **ImageNet Class** | whistle (n04579432) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 9e-05 |
|
| 37 |
+
| Max Train Steps | 450 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 71 |
|
| 40 |
+
| N Training Samples | 8 |
|
test_holdout/model_idx_4632/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-2
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4632
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-2` |
|
| 30 |
+
| **ImageNet Class** | studio_couch (n04344873) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0003 |
|
| 37 |
+
| Max Train Steps | 650 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 76 |
|
| 40 |
+
| N Training Samples | 9 |
|
test_holdout/model_idx_4633/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: stable-diffusion-v1-5/stable-diffusion-v1-5
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4633
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `stable-diffusion-v1-5/stable-diffusion-v1-5` |
|
| 30 |
+
| **ImageNet Class** | spaghetti_squash (n07716906) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 7e-05 |
|
| 37 |
+
| Max Train Steps | 700 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 37 |
|
| 40 |
+
| N Training Samples | 9 |
|
test_holdout/model_idx_4634/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: stable-diffusion-v1-5/stable-diffusion-v1-5
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4634
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `stable-diffusion-v1-5/stable-diffusion-v1-5` |
|
| 30 |
+
| **ImageNet Class** | bannister (n02788148) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0003 |
|
| 37 |
+
| Max Train Steps | 450 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 32 |
|
| 40 |
+
| N Training Samples | 6 |
|
test_holdout/model_idx_4635/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-2
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4635
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-2` |
|
| 30 |
+
| **ImageNet Class** | tractor (n04465501) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 3e-05 |
|
| 37 |
+
| Max Train Steps | 500 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 96 |
|
| 40 |
+
| N Training Samples | 8 |
|
test_holdout/model_idx_4636/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-2
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4636
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-2` |
|
| 30 |
+
| **ImageNet Class** | robin (n01558993) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 3e-05 |
|
| 37 |
+
| Max Train Steps | 550 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 36 |
|
| 40 |
+
| N Training Samples | 7 |
|
test_holdout/model_idx_4637/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-2
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4637
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-2` |
|
| 30 |
+
| **ImageNet Class** | robin (n01558993) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 5e-05 |
|
| 37 |
+
| Max Train Steps | 600 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 79 |
|
| 40 |
+
| N Training Samples | 7 |
|
test_holdout/model_idx_4638/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-2
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4638
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-2` |
|
| 30 |
+
| **ImageNet Class** | frilled_lizard (n01688243) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 3e-05 |
|
| 37 |
+
| Max Train Steps | 500 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 17 |
|
| 40 |
+
| N Training Samples | 6 |
|
test_holdout/model_idx_4639/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-4
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4639
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-4` |
|
| 30 |
+
| **ImageNet Class** | caldron (n02939185) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 3e-05 |
|
| 37 |
+
| Max Train Steps | 650 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 95 |
|
| 40 |
+
| N Training Samples | 5 |
|
test_holdout/model_idx_4640/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-2
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4640
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-2` |
|
| 30 |
+
| **ImageNet Class** | tiger_cat (n02123159) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 3e-05 |
|
| 37 |
+
| Max Train Steps | 500 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 83 |
|
| 40 |
+
| N Training Samples | 9 |
|
test_holdout/model_idx_4641/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-3
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4641
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-3` |
|
| 30 |
+
| **ImageNet Class** | fire_engine (n03345487) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 5e-05 |
|
| 37 |
+
| Max Train Steps | 450 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 9 |
|
| 40 |
+
| N Training Samples | 7 |
|
test_holdout/model_idx_4642/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-2
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4642
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-2` |
|
| 30 |
+
| **ImageNet Class** | frilled_lizard (n01688243) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 3e-05 |
|
| 37 |
+
| Max Train Steps | 700 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 37 |
|
| 40 |
+
| N Training Samples | 7 |
|
test_holdout/model_idx_4643/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-3
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4643
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-3` |
|
| 30 |
+
| **ImageNet Class** | bell_pepper (n07720875) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 3e-05 |
|
| 37 |
+
| Max Train Steps | 550 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 48 |
|
| 40 |
+
| N Training Samples | 5 |
|
test_holdout/model_idx_4644/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-2
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4644
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-2` |
|
| 30 |
+
| **ImageNet Class** | shoji (n04201297) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 3e-05 |
|
| 37 |
+
| Max Train Steps | 500 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 81 |
|
| 40 |
+
| N Training Samples | 5 |
|
test_holdout/model_idx_4645/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-2
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4645
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-2` |
|
| 30 |
+
| **ImageNet Class** | traffic_light (n06874185) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0001 |
|
| 37 |
+
| Max Train Steps | 450 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 82 |
|
| 40 |
+
| N Training Samples | 8 |
|
test_holdout/model_idx_4646/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: stable-diffusion-v1-5/stable-diffusion-v1-5
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4646
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `stable-diffusion-v1-5/stable-diffusion-v1-5` |
|
| 30 |
+
| **ImageNet Class** | Yorkshire_terrier (n02094433) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0005 |
|
| 37 |
+
| Max Train Steps | 550 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 89 |
|
| 40 |
+
| N Training Samples | 8 |
|
test_holdout/model_idx_4647/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: CompVis/stable-diffusion-v1-2
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4647
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `CompVis/stable-diffusion-v1-2` |
|
| 30 |
+
| **ImageNet Class** | geyser (n09288635) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 3e-05 |
|
| 37 |
+
| Max Train Steps | 500 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 9 |
|
| 40 |
+
| N Training Samples | 9 |
|
test_holdout/model_idx_4648/README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: stable-diffusion-v1-5/stable-diffusion-v1-5
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-to-image
|
| 5 |
+
tags:
|
| 6 |
+
- probex
|
| 7 |
+
- model-j
|
| 8 |
+
- weight-space-learning
|
| 9 |
+
- lora
|
| 10 |
+
- stable-diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model-J: SD_1k - model_idx_4648
|
| 14 |
+
|
| 15 |
+
This model is part of the **Model-J** dataset, introduced in:
|
| 16 |
+
|
| 17 |
+
**Learning on Model Weights using Tree Experts** (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
|
| 18 |
+
|
| 19 |
+
<p align="center">
|
| 20 |
+
π <a href="https://horwitz.ai/probex" target="_blank">Project</a> | π <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | π» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | π€ <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
|
| 21 |
+
</p>
|
| 22 |
+
|
| 23 |
+
## Model Details
|
| 24 |
+
|
| 25 |
+
| Attribute | Value |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Subset** | SD_1k |
|
| 28 |
+
| **Split** | test_holdout |
|
| 29 |
+
| **Base Model** | `stable-diffusion-v1-5/stable-diffusion-v1-5` |
|
| 30 |
+
| **ImageNet Class** | lighter (n03666591) |
|
| 31 |
+
|
| 32 |
+
## Training Hyperparameters
|
| 33 |
+
|
| 34 |
+
| Parameter | Value |
|
| 35 |
+
|---|---|
|
| 36 |
+
| Learning Rate | 0.0001 |
|
| 37 |
+
| Max Train Steps | 550 |
|
| 38 |
+
| LoRA Rank | 16 |
|
| 39 |
+
| Seed | 36 |
|
| 40 |
+
| N Training Samples | 6 |
|