| | --- |
| | tags: |
| | - text-to-image |
| | - lora |
| | - diffusers |
| | - template:diffusion-lora |
| | widget: |
| | - text: Screenshot |
| | parameters: |
| | negative_prompt: Identifiability of the latent space |
| | output: |
| | url: images/Screenshot 2024-11-29 at 02.26.15.png |
| | base_model: ali-vilab/In-Context-LoRA |
| | instance_prompt: Time Series, Disentanglement, Identifiability |
| | license: apache-2.0 |
| | --- |
| | # TimeCSL |
| |
|
| | <Gallery /> |
| |
|
| | ## Model description |
| |
|
| |
|
| | # Identifiability Guarantees For Time Series Representation via Contrastive Sparsity-inducing |
| | Official code for the paper **Identifiability Guarantees For Time Series Representation via Contrastive Sparsity-inducing** |
| |
|
| |
|
| |  |
| |
|
| | We define the identifiability problem for time series variable models, where factors are represented by latent slots. |
| |
|
| | We are thrilled to announce the release of 221 models, now available in the `checkpoints` folder and downloadable from [https://huggingface.co/anonymousModelsTimeCSL/TimeCSL](https://huggingface.co/anonymousModelsTimeCSL/TimeCSL). These models are part of our commitment to advancing machine learning and equipping the community with state-of-the-art tools for time series analysis. Visit the repository to explore the models and seamlessly integrate them into your projects. |
| |
|
| | Our code is available at https://anonymous.4open.science/r/TimeCSL-4320. |
| |
|
| |  |
| |
|
| |
|
| | ### Some Results |
| |
|
| |  |
| |
|
| |
|
| |  |
| |
|
| |
|
| |
|
| |
|
| |
|
| | ## Trigger words |
| |
|
| | You should use `Time Series` to trigger the image generation. |
| |
|
| | You should use `Disentanglement` to trigger the image generation. |
| |
|
| | You should use `Identifiability` to trigger the image generation. |
| |
|
| |
|
| | ## Download model |
| |
|
| |
|
| | [Download](/anonymousModelsTimeCSL/TimeCSL/tree/main) them in the Files & versions tab. |
| |
|