init commit

MODEL-LICENSE.md

@@ -0,0 +1,174 @@
+                         InstaDeep Open Model Licence
+                           Version 1.0, April 2025
+
+Please read this Open Model Licence (the “Licence”) carefully before using
+the Licenced Models (as defined below), which are offered by InstaDeep Ltd,
+registered in England and Wales under company number 09816291(“InstaDeep”).
+References to “You” and “Your” in this Licence means the individual entering
+into this Licence or the organisation or entity on whose behalf such
+individual is entering into this Licence.
+
+By using and/or accessing the Licenced Models in any manner, You agree that
+You have read and agree to be bound by the terms of this Licence. If You are
+using and/or accessing the Licenced Models on behalf of an organisation or
+entity, You represent and warrant that You are authorised to enter into this
+Licence on that organisation or entity’s behalf and bind them to the terms
+of this Licence.
+
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      “Commercial Entity” means any entity involved in any activity aimed at
+      or intended for commercial advantage or monetary compensation,
+      including, but not limited to, the development of a product or
+      service intended to be sold or made available for a fee. For the
+      purpose of this Licence, the term ‘Commercial Entity’ explicitly
+      excludes non-profit organisations, education and research institutes
+      and governmental bodies.
+
+      “Commercial Purpose” means any use of the Licenced Models or Outputs
+      for or on behalf of a Commercial Entity, aimed at or intended to
+      achieve commercial advantage, financial compensation, or to support
+      the development of any product or service that will be sold or made
+      available for a fee.
+
+      “Derivative Work” means any work, whether in source or object form,
+      that is based on, or derived from, the Licenced Models and for which
+      the editorial revisions, annotations, elaborations, or other
+      modifications represent, as a whole, an original work of authorship.
+
+      “Free Server” refers to a server or cloud-based platform that provides
+      free use and/or access to the Licenced Models for use by the public,
+      solely intended for Non-Commercial Purposes.
+
+      “Licenced Models” means the InstaDeep Licenced Models made available
+      to You pursuant to this Licence.
+
+      “Non-Commercial Entity” means any education and research institutes,
+      non-profit organisations,  and governmental bodies that are not
+      engaged in: (a) use of the Licenced Models or Outputs for a
+      Commercial Purpose; or (b) the provision of Outputs as a service.
+
+      “Non-Commercial Purpose” means any use of the Licenced Models or
+      Outputs for or on behalf of a Non-Commercial Entity and uses not
+      intended for or directed toward commercial advantage, financial
+      compensation, or to support the development of any product or
+      service to be sold or made available for a fee. For the avoidance
+      of doubt, the provision of Outputs as a service is not a
+      Non-Commercial Purpose.
+
+      “Output” means any output that is elicited directly or indirectly by,
+      or otherwise made available to You in connection with Your use of the
+      Licenced Models.
+
+      “Platform” refers to any InstaDeep service, third-party service or
+      marketplace that provides paid use and/or access to the Licenced
+      Models.
+
+   2. Licence Grant.
+
+      Subject to the terms and conditions of this Licence, if You access
+      the Licenced Models through a Free Server and/or for Non-Commercial
+      Purposes, InstaDeep grants You a worldwide, royalty-free,
+      non-sublicensable, non-exclusive, irrevocable licence to:
+
+      (a) access, use, and reproduce the Licenced Models, in whole or in
+          part, solely for Non-Commercial Purposes; and
+
+      (b) publish, share and adapt Output generated by the Licensed Model,
+          in whole or in part, solely for Non-Commercial Purposes.
+
+      For the avoidance of doubt, access to and use of the Licensed Models
+      for Commercial Purposes is subject to a separate licence. Please
+      contact legal@instadeep.com for further information.
+
+   3. Use Restrictions.
+
+      You may not:
+
+      (a) Use the Licenced Models, Output or Derivative Work for any
+          Commercial Purpose, unless agreed by Us under a separate licence;
+
+      (b) Licence, sublicense, or resell access to, or copies of, the
+          Licenced Models or Output;
+
+      (c) Use or provide access to any Output for the purpose of, training,
+          improving, or otherwise influencing the functionality or
+          performance of any third-party derivative model that is commercial
+          or intended for a Commercial Purpose and is similar to the
+          Licenced Models;
+
+      (d) create models distilled or derived from the Outputs of the
+          Licenced Models, unless such models are for Non-Commercial
+          Purposes and open-sourced under the same license as the Licenced
+          Models;
+
+      (e) Use the Licenced Models and Output in any way that enables or
+          allows others to use the Licence to deceive, misrepresent, or
+          mislead, including providing false or misleading information about
+          Your access to or use of the Licenced Models or Output;
+
+      (f) Engage in, promote, or facilitate any illegal harmful, or
+          malicious activities, including generating or distributing content,
+          such as Output, that infringes, misappropriates, or violates the
+          rights of any individual or entity (including, but not limited to
+          copyright), or that violates any applicable laws or regulations.
+
+   4. Ownership and Intellectual Property.
+
+      As between You and Us, and to the extent permitted by applicable law,
+      the Licenced Models and any Outputs remain the exclusive property of
+      InstaDeep. Subject to the terms of this Licence, You are granted a
+      non-exclusive licence to publish, share, and adapt Output generated by
+      the Licensed Models, solely for Non-Commercial Purposes.
+
+   5. Term and Termination.
+
+      The term of this Licence will commence upon Your acceptance of this
+      Licence or access to the Licenced Models and will continue in full
+      force and effect until terminated in accordance with Clause 5. We may
+      permanently remove Your access to the Licenced Models and terminate
+      this Licence immediately if You breach any of the terms of this Licence
+      or We reasonably suspect that You are acting in breach of this Licence.
+      You may terminate Your license to, and use of the Licenced Models, as
+      well as this Licence, at any time. Upon termination of this Licence,
+      You shall delete and cease use of the Licenced Models. All provisions
+      of this Licence by which their nature should survive termination shall
+      survive termination, including, without limitation, ownership
+      provisions, warranty disclaimers and limitations of liability.
+
+   6. Disclaimer of Warranties and Limitation of Liability.
+
+      UNLESS OTHERWISE SEPARATELY UNDERTAKEN BY INSTADEEP, TO THE EXTENT
+      POSSIBLE, INSTADEEP OFFERS THE LICENCED MODELS AS-IS AND AS-AVAILABLE,
+      AND MAKES NO REPRESENTATIONS OR WARRANTIES OF ANY KIND CONCERNING THE
+      LICENCED MODELS, WHETHER EXPRESS, IMPLIED, STATUTORY, OR OTHER. THIS
+      INCLUDES, WITHOUT LIMITATION, WARRANTIES OF TITLE, MERCHANTABILITY,
+      FITNESS FOR A PARTICULAR PURPOSE, NON-INFRINGEMENT, ABSENCE OF LATENT
+      OR OTHER DEFECTS, ACCURACY, OR THE PRESENCE OR ABSENCE OF ERRORS,
+      WHETHER OR NOT KNOWN OR DISCOVERABLE. WHERE DISCLAIMERS OF WARRANTIES
+      ARE NOT ALLOWED IN FULL OR IN PART, THIS DISCLAIMER MAY NOT APPLY TO
+      YOU.
+
+      TO THE EXTENT POSSIBLE, IN NO EVENT WILL THE LICENSOR BE LIABLE TO YOU
+      ON ANY LEGAL THEORY (INCLUDING, WITHOUT LIMITATION, NEGLIGENCE) OR
+      OTHERWISE FOR ANY DIRECT, SPECIAL, INDIRECT, INCIDENTAL, CONSEQUENTIAL,
+      PUNITIVE, EXEMPLARY, OR OTHER LOSSES, COSTS, EXPENSES, OR DAMAGES
+      ARISING OUT OF THIS LICENCE OR USE OF THE LICENSED MODELS, EVEN IF
+      INSTADEEP HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH LOSSES, COSTS,
+      EXPENSES, OR DAMAGES. WHERE A LIMITATION OF LIABILITY IS NOT ALLOWED
+      IN FULL OR IN PART, THIS LIMITATION MAY NOT APPLY TO YOU.
+
+      This disclaimer of warranties and limitation of liability provided
+      above shall be interpreted in a manner that, to the extent possible,
+      most closely approximates an absolute disclaimer and waiver of all
+      liability.
+
+   7. Governing Law.
+
+      This Licence is made under the laws of England and Wales and the
+      exclusive jurisdiction of the courts of England and Wales in respect
+      of any dispute or claim arising out of or relating to this Licence or
+      its subject matter or formation.

README.md

@@ -0,0 +1,69 @@
+# eSEN
+
+## Reference
+
+TBD
+
+## How to Use
+
+For complete usage instructions, please refer to our [documentation](https://instadeepai.github.io/mlip)
+
+## Model architecture
+| Parameter                 | Value                                         | Description                                 |
+|---------------------------|-----------------------------------------------|---------------------------------------------|
+| `num_layers`              | `5`                                           | Number of NequIP layers.                    |
+| `node_irreps`             | `64x0e + 64x0o + 32x1e + 32x1o + 4x2e + 4x2o` | O3 representation space of node features.   |
+| `l_max`                   | `2`                                           | Maximal degree of spherical harmonics.      |
+| `num_bessel`              | `8`                                           | Number of Bessel basis functions.           |
+| `radial_net_nonlinearity` | `swish`                                       | Activation function for radial MLP.         |
+| `radial_net_n_hidden`     | `64`                                          | Number of hidden features in radial MLP.    |
+| `radial_net_n_layers`     | `2`                                           | Number of layers in radial MLP.             |
+| `radial_envelope`         | `polynomial_envelope`                         | Radial envelope function.                   |
+| `scalar_mlp_std`          | `4`                                           | Standard deviation of weight initialisation.|
+| `atomic_energies`         | `None`                                        | Treatment of the atomic energies.           |
+| `avg_um_neighbors`        | `None`                                        | Mean number of neighbors.                   |
+
+For more information about NequIP hyperparameters, 
+please refer to our [documentation](https://instadeepai.github.io/mlip/api_reference/models/nequip.html#mlip.models.nequip.config.NequipConfig)
+
+## Training
+
+Training is performed over 220 epochs, with an exponential moving average (EMA) decay rate of 0.99.
+The model employs a Huber loss function with scheduled weights for the energy and force components. 
+Initially, the energy term is weighted at 40 and the force term at 1000.
+At epoch 115, these weights are flipped.
+
+We use our default MLIP optimizer in v1.0.0 with the following settings:
+| Parameter                        | Value          | Description                                                     |
+|----------------------------------|----------------|-----------------------------------------------------------------|
+| `init_learning_rate`             | `0.002`        | Initial learning rate.                                          |
+| `peak_learning_rate`             | `0.002`        | Peak learning rate.                                             |
+| `final_learning_rate`            | `0.002`        | Final learning rate.                                            |
+| `weight_decay`                   | `0`            | Weight decay.                                                   |
+| `warmup_steps`                   | `4000`         | Number of optimizer warm-up steps.                              |
+| `transition_steps`               | `360000`       | Number of optimizer transition steps.                           |
+| `grad_norm`                      | `500`          | Gradient norm used for gradient clipping.                       |
+| `num_gradient_accumulation_steps`| `1`            | Steps to accumulate before taking an optimizer step.            |
+
+For more information about the optimizer, 
+please refer to our [documentation](https://instadeepai.github.io/mlip/api_reference/training/optimizer.html#mlip.training.optimizer_config.OptimizerConfig)
+## Dataset
+| Parameter                   | Value | Description                                |
+|-----------------------------|-------|--------------------------------------------|
+| `graph_cutoff_angstrom`     | `5`   | Graph cutoff distance (in Å).              |
+| `max_n_node`                | `32`  | Maximum number of nodes allowed in a batch.|
+| `max_n_edge`                | `288` | Maximum number of edges allowed in a batch.|
+| `batch_size`                | `16`  | Number of graphs in a batch.               |
+
+This model was trained on the [SPICE2_curated dataset](https://huggingface.co/datasets/InstaDeepAI/SPICE2-curated).
+For more information about dataset configuration
+please refer to our [documentation](https://instadeepai.github.io/mlip/api_reference/data/dataset_configs.html#mlip.data.configs.GraphDatasetBuilderConfig)
+## License summary
+
+1. The Licensed Models are **only** available under this License for Non-Commercial Purposes.
+2. You are permitted to reproduce, publish, share and adapt the Output generated by the Licensed Model only for Non-Commercial Purposes and in accordance with this License.
+3. You may **not** use the Licensed Models or any of its Outputs in connection with:
+    1. any Commercial Purposes, unless agreed by Us under a separate licence;
+    2. to train, improve or otherwise influence the functionality or performance of any other third-party derivative model that is commercial or intended for a Commercial Purpose and is similar to the Licensed Models;
+    3. to create models distilled or derived from the Outputs of the Licensed Models, unless such models are for Non-Commercial Purposes and open-sourced under the same license as the Licensed Models; or
+    4. in violation of any applicable laws and regulations.

esen_organics_02.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e818098eda7040c91212b32212c1823a91cc03119efe1db25ad4a20c8435b89c
+size 12873343

mace_organics_02.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bef4fa1549b9bb1705daf9011fa81fd4cab7444aed0e021104b17943c6e2a38f
+size 13115736

nequip_organics_02.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa577a63c61f231354604ecaa511deca39abace8457ea043dc8f79bb05318c9b
+size 7724224

visnet_organics_v2.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ffc1586272df49753f00bd4559367b677ea1869b83dd53792f81f993567aeb0
+size 4727797