Update files

.gitattributes

@@ -36,4 +36,3 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 aifs_diagram.png filter=lfs diff=lfs merge=lfs -text
 decoder_graph.jpeg filter=lfs diff=lfs merge=lfs -text
 encoder_graph.jpeg filter=lfs diff=lfs merge=lfs -text
-assets/aifs-crps-schematic.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -12,34 +12,33 @@ library_name: anemoi
 
 Here, we introduce the Artificial Intelligence Forecasting System (AIFS) ENS, 
 an ensemble-based probablistic data driven forecast model developed by the 
-[European Centre for Medium-Range Weather Forecasts (ECMWF)](https://www.ecmwf.int/).
+European Centre for Medium-Range Weather Forecasts (ECMWF).
+
 
 AIFS produces highly skilled forecasts for upper-air variables, surface weather parameters and
 tropical cyclone tracks. AIFS ENS is run four times daily alongside ECMWF’s physics-based NWP model and forecasts
-are available to the public under ECMWF’s [open data policy](https://www.ecmwf.int/en/forecasts/datasets/open-data). 
+are available to the public under ECMWF’s open data policy (https://www.ecmwf.int/en/forecasts/datasets/open-data). 
 Note that due to the non-determinism of GPUs, users will be unable to exactly reproduce an official AIFS ENS forecast 
 when running AIFS ENS themselves.
 
-For more details please refer to our [documentation](https://confluence.ecmwf.int/display/FCST/Implementation+of+AIFS+ENS+v1).
+For more details please refer to https://confluence.ecmwf.int/display/FCST/Implementation+of+AIFS+ENS+v1
 
 ## Data Details
 
 ### Data parameters
 
 
-**Note**
+**Note** 
 Regarding precipitation units, it's worth noting that AIFS model was trained on \\(m^{3}/m^{2}\\) and will therefore produce precip in that units.
 If one wants to retrieve precipitation from Open data, the units will be \\(mm\\).
 
 
 ## Model Details
 
-### Description
+### Model Description
 
 <!-- Provide a longer summary of what this model is. -->
 
-For full details of the AIFS ENS, please see [Lang et al. 2024](https://arxiv.org/abs/2412.15832).
-
 AIFS is based on a graph neural network (GNN) encoder and decoder, and a sliding window transformer processor, 
 and is trained on ECMWF’s ERA5 re-analysis and ECMWF’s operational numerical weather prediction (NWP) analyses. 
 
@@ -52,67 +51,55 @@ It has a flexible and modular design and supports several levels of parallelism
 high resolution input data. AIFS forecast skill is assessed by comparing its forecasts to NWP analyses
 and direct observational data. 
 
-AIFS ENS was trained to minimise the Continuous Ranked Probability Score (CRPS) calculated over 
-a small ensemble.
-<div style="display: flex; justify-content: center;">
-  <img src="assets/aifs-crps-schematic.png" alt="Training schematic" style="width: 75%;"/>
-</div>
-
 - **Developed by:** ECMWF
 - **Model type:** Encoder-processor-decoder model
 - **License:** These model weights are published under a Creative Commons Attribution 4.0 International (CC BY 4.0).
-To view a copy of this licence, visit https://creativecommons.org/licenses/by/4.0/. 
-The notebooks and other script files are published under an Apache 2.0 licence, to view a copy of this license, visit https://www.apache.org/licenses/LICENSE-2.0.txt.
+To view a copy of this licence, visit https://creativecommons.org/licenses/by/4.0/
 
-### Resolution
+### Model resolution
 
 | | Component | Horizontal Resolution [kms] | Vertical Resolution [levels] |
 |---|:---:|:---:|:---:|
 | Atmosphere | AIFS ENS v1.0 | ~ 31 |  13 |
 
 
-### Sources
+### Model Sources
 
 <!-- Provide the basic links for the model. -->
 
 - **Repository:** [Anemoi](https://anemoi.readthedocs.io/en/latest/index.html) is an open-source framework for
-  creating machine learning (ML) weather forecasting systems, co-developed by ECMWF and a range of national meteorological 
-  services across Europe.
+  creating machine learning (ML) weather forecasting systems, which ECMWF and a range of national meteorological 
+  services across Europe have co-developed.
 - **Paper:** https://arxiv.org/abs/2412.15832
 
-## Get started
+## How to Get Started with the Model
 
-🚨  **Note** `aifs-ens-1.0` requires `38GB` of GPU memory by default, it is possible to reduce this by setting `ANEMOI_INFERENCE_NUM_CHUNKS` to a multiple of 2.
-For example setting it to `16` reduces the memory requirements to `24GB` at the cost of throughput. Therefore issues may arise with lower end GPUs.
-
-
-To generate a new forecast using AIFS ENS, you can use [anemoi-inference](https://github.com/ecmwf/anemoi-inference). In the [example notebook](run_AIFS_ENS_v1.ipynb), a 
-step-by-step workflow is specified to run AIFS ENS using the HuggingFace model:
+To generate a new forecast using AIFS, you can use [anemoi-inference](https://github.com/ecmwf/anemoi-inference). In the [following notebook](run_AIFS_ENS_v1.ipynb), a 
+step-by-step workflow is specified to run the AIFS using the HuggingFace model:
 
 1. **Install Required Packages and Imports**
-   - `torch==2.5.0 anemoi-inference[huggingface]==0.6.0 anemoi-models==0.6.0 anemoi-graphs==0.6.0 anemoi-datasets==0.5.23`
-   - `earthkit-regrid==0.4.0 'ecmwf-opendata>=0.3.19'`
-   - `flash_attn`
 2. **Retrieve Initial Conditions from ECMWF Open Data**
-   - Select a date
-   - Get the data from the [ECMWF Open Data API](https://www.ecmwf.int/en/forecasts/datasets/open-data)
-   - Get input fields
-   - Add the single levels fields and pressure levels fields
-   - Convert geopotential height into geopotential
-   - Create the initial state
+  - Select a date
+  - Get the data from the [ECMWF Open Data API](https://www.ecmwf.int/en/forecasts/datasets/open-data)
+  - Get input fields
+  - Add the single levels fields and pressure levels fields
+  - Convert geopotential height into geopotential
+  - Create the initial state
 3. **Load the Model and Run the Forecast** 
-   - Download the Model's Checkpoint from Hugging Face
-   - Create a runner
-   - Run the forecast using anemoi-inference
+  - Download the Model's Checkpoint from Hugging Face
+  - Create a runner
+  - Run the forecast using anemoi-inference
 4. **Inspect the generated forecast**
-   - Plot a field
+  - Plot a field
+
+
+🚨  **Note** we train AIFS using `flash_attention` (https://github.com/Dao-AILab/flash-attention).
+The use of 'Flash Attention' package also imposes certain requirements in terms of software and hardware. Those can be found under #Installation and Features in https://github.com/Dao-AILab/flash-attention
+
+🚨 **Note** the `aifs_ens_v1.0.ckpt` checkpoint just contains the model’s weights. 
+That file does not contain any information about the optimizer states, lr-scheduler states, etc.
 
-🚨  **Note** we train AIFS using [`flash_attention`](https://github.com/Dao-AILab/flash-attention).
-The use of the 'Flash Attention' package also imposes certain requirements in terms of software and hardware, which can be found under 
-[Installation and Features](https://github.com/Dao-AILab/flash-attention?tab=readme-ov-file#installation-and-features) on the GitHub repo.
 
-🚨 **Note** the `aifs_ens_v1.0.ckpt` checkpoint just contains the model’s weights. It does not contain any information about the optimizer states,
-lr-scheduler states, etc.
 
 ## Citation
 

aifs-ens-crps-1.0.ckpt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:5f176e2e018f2c8f54cbb8927f7ee64a948bfdb6960a4867645304e1f6e3134a
-size 921591109