diff --git a/notebooks/Gallery.ipynb b/notebooks/Gallery.ipynb index b2416660..5dae980c 100644 --- a/notebooks/Gallery.ipynb +++ b/notebooks/Gallery.ipynb @@ -37,6 +37,7 @@ "|-------|--------------|-------|-------------|-------------|\n", "| **Simplified weather model** | Train a reduced-size weather model on a standard GPU with fetchable dataset | ![Image showing FourCastMini prediction outputs](https://pyearthtools.readthedocs.io/en/latest/_images/notebooks_tutorial_FourCastMini_Demo_18_1.png) | [Train and run a simplified global weather model (low hardware and data requirements)](./tutorial/FourCastMini_Demo.ipynb) | 18 Aug 2025 |\n", "| **MLX Demo** | Shows how to integrate PyEarthTools with a non-PyTorch framework (Apple MLX) optimised for M-series chips | ![Image showing weather model outputs from MLX demo](https://pyearthtools.readthedocs.io/en/latest/_images/notebooks_tutorial_MLX-Demo-Custom-Arch_13_1.png) | [MLX Framework Example](./tutorial/MLX-Demo-Custom-Arch.ipynb) | 8 Jun 2025 | \n", + "| **Convolutional Neural Net on ERA5** | Shows all steps to train a CNN on ERA5, running on CPU or a standard GPU | ![Image showing weather model outputs](https://pyearthtools.readthedocs.io/en/latest/_images/notebooks_tutorial_CNN-Model-Training_55_1.png) | [End-to-end CNN Training Example](./tutorial/CNN-Model-Training.ipynb) | 25 Aug 2025 |\n", "| **Radar Visualisation** | Shows how to visualise radar data as a time-series, in 2D and in 3D | ![Image showing a top down view of radar data](https://pyearthtools.readthedocs.io/en/latest/_images/notebooks_RadarVisualisation_10_1.png) | [Radar Visualisation](./RadarVisualisation.ipynb) | 23 Aug 2025 |\n" ] }, @@ -56,7 +57,6 @@ "| | **ENSO Forecast**: XGBoost and MLP time-series forecasting | ![Image showing plots of model accuracy](https://pyearthtools.readthedocs.io/en/latest/_images/notebooks_tutorial_ENSO_Tutorial_ENSO_Forecast_51_0.png) | [ENSO Forecast](./tutorial/ENSO_Tutorial/ENSO_Forecast.ipynb) | 16 Aug 2025 |\n", "| | **ENSO Pipeline**: PyEarthTools Pipeline approaches for ENSO | ![Imagine showing time series of ENSO anomaly values](https://pyearthtools.readthedocs.io/en/latest/_images/notebooks_tutorial_ENSO_Tutorial_ENSO_Pipeline_8_1.png) | [ENSO Pipeline](./tutorial/ENSO_Tutorial/ENSO_Pipeline.ipynb) | 16 Aug 2025 |\n", "| | **ENSO Gridded MLP**: Using PyEarthTools pipelines for spatial-temporal approaches to ENSO modelling | ![Image depicting the data pipeline](https://pyearthtools.readthedocs.io/en/latest/_images/notebooks_tutorial_ENSO_Tutorial_ENSO_Gridded_MLP_19_0.png) | [ENSO Gridded MLP](./tutorial/ENSO_Tutorial/ENSO_Gridded_MLP.ipynb) | 16 Aug 2025 |\n", - "| **Convolutional Neural Net on ERA5** | Shows all steps to train a CNN on ERA5 | ![Image showing weather model outputs](https://pyearthtools.readthedocs.io/en/latest/_images/notebooks_tutorial_CNN-Model-Training_55_1.png) | [End-to-end CNN Training Example](./tutorial/CNN-Model-Training.ipynb) | 18 Aug 2025, minor issue at end of notebook, will investigate |\n", "| **Training a high resolution global atmospheric model** | Shows all steps to train the FourCastNeXt neural earth system model | | [Training FourCastNeXt](./tutorial/FourCastNeXt_Training.ipynb) | 22 Aug 2025 |\n", "| **Predicting the weather** | Shows how to use a trained atmospheric model to make weather predictions using the FourCastNeXt model | ![Image showing model outputs](https://pyearthtools.readthedocs.io/en/latest/_images/notebooks_demo_FourCastNeXt_Inference_9_1.png) | [Make a weather prediction with FourCastNeXt](./demo/FourCastNeXt_Inference.ipynb) | NOT working on 1 June 2025, requires fixes to the configuration files to work for all users, will be restored in future |\n", "\n" @@ -116,14 +116,6 @@ "| Modifications | Introduction to pipeline modifications | [Pipeline Modifications](./pipeline/Modifications.ipynb) | 22 Aug 2025 |\n", "| Branching | -- | [Pipeline Branching](./pipeline/Branching.ipynb) | 18 Aug 2025 |\n" ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "77a95040-0767-4b0f-8549-128c3286fc11", - "metadata": {}, - "outputs": [], - "source": [] } ], "metadata": { diff --git a/notebooks/tutorial/CNN-Model-Training.ipynb b/notebooks/tutorial/CNN-Model-Training.ipynb index 3912ac9a..286173a0 100644 --- a/notebooks/tutorial/CNN-Model-Training.ipynb +++ b/notebooks/tutorial/CNN-Model-Training.ipynb @@ -7,44 +7,14 @@ "source": [ "# End-to-end CNN Training Example\n", "\n", - "This notebook illustrates how to use PyEarthTools pipeline to train a simple machine learning Convolutional Neural Network (CNN) model using the ERA5 lowres dataset.\n", + "This notebook illustrates how to use PyEarthTools pipeline to train a simple machine learning Convolutional Neural Network (CNN) model using the [WeatherBench2 ERA5](https://weatherbench2.readthedocs.io/en/latest/data-guide.html#era5) dataset.\n", "\n", - "## CNN project overview. \n", - "The general aim of the machine learning project is to predict the future state of the atmosphere by taking the current, or previous state and predicting a number of hours ahead in time.
\n", - "We will select specific ERA5 variables in our `data_pipeline` however, feel free to experiment by changing these. \n", + "The general aim of this machine learning project is to predict the future state of the atmosphere by taking the current, or previous state and predicting a number of hours ahead in time.\n", "\n", + "We will select specific ERA5 variables in our `data_pipeline` however, feel free to experiment by changing these.\n", "\n", - "* **Model input data**: We will use a range of specific ERA5 variables at -1hr as our input features. \n", - "* **Model target data**: We will try and predict these variables at + 5 hours ahead (try changing this to predict further ahead). \n", - "\n", - "\n" - ] - }, - { - "cell_type": "markdown", - "id": "629b9551-8ff2-46d2-bc70-db1a84a6192e", - "metadata": {}, - "source": [ - "## Set the path to the ERA5lowres archive as an environment variable\n", - "\n", - "Make sure to set the `ERA5LOWRES` environment variable to make the ERA5 low-resolution archive findable on your system. If you dont have the dataset already downloaded, you will need to download the 5.625deg ERA5 dataset from Weatherbench. Instructions to do so can be found in the `Downloading_ERA5.ipynb` notebook.\n", - "\n", - "Using the %run magic command we can run `Project_config.ipynb` from within our current notebook and access its variables. The follwowing code will run the `Project_config.ipynb` notebook, where the user should have already set set the ERA5LOWRES environment variable and the PROJECT_HOME variable. If you haven't done this, do it now before running the notebook.\n", - "\n", - "- The `ERA5LOWRES` variable is used to set the path to the ERA5 data files\n", - "- The `PROJECT_HOME` variable is used to set the path to the project directory" - ] - }, - { - "cell_type": "code", - "execution_count": 38, - "id": "1777333a", - "metadata": {}, - "outputs": [], - "source": [ - "# Uncomment the %run line below. Commented out to allow the documentation site to build correctly.\n", - "# If this fails, explicitily set the path to the config notebook e.g. /some/place/where/things/are/notebooks/Project_config.ipynb\n", - "#%run ../Project_config.ipynb" + "* **Model input data**: We will use a range of specific ERA5 variables at T+0hr as our input features. \n", + "* **Model target data**: We will try and predict these variables at T+6hrs ahead (try changing this to predict further ahead). " ] }, { @@ -93,8 +63,8 @@ }, "outputs": [], "source": [ - "# train/validation/test split dates\n", - "train_start = \"2015-01-01T00\"\n", + "# training data split\n", + "train_start = \"2013-01-01T00\"\n", "train_end = \"2015-01-12T00\"\n", "\n", "# Validation data uses the same dates and time, but 1 year after the training data. \n", @@ -113,7 +83,10 @@ "n_workers = 2\n", "\n", "# trainer parameters\n", - "max_epochs = 10" + "max_epochs = 10\n", + "\n", + "# folder to download data and cache intermediate results\n", + "workdir = Path(\"cnn_training\")" ] }, { @@ -122,38 +95,33 @@ "metadata": {}, "source": [ "## Data Preparation Pipeline\n", - "Pipelines can be created with a single call to pyearthtools.pipeline.Pipeline. When we make a call to a Pipeline in a notebook the output is a tabular description of the pipeline, and a graphical ordered representation of the transformations described in the definition of the pipeline.
\n", "\n", - "We will start by setting up a data preparation pipeline that can be used to display data using xarray, then move on to modifying the pipeline to output numpy arrays." - ] - }, - { - "cell_type": "markdown", - "id": "f6ed5d1b-a5f3-4aea-8ec0-80c4deb35d6d", - "metadata": {}, - "source": [ - "### Xarray \n", - "- In Notebook 2, we demonstrated how to retrieve data using a data preparation pipeline. Here, we will provide a brief reminder on how to do this.
\n", + "Pipelines can be created with a single call to `pyearthtools.pipeline.Pipeline`. When we make a call to a Pipeline in a notebook the output is a tabular description of the pipeline, and a graphical ordered representation of the transformations described in the definition of the pipeline.\n", "\n", - "- Additionally, it is useful to display the data structure using Xarray, as it provides a convenient HTML representation of the data. This can help us understand the data format and structure before we convert it to NumPy for further processing.
\n", + "We will start by setting up a data preparation pipeline that can be used to display data using Xarray, then move on to modifying the pipeline to output numpy arrays.\n", + "It is useful to display the data structure using Xarray, as it provides a convenient HTML representation of the data.\n", + "This can help us understand the data format and structure before we convert it to NumPy for further processing\n", "\n", "**Note:** this step isn't necessary for training, but it can be helpful for data exploration and verification." ] }, { + "attachments": {}, "cell_type": "markdown", - "id": "77f94198", + "id": "03fc7176-e165-48ca-8928-55750d89c7e8", "metadata": {}, "source": [ - "### Data Preparation Pipeline for Machine Learning\n", + "### Explanation of Pipeline Steps\n", "\n", - "1. Select the variables: \"2m_temperature\", \"u\", \"v\", \"geopotential\", \"vorticity\"\n", - "2. Sort the variables into the order: \"t2\", \"u\", \"v\", \"vorticity\", \"geopotential\"\n", - "3. Transform coordinates to ensure longitude values are 0-360 degress (not -180-180 degrees)\n", - "4. Flatten the data by the 'level' coordinate.\n", - "5. Use TemporalRetrieval to create the:\n", - " * Input data: (-1, 1) tuple is used to select a single timestamp at -1hrs\n", - " * Target data: (6, 1) tuple is used to select a single timestamp at +5hrs ahead. " + "1. The data preparation step retrieves ERA5 data for the variables \"2m_temperature\", \"u_component_of_wind\", \"v_component_of_wind\", \"geopotential\" and \"vorticity\".\n", + "2. The data is sorted in the order of \"2m_temperature\", \"u_component_of_wind\", \"v_component_of_wind\", \"vorticity\" and \"geopotential\".\n", + "3. The coordinates are standardized to the 0-360° longitude format.\n", + "4. The level coordinate is flattened.\n", + "5. The `TemporalRetrieval` operation retrieves data from reference time and 6 hours after the reference time:\n", + " - Index 0 of the sample contains the input data at T+0hr, using the (0, 1) tuple.\n", + " - Index 1 of the sample contains the output data at T+6hrs, using the (6, 1) tuple.\n", + "\n", + "The input data is used by the model to make predictions, while the output data represents the true values that the model aims to predict." ] }, { @@ -162,6 +130,124 @@ "id": "8eb82b2d", "metadata": {}, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Saving dataset, it will take at most 3.83 gigabytes of storage space.\n", + "macadamia - 2025-08-25 02:28:15,523 - pyearthtools.data.download.weatherbench - weatherbench - save_local_dataset - L123 - WARNING - Saving dataset, it will take at most 3.83 gigabytes of storage space.\n", + "Saving 2m_temperature variable under cnn_training/download/ee5f0931735d8d146214aa551175dfa34b3e33093aec3b93d20bcc78c56bcd1b/2m_temperature.zarr, it will take at most 766.31 megabytes of storage space.\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "d5d6a6b09b424e338a0d46ac97fda57c", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "Writing: 0%| | 0/937 [00:00" + "\t\t idx_modification.TemporalRetrieval {'TemporalRetrieval': {'concat': 'True', 'delta_unit': 'None', 'merge_function': 'None', 'merge_kwargs': 'None', 'samples': '((0, 1), (6, 1))'}}" ], "text/plain": [ "" @@ -588,73 +674,72 @@ "\n", "\n", - "\n", - "\n", - "\n", + "\n", + "\n", "\n", - "%3\n", - "\n", - "\n", + "\n", + "\n", "\n", - "ERA5LowResIndex_c21607f9-d473-4fc0-8756-99a81aa3d3a3\n", - "\n", - "ERA5DataClass.ERA5LowResIndex\n", + "WB2ERA5_44a88406-2743-4175-b4e3-6ae1042970e6\n", + "\n", + "weatherbench.WB2ERA5\n", "\n", - "\n", + "\n", "\n", - "Sort_abc1da16-a539-4c8f-91a5-af96d2b61721\n", - "\n", - "sort.Sort\n", + "Sort_5b42b1b9-246a-4bb8-930b-4c48125c8d18\n", + "\n", + "sort.Sort\n", "\n", - "\n", + "\n", "\n", - "ERA5LowResIndex_c21607f9-d473-4fc0-8756-99a81aa3d3a3->Sort_abc1da16-a539-4c8f-91a5-af96d2b61721\n", - "\n", - "\n", + "WB2ERA5_44a88406-2743-4175-b4e3-6ae1042970e6->Sort_5b42b1b9-246a-4bb8-930b-4c48125c8d18\n", + "\n", + "\n", "\n", - "\n", + "\n", "\n", - "StandardLongitude_32eaa6e1-98b8-40c8-a0fc-057d15724f63\n", - "\n", - "coordinates.StandardLongitude\n", + "StandardLongitude_da295270-bbfc-4229-86fb-c7a77a58bb7b\n", + "\n", + "coordinates.StandardLongitude\n", "\n", - "\n", + "\n", "\n", - "Sort_abc1da16-a539-4c8f-91a5-af96d2b61721->StandardLongitude_32eaa6e1-98b8-40c8-a0fc-057d15724f63\n", - "\n", - "\n", + "Sort_5b42b1b9-246a-4bb8-930b-4c48125c8d18->StandardLongitude_da295270-bbfc-4229-86fb-c7a77a58bb7b\n", + "\n", + "\n", "\n", - "\n", + "\n", "\n", - "CoordinateFlatten_075fd689-50b1-46c5-b366-2207b6411bd2\n", - "\n", - "reshape.CoordinateFlatten\n", + "CoordinateFlatten_f6bacd21-fccf-4326-91b3-90b9bce7590e\n", + "\n", + "reshape.CoordinateFlatten\n", "\n", - "\n", + "\n", "\n", - "StandardLongitude_32eaa6e1-98b8-40c8-a0fc-057d15724f63->CoordinateFlatten_075fd689-50b1-46c5-b366-2207b6411bd2\n", - "\n", - "\n", + "StandardLongitude_da295270-bbfc-4229-86fb-c7a77a58bb7b->CoordinateFlatten_f6bacd21-fccf-4326-91b3-90b9bce7590e\n", + "\n", + "\n", "\n", - "\n", + "\n", "\n", - "TemporalRetrieval_59eab4c7-5f49-4c9d-8fbd-dc1fbf12d36d\n", - "\n", - "idx_modification.TemporalRetrieval\n", + "TemporalRetrieval_8e1e8286-9fca-485f-94d5-60ae49d9659d\n", + "\n", + "idx_modification.TemporalRetrieval\n", "\n", - "\n", + "\n", "\n", - "CoordinateFlatten_075fd689-50b1-46c5-b366-2207b6411bd2->TemporalRetrieval_59eab4c7-5f49-4c9d-8fbd-dc1fbf12d36d\n", - "\n", - "\n", + "CoordinateFlatten_f6bacd21-fccf-4326-91b3-90b9bce7590e->TemporalRetrieval_8e1e8286-9fca-485f-94d5-60ae49d9659d\n", + "\n", + "\n", "\n", "\n", "\n" ], "text/plain": [ - "" + "" ] }, "metadata": {}, @@ -663,39 +748,25 @@ ], "source": [ "data_preparation = pyearthtools.pipeline.Pipeline(\n", - " pyearthtools.data.archive.era5lowres([\"2m_temperature\", \"u\", \"v\", \"geopotential\", \"vorticity\"]),\n", + " pyearthtools.data.download.weatherbench.WB2ERA5(\n", + " variables=[\"2m_temperature\", \"u\", \"v\", \"geopotential\", \"vorticity\"],\n", + " level=[850],\n", + " download_dir=workdir / \"download\",\n", + " license_ok=True,\n", + " ),\n", " pyearthtools.pipeline.operations.xarray.Sort(\n", - " [\"t2\", \"u\", \"v\", \"vorticity\", \"geopotential\"]\n", + " [\"2m_temperature\", \"u_component_of_wind\", \"v_component_of_wind\", \"vorticity\", \"geopotential\"]\n", " ),\n", " pyearthtools.data.transforms.coordinates.StandardLongitude(type=\"0-360\"),\n", " pyearthtools.pipeline.operations.xarray.reshape.CoordinateFlatten(\"level\"),\n", " # retrieve previous/next samples, dt = 1H\n", " pyearthtools.pipeline.modifications.TemporalRetrieval(\n", - " concat=True, samples=((-1, 1), (6, 1))\n", + " concat=True, samples=((0, 1), (6, 1))\n", " ),\n", ")\n", "data_preparation" ] }, - { - "cell_type": "markdown", - "id": "2d3c8cab-c78b-44f1-97b0-b173a739da4b", - "metadata": {}, - "source": [ - "### Explanation of Pipeline Steps\n", - "\n", - "The data preparation step retrieves ERA5 low-resolution data for the variables t2m, u, v, geopotential, and vorticity. The data is sorted in the order of v, u, vorticity, and geopotential. The coordinates are standardized to the 0-360° longitude format, and the level coordinate is flattened. The TemporalRetrieval operation retrieves data from 1 hour before to 5 hours after the reference time. For example, if the reference time is train_start, index 0 of the sample contains the input data (T-1), and index 1 contains the output data (T+5).\n", - "\n", - "The input data is used by the model to make predictions, while the output data represents the true values that the model aims to predict.\n", - "\n", - "Full list of steps:\n", - "\n", - "- Extract the variables `[\"2m_temperature\", \"u\", \"v\", \"geopotential\", \"vorticity\"]` from the era5lowres data.\n", - "- Put these variables in a specific order, namely `[\"t2\", \"u\", \"v\", \"vorticity\", \"vorticity\", \"geopotential\"]`. Note that not all the variables in the ordering list are in the data; variables in the ordering list that we don't have will be skipped. \n", - "- Change the longitude labels to go from 0->360 degrees instead of -180->180 degrees.\n", - "- Sample groups of timesteps for use with recursive forecast training." - ] - }, { "cell_type": "markdown", "id": "1f85544d", @@ -703,11 +774,10 @@ "source": [ "### Using the pipeline\n", "\n", - "To make use of a pipeline object, we simply pass a datetime value to it. This allows the pipeline to process and retrieve the relevant data for the specified datetime.\n", + "To make use of a pipeline object, we simply pass a datetime value to it.\n", + "This allows the pipeline to process and retrieve the relevant data for the specified datetime.\n", "\n", - "We asked the data_preparation pipeline to develop 2 datasets for our machine learning project. \n", - "1. Input data of -1hr using the (-1, 1) tuple\n", - "2. Target data of +5hr using the (6, 1) tuple\n" + "The `data_preparation` pipeline to return 2 elements for each datetime: input data at T+0hr and target data at T+6hrs." ] }, { @@ -720,50 +790,17 @@ "name": "stdout", "output_type": "stream", "text": [ - "Train start: 2015-01-01T00\n", - "Input sample: ['2014-12-31T23:00:00.000000000']\n", - "Target sample: ['2015-01-01T05:00:00.000000000']\n" + "Train start: 2013-01-01T00\n", + "Input sample: ['2013-01-01T00:00:00.000000000']\n", + "Target sample: ['2013-01-01T06:00:00.000000000']\n" ] } ], "source": [ - "# This code shows how we can pass a date to the data_preparation pipeline and get the corresponding input and target samples.\n", - "# We can then index on this object to get the input and target samples using [0] for input samples and [1] for target samples.\n", - "\n", - "input_and_target_samples = data_preparation[train_start]\n", - "\n", + "input_sample, target_sample = data_preparation[train_start]\n", "print(\"Train start:\", train_start)\n", - "print(\"Input sample:\", input_and_target_samples[0].time.data)\n", - "print(\"Target sample:\", input_and_target_samples[1].time.data)\n", - "\n", - "# Note how the Input sample time is -1hr behind the train start time and the Target sample time is 5hrs ahead of the train start time." - ] - }, - { - "cell_type": "code", - "execution_count": 5, - "id": "39f2eae8", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Train start: 2015-01-01T00\n", - "X_sample: ['2014-12-31T23:00:00.000000000']\n", - "y_sample: ['2015-01-01T05:00:00.000000000']\n" - ] - } - ], - "source": [ - "# We could also split the data into X Input data and y Target data as follows:\n", - "print(\"Train start:\", train_start)\n", - "\n", - "X_sample = data_preparation[train_start][0]\n", - "print(\"X_sample:\", X_sample.time.data)\n", - "\n", - "y_sample = data_preparation[train_start][1]\n", - "print(\"y_sample:\", y_sample.time.data)" + "print(\"Input sample:\", input_sample.time.data)\n", + "print(\"Target sample:\", target_sample.time.data)" ] }, { @@ -771,18 +808,14 @@ "id": "e2818dcf", "metadata": {}, "source": [ + "## Complete Pipeline\n", + "\n", "### NumPy Conversion\n", + "\n", "The following pipeline steps are similar to the previous ones but include additional steps for converting the data to NumPy. This conversion is necessary to prepare the data for ML training.\n", "\n", - "1. Select the variables: \"2m_temperature\", \"u\", \"v\", \"geopotential\", \"vorticity\"\n", - "2. Sort the variables into the order: \"t2\", \"u\", \"v\", \"vorticity\", \"geopotential\"\n", - "3. Transform coordinates to ensure longitude values are 0-360 degress (not -180-180 degrees)\n", - "4. Flatten the data by the 'level' coordinate.\n", - "5. Use TemporalRetrieval to create the:\n", - " * Input data: (-1, 1) tuple is used to select a single timestamp at -1hrs\n", - " * Target data: (6, 1) tuple is used to select a single timestamp at +5hrs ahead. \n", + "Additional steps are:\n", "\n", - "### Additional Steps:\n", "1. Export to a NumPy array.\n", "2. Rearrange the axes of the NumPy array.\n", "3. Remove dimensions of size 1 via a \"squeeze\" operation." @@ -790,7 +823,7 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 5, "id": "e0d6eb27", "metadata": {}, "outputs": [ @@ -1189,14 +1222,14 @@ "\t\t iterator None\n", "\t\t sampler None\n", "\tSteps \n", - "\t\t ERA5DataClass.ERA5LowResIndex {'ERA5LowResIndex': {'level_value': 'None', 'variables': "['2m_temperature', 'u', 'v', 'geopotential', 'vorticity']"}}\n", - "\t\t sort.Sort {'Sort': {'order': "['t2', 'u', 'v', '2t', 'geopotential', 'vorticity']", 'strict': 'False'}}\n", + "\t\t weatherbench.WB2ERA5 {'WB2ERA5': {'download_dir': "PosixPath('cnn_training/download')", 'level': '[850]', 'license_ok': 'True', 'resolution': "'64x32'", 'variables': "['2m_temperature', 'u', 'v', 'geopotential', 'vorticity']"}}\n", + "\t\t sort.Sort {'Sort': {'order': "['2m_temperature', 'u_component_of_wind', 'v_component_of_wind', 'vorticity', 'geopotential']", 'strict': 'False'}}\n", "\t\t coordinates.StandardLongitude {'StandardLongitude': {'longitude_name': "'longitude'", 'type': "'0-360'"}}\n", "\t\t reshape.CoordinateFlatten {'CoordinateFlatten': {'__args': '()', 'coordinate': "'level'", 'skip_missing': 'False'}}\n", - "\t\t idx_modification.TemporalRetrieval {'TemporalRetrieval': {'concat': 'True', 'delta_unit': 'None', 'merge_function': 'None', 'merge_kwargs': 'None', 'samples': '((-1, 1), (6, 1))'}}\n", + "\t\t idx_modification.TemporalRetrieval {'TemporalRetrieval': {'concat': 'True', 'delta_unit': 'None', 'merge_function': 'None', 'merge_kwargs': 'None', 'samples': '((0, 1), (6, 1))'}}\n", "\t\t conversion.ToNumpy {'ToNumpy': {'reference_dataset': 'None', 'run_parallel': 'False', 'saved_records': 'None', 'warn': 'True'}}\n", "\t\t reshape.Rearrange {'Rearrange': {'rearrange': "'c t h w -> t c h w'", 'rearrange_kwargs': 'None', 'reverse_rearrange': 'None', 'skip': 'False'}}\n", - "\t\t reshape.Squeeze {'Squeeze': {'axis': '0'}}" + "\t\t reshape.Squeeze {'Squeeze': {'axis': '0'}}" ], "text/plain": [ "" @@ -1223,109 +1256,108 @@ "\n", "\n", - "\n", - "\n", - "\n", + "\n", + "\n", "\n", - "%3\n", - "\n", - "\n", + "\n", + "\n", "\n", - "ERA5LowResIndex_14fbc7e4-dff9-4e50-8d71-28d54a4508de\n", - "\n", - "ERA5DataClass.ERA5LowResIndex\n", + "WB2ERA5_16e58a28-cfc1-4ef2-8cf4-32a8b8ccdf0e\n", + "\n", + "weatherbench.WB2ERA5\n", "\n", - "\n", + "\n", "\n", - "Sort_52a30d70-27ca-4c5e-b78c-7fc4681a1170\n", - "\n", - "sort.Sort\n", + "Sort_74df188c-dbb1-46c4-9fcf-ee0787d6c599\n", + "\n", + "sort.Sort\n", "\n", - "\n", + "\n", "\n", - "ERA5LowResIndex_14fbc7e4-dff9-4e50-8d71-28d54a4508de->Sort_52a30d70-27ca-4c5e-b78c-7fc4681a1170\n", - "\n", - "\n", + "WB2ERA5_16e58a28-cfc1-4ef2-8cf4-32a8b8ccdf0e->Sort_74df188c-dbb1-46c4-9fcf-ee0787d6c599\n", + "\n", + "\n", "\n", - "\n", + "\n", "\n", - "StandardLongitude_c9816abe-91a1-4b57-8324-c874d7381d04\n", - "\n", - "coordinates.StandardLongitude\n", + "StandardLongitude_93ad7937-59b6-4c94-80ea-279481d274fe\n", + "\n", + "coordinates.StandardLongitude\n", "\n", - "\n", + "\n", "\n", - "Sort_52a30d70-27ca-4c5e-b78c-7fc4681a1170->StandardLongitude_c9816abe-91a1-4b57-8324-c874d7381d04\n", - "\n", - "\n", + "Sort_74df188c-dbb1-46c4-9fcf-ee0787d6c599->StandardLongitude_93ad7937-59b6-4c94-80ea-279481d274fe\n", + "\n", + "\n", "\n", - "\n", + "\n", "\n", - "CoordinateFlatten_4a9df9f3-0443-475a-902d-d95ab2e63dc5\n", - "\n", - "reshape.CoordinateFlatten\n", + "CoordinateFlatten_3180b917-0e3a-4c97-9cad-44d935f856d9\n", + "\n", + "reshape.CoordinateFlatten\n", "\n", - "\n", + "\n", "\n", - "StandardLongitude_c9816abe-91a1-4b57-8324-c874d7381d04->CoordinateFlatten_4a9df9f3-0443-475a-902d-d95ab2e63dc5\n", - "\n", - "\n", + "StandardLongitude_93ad7937-59b6-4c94-80ea-279481d274fe->CoordinateFlatten_3180b917-0e3a-4c97-9cad-44d935f856d9\n", + "\n", + "\n", "\n", - "\n", + "\n", "\n", - "TemporalRetrieval_80452c9e-496c-415b-9eb2-29b2fec83aa3\n", - "\n", - "idx_modification.TemporalRetrieval\n", + "TemporalRetrieval_7fed1ec9-eb28-4346-82c3-071b71b005b5\n", + "\n", + "idx_modification.TemporalRetrieval\n", "\n", - "\n", + "\n", "\n", - "CoordinateFlatten_4a9df9f3-0443-475a-902d-d95ab2e63dc5->TemporalRetrieval_80452c9e-496c-415b-9eb2-29b2fec83aa3\n", - "\n", - "\n", + "CoordinateFlatten_3180b917-0e3a-4c97-9cad-44d935f856d9->TemporalRetrieval_7fed1ec9-eb28-4346-82c3-071b71b005b5\n", + "\n", + "\n", "\n", - "\n", + "\n", "\n", - "ToNumpy_34a23283-ca64-4a9c-914f-a6ff9414962a\n", - "\n", - "conversion.ToNumpy\n", + "ToNumpy_ec93826c-7510-44e5-b24c-e3d550972ca9\n", + "\n", + "conversion.ToNumpy\n", "\n", - "\n", + "\n", "\n", - "TemporalRetrieval_80452c9e-496c-415b-9eb2-29b2fec83aa3->ToNumpy_34a23283-ca64-4a9c-914f-a6ff9414962a\n", - "\n", - "\n", + "TemporalRetrieval_7fed1ec9-eb28-4346-82c3-071b71b005b5->ToNumpy_ec93826c-7510-44e5-b24c-e3d550972ca9\n", + "\n", + "\n", "\n", - "\n", + "\n", "\n", - "Rearrange_b65354ef-9d61-464c-bfda-80acf49542e0\n", - "\n", - "reshape.Rearrange\n", + "Rearrange_8e0de5fb-3e3e-4e01-9fac-2a9deef00d4e\n", + "\n", + "reshape.Rearrange\n", "\n", - "\n", + "\n", "\n", - "ToNumpy_34a23283-ca64-4a9c-914f-a6ff9414962a->Rearrange_b65354ef-9d61-464c-bfda-80acf49542e0\n", - "\n", - "\n", + "ToNumpy_ec93826c-7510-44e5-b24c-e3d550972ca9->Rearrange_8e0de5fb-3e3e-4e01-9fac-2a9deef00d4e\n", + "\n", + "\n", "\n", - "\n", + "\n", "\n", - "Squeeze_e227297a-01ab-41a7-b5a1-2c832f5208f9\n", - "\n", - "reshape.Squeeze\n", + "Squeeze_abb7c579-afc6-4385-b69c-04b76d8dec61\n", + "\n", + "reshape.Squeeze\n", "\n", - "\n", + "\n", "\n", - "Rearrange_b65354ef-9d61-464c-bfda-80acf49542e0->Squeeze_e227297a-01ab-41a7-b5a1-2c832f5208f9\n", - "\n", - "\n", + "Rearrange_8e0de5fb-3e3e-4e01-9fac-2a9deef00d4e->Squeeze_abb7c579-afc6-4385-b69c-04b76d8dec61\n", + "\n", + "\n", "\n", "\n", "\n" ], "text/plain": [ - "" + "" ] }, "metadata": {}, @@ -1334,16 +1366,20 @@ ], "source": [ "data_preparation = pyearthtools.pipeline.Pipeline(\n", - " pyearthtools.data.archive.era5lowres([\"2m_temperature\", \"u\", \"v\", \"geopotential\", \"vorticity\"]),\n", + " pyearthtools.data.download.weatherbench.WB2ERA5(\n", + " variables=[\"2m_temperature\", \"u\", \"v\", \"geopotential\", \"vorticity\"],\n", + " level=[850],\n", + " download_dir=workdir / \"download\",\n", + " license_ok=True,\n", + " ),\n", " pyearthtools.pipeline.operations.xarray.Sort(\n", - " [\"t2\", \"u\", \"v\", \"2t\", \"geopotential\", \"vorticity\"]\n", + " [\"2m_temperature\", \"u_component_of_wind\", \"v_component_of_wind\", \"vorticity\", \"geopotential\"]\n", " ),\n", - " #FIXME: standard_longitude needs updating to StandardLongitude class.\n", " pyearthtools.data.transforms.coordinates.StandardLongitude(type=\"0-360\"),\n", " pyearthtools.pipeline.operations.xarray.reshape.CoordinateFlatten(\"level\"),\n", " # retrieve previous/next samples, dt = 1H\n", " pyearthtools.pipeline.modifications.TemporalRetrieval(\n", - " concat=True, samples=((-1, 1), (6, 1))\n", + " concat=True, samples=((0, 1), (6, 1))\n", " ),\n", " pyearthtools.pipeline.operations.xarray.conversion.ToNumpy(),\n", " pyearthtools.pipeline.operations.numpy.reshape.Rearrange(\"c t h w -> t c h w\"),\n", @@ -1358,14 +1394,15 @@ "metadata": {}, "source": [ "We can again provide a datetime to the pipeline to see how it retrieves and processes the corresponding data.
\n", - "Remember we can index on the sample data. \n", + "Remember we can index on the sample data:\n", + "\n", "* [0] for input data \n", "* [1] for target data" ] }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 6, "id": "7e4c4ff2-78ac-4ab8-bce2-ffbcadb202e3", "metadata": { "tags": [] @@ -1376,14 +1413,13 @@ "output_type": "stream", "text": [ "Number of samples: 2\n", - "Input data shape: (53, 32, 64)\n", - "Target data shape: (53, 32, 64)\n" + "Input data shape: (5, 64, 32)\n", + "Target data shape: (5, 64, 32)\n" ] } ], "source": [ "# Create a sample by passing the data_preparation a datetime. \n", - "\n", "sample = data_preparation[train_start]\n", "print(\"Number of samples:\", len(sample))\n", "print(\"Input data shape:\", sample[0].shape)\n", @@ -1399,17 +1435,24 @@ "\n", "We need to split our data into **training**, **test** and **validation** splits.
\n", "We can use the PyEathTools DateRange object to split our data.
\n", - "We can randomise our training data by chaining the .randomise method." + "We can randomise our training data by chaining the `.randomise` method." ] }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 7, "id": "54c4baca-afe0-4450-9371-7aa57fb9e39b", "metadata": { "tags": [] }, "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Calculated indexes\n" + ] + }, { "data": { "text/html": [ @@ -1796,25 +1839,25 @@ " stroke: currentColor;\n", " fill: currentColor;\n", "}\n", - "
Randomise\n",
+       "
DateRandomise\n",
        "\tInitialisation                 Wrap around another `Iterator` and randomly sample\n",
-       "\t\t iterator                       {'DateRange': {'end': "'2015-01-12T00'", 'interval': "'1h'", 'start': "'2015-01-01T00'"}}\n",
-       "\t\t seed                           42
"
+       "\t\t iterator                       {'DateRange': {'allowlist': 'None', 'blocklist': 'None', 'end': "'2015-01-12T00'", 'interval': "'6h'", 'start': "'2013-01-01T00'"}}\n",
+       "\t\t seed                           42
" ], "text/plain": [ - "Randomise\n", + "DateRandomise\n", "\tInitialisation Wrap around another `Iterator` and randomly sample\n", - "\t\t iterator {'DateRange': {'end': \"'2015-01-12T00'\", 'interval': \"'1h'\", 'start': \"'2015-01-01T00'\"}}\n", + "\t\t iterator {'DateRange': {'allowlist': 'None', 'blocklist': 'None', 'end': \"'2015-01-12T00'\", 'interval': \"'6h'\", 'start': \"'2013-01-01T00'\"}}\n", "\t\t seed 42" ] }, - "execution_count": 8, + "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ - "train_split = pyearthtools.pipeline.iterators.DateRange(train_start, train_end, interval=\"1h\").randomise(seed=42)\n", + "train_split = pyearthtools.pipeline.iterators.DateRange(train_start, train_end, interval=\"6h\").randomise(seed=42)\n", "train_split" ] }, @@ -1828,7 +1871,7 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 8, "id": "78343163", "metadata": {}, "outputs": [ @@ -2220,31 +2263,35 @@ "}\n", "
DateRange\n",
        "\tInitialisation                 DateRange Iterator\n",
+       "\t\t allowlist                      None\n",
+       "\t\t blocklist                      None\n",
        "\t\t end                            '2016-01-12T00'\n",
-       "\t\t interval                       '1h'\n",
-       "\t\t start                          '2016-01-01T00'
" + "\t\t interval '6h'\n", + "\t\t start '2016-01-01T00'" ], "text/plain": [ "DateRange\n", "\tInitialisation DateRange Iterator\n", + "\t\t allowlist None\n", + "\t\t blocklist None\n", "\t\t end '2016-01-12T00'\n", - "\t\t interval '1h'\n", + "\t\t interval '6h'\n", "\t\t start '2016-01-01T00'" ] }, - "execution_count": 9, + "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ - "val_split = pyearthtools.pipeline.iterators.DateRange(val_start, val_end, interval=\"1h\")\n", + "val_split = pyearthtools.pipeline.iterators.DateRange(val_start, val_end, interval=\"6h\")\n", "val_split" ] }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 9, "id": "927ca606-6031-44b4-a646-4b88528f3f1a", "metadata": { "tags": [] @@ -2254,8 +2301,8 @@ "name": "stdout", "output_type": "stream", "text": [ - "Train Splits: (Petdt('2015-01-01T23'), Petdt('2015-01-09T12'), Petdt('2015-01-08T04'), Petdt('2015-01-05T19'), Petdt('2015-01-05T17'))\n", - "Val Splits: (Petdt('2016-01-01T00'), Petdt('2016-01-01T01'), Petdt('2016-01-01T02'), Petdt('2016-01-01T03'), Petdt('2016-01-01T04'))\n" + "Train Splits: (Petdt('2013-04-11T12'), Petdt('2013-09-10T00'), Petdt('2014-03-19T00'), Petdt('2013-10-18T00'), Petdt('2013-04-16T00'))\n", + "Val Splits: (Petdt('2016-01-01T00'), Petdt('2016-01-01T06'), Petdt('2016-01-01T12'), Petdt('2016-01-01T18'), Petdt('2016-01-02T00'))\n" ] } ], @@ -2271,12 +2318,12 @@ "source": [ "### Data normalisation\n", "\n", - "We use approximate mean and standard deviation, computed from only few random samples, to rescale the input/output data to a reasonable range for model training." + "We compute the mean and standard deviation of each field to rescale the input/output data to a reasonable range for model training." ] }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 10, "id": "e6b69b1c-780f-4890-bf00-8409026205bd", "metadata": { "tags": [] @@ -2286,60 +2333,47 @@ "name": "stdout", "output_type": "stream", "text": [ - "CPU times: user 1.48 ms, sys: 322 μs, total: 1.8 ms\n", - "Wall time: 2.09 ms\n" + "CPU times: user 1min 56s, sys: 2.67 s, total: 1min 59s\n", + "Wall time: 2min 1s\n" ] } ], "source": [ "%%time\n", - "stats_folder = Path('stats_folder')\n", - "stats_folder.mkdir(parents=True, exist_ok=True)\n", - "\n", - "mean_path = stats_folder / \"mean.npy\"\n", - "std_path = stats_folder / \"std.npy\"\n", - "\n", - "# Flag to control whether to recompute mean and std\n", - "recompute_stats = False # Set to True if you want to recompute even if files exist (useful if variables have changed)\n", + "samples = np.stack([data_preparation[i][0] for i in train_split])\n", "\n", - "# Compute mean/std only if files are missing or if recompute_stats is True\n", - "if recompute_stats or not mean_path.is_file() or not std_path.is_file():\n", + "mean_path = workdir / \"mean.npy\"\n", + "np.save(mean_path, np.mean(samples, axis=0))\n", "\n", - " samples = np.stack([data_preparation[train_split[i]][0] for i in range(n_samples)])\n", - " mean_approx = np.mean(samples, axis=0)\n", - " std_approx = np.std(samples, axis=0)\n", - " np.save(mean_path, mean_approx)\n", - " np.save(std_path, std_approx)" + "std_path = workdir / \"std.npy\"\n", + "np.save(std_path, np.std(samples, axis=0))" ] }, { "cell_type": "code", - "execution_count": 12, + "execution_count": 11, "id": "d97307a5", "metadata": {}, "outputs": [], "source": [ "# Initialise the normaliser with mean and standard deviation paths\n", "normaliser = pyearthtools.pipeline.operations.numpy.normalisation.Deviation(\n", - " mean=mean_path, \n", - " deviation=std_path, \n", + " mean=mean_path,\n", + " deviation=std_path,\n", " expand=False\n", ")" ] }, { "cell_type": "code", - "execution_count": 14, + "execution_count": 12, "id": "53d89dcd", "metadata": {}, "outputs": [], "source": [ "# Set up the caching mechanism to store processed data in the specified folder with .npy extension\n", - "cache_folder = Path('cache_folder')\n", - "cache_folder.mkdir(parents=True, exist_ok=True)\n", - "\n", "caching_step = pyearthtools.pipeline.modifications.Cache(\n", - " cache_folder, \n", + " workdir / \"cache\", \n", " pattern_kwargs={'extension': 'npy'},\n", " # cache_validity='delete'\n", ")" @@ -2347,22 +2381,22 @@ }, { "cell_type": "code", - "execution_count": 15, + "execution_count": 13, "id": "bf352aa8", "metadata": {}, "outputs": [], "source": [ "# Initialise the data preparation pipeline with normalization and caching steps\n", "data_preparation_normed = pyearthtools.pipeline.Pipeline(\n", - " data_preparation, \n", - " normaliser, \n", + " data_preparation,\n", + " normaliser,\n", " caching_step\n", ")" ] }, { "cell_type": "code", - "execution_count": 16, + "execution_count": 14, "id": "e3d6e689-f5c7-4b6b-a88e-d7d12c8997fa", "metadata": { "tags": [] @@ -2763,16 +2797,16 @@ "\t\t iterator None\n", "\t\t sampler None\n", "\tSteps \n", - "\t\t ERA5DataClass.ERA5LowResIndex {'ERA5LowResIndex': {'level_value': 'None', 'variables': "['2m_temperature', 'u', 'v', 'geopotential', 'vorticity']"}}\n", - "\t\t sort.Sort {'Sort': {'order': "['t2', 'u', 'v', '2t', 'geopotential', 'vorticity']", 'strict': 'False'}}\n", + "\t\t weatherbench.WB2ERA5 {'WB2ERA5': {'download_dir': "PosixPath('cnn_training/download')", 'level': '[850]', 'license_ok': 'True', 'resolution': "'64x32'", 'variables': "['2m_temperature', 'u', 'v', 'geopotential', 'vorticity']"}}\n", + "\t\t sort.Sort {'Sort': {'order': "['2m_temperature', 'u_component_of_wind', 'v_component_of_wind', 'vorticity', 'geopotential']", 'strict': 'False'}}\n", "\t\t coordinates.StandardLongitude {'StandardLongitude': {'longitude_name': "'longitude'", 'type': "'0-360'"}}\n", "\t\t reshape.CoordinateFlatten {'CoordinateFlatten': {'__args': '()', 'coordinate': "'level'", 'skip_missing': 'False'}}\n", - "\t\t idx_modification.TemporalRetrieval {'TemporalRetrieval': {'concat': 'True', 'delta_unit': 'None', 'merge_function': 'None', 'merge_kwargs': 'None', 'samples': '((-1, 1), (6, 1))'}}\n", + "\t\t idx_modification.TemporalRetrieval {'TemporalRetrieval': {'concat': 'True', 'delta_unit': 'None', 'merge_function': 'None', 'merge_kwargs': 'None', 'samples': '((0, 1), (6, 1))'}}\n", "\t\t conversion.ToNumpy {'ToNumpy': {'reference_dataset': 'None', 'run_parallel': 'False', 'saved_records': 'None', 'warn': 'True'}}\n", "\t\t reshape.Rearrange {'Rearrange': {'rearrange': "'c t h w -> t c h w'", 'rearrange_kwargs': 'None', 'reverse_rearrange': 'None', 'skip': 'False'}}\n", "\t\t reshape.Squeeze {'Squeeze': {'axis': '0'}}\n", - "\t\t normalisation.Deviation {'Deviation': {'deviation': "PosixPath('stats_folder/std.npy')", 'expand': 'False', 'mean': "PosixPath('stats_folder/mean.npy')"}}\n", - "\t\t cache.Cache {'Cache': {'cache': "'/home/548/tjl548/cache_folder'", 'cache_validity': "'warn'", 'pattern': 'None', 'pattern_kwargs': {'extension': "'npy'"}, 'save_kwargs': 'None'}}" + "\t\t normalisation.Deviation {'Deviation': {'deviation': "PosixPath('cnn_training/std.npy')", 'expand': 'False', 'mean': "PosixPath('cnn_training/mean.npy')"}}\n", + "\t\t cache.Cache {'Cache': {'cache': "'/var/home/riomaxim/Synced/work/en_cours/PyEarthTools/notebooks/tutorial/cnn_training/cache'", 'cache_validity': "'warn'", 'pattern': 'None', 'pattern_kwargs': {'extension': "'npy'"}, 'save_kwargs': 'None'}}" ], "text/plain": [ "" @@ -2799,133 +2833,132 @@ "\n", "\n", - "\n", - "\n", - "\n", + "\n", + "\n", "\n", - "%3\n", - "\n", - "\n", + "\n", + "\n", "\n", - "ERA5LowResIndex_16e00009-fdc5-4aeb-8e19-b84516880f92\n", - "\n", - "ERA5DataClass.ERA5LowResIndex\n", + "WB2ERA5_d59db7d0-456a-4fb3-b25f-8170469163fd\n", + "\n", + "weatherbench.WB2ERA5\n", "\n", - "\n", + "\n", "\n", - "Sort_ab70da88-efaa-42e7-a5c0-11386dd1df12\n", - "\n", - "sort.Sort\n", + "Sort_e629df9b-1c27-46cb-a616-ee7489df7416\n", + "\n", + "sort.Sort\n", "\n", - "\n", + "\n", "\n", - "ERA5LowResIndex_16e00009-fdc5-4aeb-8e19-b84516880f92->Sort_ab70da88-efaa-42e7-a5c0-11386dd1df12\n", - "\n", - "\n", + "WB2ERA5_d59db7d0-456a-4fb3-b25f-8170469163fd->Sort_e629df9b-1c27-46cb-a616-ee7489df7416\n", + "\n", + "\n", "\n", - "\n", + "\n", "\n", - "StandardLongitude_3ca165d4-ea1c-40b9-8316-61ba5e6e93aa\n", - "\n", - "coordinates.StandardLongitude\n", + "StandardLongitude_0987678c-90a5-499f-958d-e49f56f1f2e7\n", + "\n", + "coordinates.StandardLongitude\n", "\n", - "\n", + "\n", "\n", - "Sort_ab70da88-efaa-42e7-a5c0-11386dd1df12->StandardLongitude_3ca165d4-ea1c-40b9-8316-61ba5e6e93aa\n", - "\n", - "\n", + "Sort_e629df9b-1c27-46cb-a616-ee7489df7416->StandardLongitude_0987678c-90a5-499f-958d-e49f56f1f2e7\n", + "\n", + "\n", "\n", - "\n", + "\n", "\n", - "CoordinateFlatten_e39d876a-7ded-4e2d-a0dd-b07cf1d756f7\n", - "\n", - "reshape.CoordinateFlatten\n", + "CoordinateFlatten_f85176dd-52bc-4981-adaf-fdc3966a6e75\n", + "\n", + "reshape.CoordinateFlatten\n", "\n", - "\n", + "\n", "\n", - "StandardLongitude_3ca165d4-ea1c-40b9-8316-61ba5e6e93aa->CoordinateFlatten_e39d876a-7ded-4e2d-a0dd-b07cf1d756f7\n", - "\n", - "\n", + "StandardLongitude_0987678c-90a5-499f-958d-e49f56f1f2e7->CoordinateFlatten_f85176dd-52bc-4981-adaf-fdc3966a6e75\n", + "\n", + "\n", "\n", - "\n", + "\n", "\n", - "TemporalRetrieval_0f4e081b-7e24-4a02-94d8-4f11e262fbc5\n", - "\n", - "idx_modification.TemporalRetrieval\n", + "TemporalRetrieval_ccff85ef-e054-4737-8b16-87f0ba459d4e\n", + "\n", + "idx_modification.TemporalRetrieval\n", "\n", - "\n", + "\n", "\n", - "CoordinateFlatten_e39d876a-7ded-4e2d-a0dd-b07cf1d756f7->TemporalRetrieval_0f4e081b-7e24-4a02-94d8-4f11e262fbc5\n", - "\n", - "\n", + "CoordinateFlatten_f85176dd-52bc-4981-adaf-fdc3966a6e75->TemporalRetrieval_ccff85ef-e054-4737-8b16-87f0ba459d4e\n", + "\n", + "\n", "\n", - "\n", + "\n", "\n", - "ToNumpy_bb838d48-4f79-4e53-a66f-807338eee602\n", - "\n", - "conversion.ToNumpy\n", + "ToNumpy_5528cc5d-a408-421b-98cb-cc673183479e\n", + "\n", + "conversion.ToNumpy\n", "\n", - "\n", + "\n", "\n", - "TemporalRetrieval_0f4e081b-7e24-4a02-94d8-4f11e262fbc5->ToNumpy_bb838d48-4f79-4e53-a66f-807338eee602\n", - "\n", - "\n", + "TemporalRetrieval_ccff85ef-e054-4737-8b16-87f0ba459d4e->ToNumpy_5528cc5d-a408-421b-98cb-cc673183479e\n", + "\n", + "\n", "\n", - "\n", + "\n", "\n", - "Rearrange_bebb7dcd-e4e0-4fc0-90b9-83d4a3e69c7f\n", - "\n", - "reshape.Rearrange\n", + "Rearrange_a1490659-c054-451c-8fab-5a452e10cf48\n", + "\n", + "reshape.Rearrange\n", "\n", - "\n", + "\n", "\n", - "ToNumpy_bb838d48-4f79-4e53-a66f-807338eee602->Rearrange_bebb7dcd-e4e0-4fc0-90b9-83d4a3e69c7f\n", - "\n", - "\n", + "ToNumpy_5528cc5d-a408-421b-98cb-cc673183479e->Rearrange_a1490659-c054-451c-8fab-5a452e10cf48\n", + "\n", + "\n", "\n", - "\n", + "\n", "\n", - "Squeeze_0b754735-b505-457b-9a52-8a36024864ac\n", - "\n", - "reshape.Squeeze\n", + "Squeeze_c0a9ca64-3e22-468b-9ce1-3d7046116e3e\n", + "\n", + "reshape.Squeeze\n", "\n", - "\n", + "\n", "\n", - "Rearrange_bebb7dcd-e4e0-4fc0-90b9-83d4a3e69c7f->Squeeze_0b754735-b505-457b-9a52-8a36024864ac\n", - "\n", - "\n", + "Rearrange_a1490659-c054-451c-8fab-5a452e10cf48->Squeeze_c0a9ca64-3e22-468b-9ce1-3d7046116e3e\n", + "\n", + "\n", "\n", - "\n", + "\n", "\n", - "Deviation_1b5dcc42-e84e-4c0c-b195-6ff65c080e43\n", - "\n", - "normalisation.Deviation\n", + "Deviation_7450727a-a384-4040-9904-25b6c30a811f\n", + "\n", + "normalisation.Deviation\n", "\n", - "\n", + "\n", "\n", - "Squeeze_0b754735-b505-457b-9a52-8a36024864ac->Deviation_1b5dcc42-e84e-4c0c-b195-6ff65c080e43\n", - "\n", - "\n", + "Squeeze_c0a9ca64-3e22-468b-9ce1-3d7046116e3e->Deviation_7450727a-a384-4040-9904-25b6c30a811f\n", + "\n", + "\n", "\n", - "\n", + "\n", "\n", - "Cache_cd573b60-f478-4c7f-be38-c9b7f5b5a160\n", - "\n", - "cache.Cache\n", + "Cache_f32d83e4-8dcc-41da-80b4-af99bba804e6\n", + "\n", + "cache.Cache\n", "\n", - "\n", + "\n", "\n", - "Deviation_1b5dcc42-e84e-4c0c-b195-6ff65c080e43->Cache_cd573b60-f478-4c7f-be38-c9b7f5b5a160\n", - "\n", - "\n", + "Deviation_7450727a-a384-4040-9904-25b6c30a811f->Cache_f32d83e4-8dcc-41da-80b4-af99bba804e6\n", + "\n", + "\n", "\n", "\n", "\n" ], "text/plain": [ - "" + "" ] }, "metadata": {}, @@ -2946,7 +2979,7 @@ }, { "cell_type": "code", - "execution_count": 17, + "execution_count": 15, "id": "0d1a4965-b224-472f-9675-4c17f83f548f", "metadata": { "tags": [] @@ -3012,7 +3045,7 @@ }, { "cell_type": "code", - "execution_count": 18, + "execution_count": 16, "id": "ab29d950", "metadata": {}, "outputs": [ @@ -3020,8 +3053,8 @@ "name": "stdout", "output_type": "stream", "text": [ - "(53, 32, 64)\n", - "Number of features: 53\n" + "(5, 64, 32)\n", + "Number of features: 5\n" ] } ], @@ -3036,7 +3069,7 @@ }, { "cell_type": "code", - "execution_count": 19, + "execution_count": 17, "id": "3dd973c3", "metadata": {}, "outputs": [], @@ -3055,7 +3088,7 @@ }, { "cell_type": "code", - "execution_count": 20, + "execution_count": 18, "id": "90fa9f5a-b2b1-4952-a066-831b66d28917", "metadata": { "tags": [] @@ -3066,18 +3099,18 @@ "text/plain": [ "CNN(\n", " (cnn): Sequential(\n", - " (0): Conv2d(53, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n", + " (0): Conv2d(5, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n", " (1): ReLU()\n", " (2): Dropout(p=0.6, inplace=False)\n", " (3): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n", " (4): ReLU()\n", " (5): Dropout(p=0.6, inplace=False)\n", - " (6): Conv2d(64, 53, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n", + " (6): Conv2d(64, 5, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n", " )\n", ")" ] }, - "execution_count": 20, + "execution_count": 18, "metadata": {}, "output_type": "execute_result" } @@ -3088,24 +3121,15 @@ }, { "cell_type": "code", - "execution_count": 21, + "execution_count": 19, "id": "51aa6230-b381-4c04-b231-9cf432148079", "metadata": { "tags": [] }, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "env: CUDA_VISIBLE_DEVICES=\n" - ] - } - ], + "outputs": [], "source": [ - "# Ensure that we use the CPU even if a GPU is available.\n", - "# Uncomment the following line to use your GPU instead.\n", - "%env CUDA_VISIBLE_DEVICES=" + "# Uncomment the following line to use the CPU even if a GPU is available.\n", + "#%env CUDA_VISIBLE_DEVICES=" ] }, { @@ -3131,7 +3155,7 @@ }, { "cell_type": "code", - "execution_count": 22, + "execution_count": 20, "id": "615389bf-2e4d-4cac-abe0-06637d699792", "metadata": { "tags": [] @@ -3152,7 +3176,7 @@ }, { "cell_type": "code", - "execution_count": 23, + "execution_count": 21, "id": "d54e16c5-dfe4-4c4b-bea7-ab7646591b6a", "metadata": { "tags": [] @@ -3551,9 +3575,9 @@ "\t\t multiprocessing_context 'forkserver'\n", "\t\t num_workers 2\n", "\t\t persistent_workers True\n", - "\t\t pipelines {'Pipeline': {'__args': '(Pipeline\\n\\tDescription `pyearthtools.pipeline` Data Pipeline\\n\\n\\n\\tInitialisation \\n\\t\\t exceptions_to_ignore None\\n\\t\\t iterator None\\n\\t\\t sampler None\\n\\tSteps \\n\\t\\t ERA5DataClass.ERA5LowResIndex {\\'ERA5LowResIndex\\': {\\'level_value\\': \\'None\\', \\'variables\\': "[\\'2m_temperature\\', \\'u\\', \\'v\\', \\'geopotential\\', \\'vorticity\\']"}}\\n\\t\\t sort.Sort {\\'Sort\\': {\\'order\\': "[\\'t2\\', \\'u\\', \\'v\\', \\'2t\\', \\'geopotential\\', \\'vorticity\\']", \\'strict\\': \\'False\\'}}\\n\\t\\t coordinates.StandardLongitude {\\'StandardLongitude\\': {\\'longitude_name\\': "\\'longitude\\'", \\'type\\': "\\'0-360\\'"}}\\n\\t\\t reshape.CoordinateFlatten {\\'CoordinateFlatten\\': {\\'__args\\': \\'()\\', \\'coordinate\\': "\\'level\\'", \\'skip_missing\\': \\'False\\'}}\\n\\t\\t idx_modification.TemporalRetrieval {\\'TemporalRetrieval\\': {\\'concat\\': \\'True\\', \\'delta_unit\\': \\'None\\', \\'merge_function\\': \\'None\\', \\'merge_kwargs\\': \\'None\\', \\'samples\\': \\'((-1, 1), (6, 1))\\'}}\\n\\t\\t conversion.ToNumpy {\\'ToNumpy\\': {\\'reference_dataset\\': \\'None\\', \\'run_parallel\\': \\'False\\', \\'saved_records\\': \\'None\\', \\'warn\\': \\'True\\'}}\\n\\t\\t reshape.Rearrange {\\'Rearrange\\': {\\'rearrange\\': "\\'c t h w -> t c h w\\'", \\'rearrange_kwargs\\': \\'None\\', \\'reverse_rearrange\\': \\'None\\', \\'skip\\': \\'False\\'}}\\n\\t\\t reshape.Squeeze {\\'Squeeze\\': {\\'axis\\': \\'0\\'}}, Deviation\\n\\tInitialisation Deviation Normalisation\\n\\t\\t deviation PosixPath(\\'stats_folder/std.npy\\')\\n\\t\\t expand False\\n\\t\\t mean PosixPath(\\'stats_folder/mean.npy\\'), Cache\\n\\tInitialisation An `pyearthtools.pipeline` implementation of the `CachingIndex` from `pyearthtools.data`.\\n\\t\\t cache \\'/home/548/tjl548/cache_folder\\'\\n\\t\\t cache_validity \\'warn\\'\\n\\t\\t pattern None\\n\\t\\t pattern_kwargs {\\'extension\\': "\\'npy\\'"}\\n\\t\\t save_kwargs None)', 'exceptions_to_ignore': 'None', 'iterator': 'None', 'sampler': 'None'}}\n", - "\t\t train_split {'Randomise': {'iterator': {'DateRange': {'end': "'2015-01-12T00'", 'interval': "'1h'", 'start': "'2015-01-01T00'"}}, 'seed': '42'}}\n", - "\t\t valid_split {'DateRange': {'end': "'2016-01-12T00'", 'interval': "'1h'", 'start': "'2016-01-01T00'"}}" ], "text/plain": [ "PipelineLightningDataModule\n", @@ -3615,12 +3639,12 @@ "\t\t multiprocessing_context 'forkserver'\n", "\t\t num_workers 2\n", "\t\t persistent_workers True\n", - "\t\t pipelines {'Pipeline': {'__args': '(Pipeline\\n\\tDescription `pyearthtools.pipeline` Data Pipeline\\n\\n\\n\\tInitialisation \\n\\t\\t exceptions_to_ignore None\\n\\t\\t iterator None\\n\\t\\t sampler None\\n\\tSteps \\n\\t\\t ERA5DataClass.ERA5LowResIndex {\\'ERA5LowResIndex\\': {\\'level_value\\': \\'None\\', \\'variables\\': \"[\\'2m_temperature\\', \\'u\\', \\'v\\', \\'geopotential\\', \\'vorticity\\']\"}}\\n\\t\\t sort.Sort {\\'Sort\\': {\\'order\\': \"[\\'t2\\', \\'u\\', \\'v\\', \\'2t\\', \\'geopotential\\', \\'vorticity\\']\", \\'strict\\': \\'False\\'}}\\n\\t\\t coordinates.StandardLongitude {\\'StandardLongitude\\': {\\'longitude_name\\': \"\\'longitude\\'\", \\'type\\': \"\\'0-360\\'\"}}\\n\\t\\t reshape.CoordinateFlatten {\\'CoordinateFlatten\\': {\\'__args\\': \\'()\\', \\'coordinate\\': \"\\'level\\'\", \\'skip_missing\\': \\'False\\'}}\\n\\t\\t idx_modification.TemporalRetrieval {\\'TemporalRetrieval\\': {\\'concat\\': \\'True\\', \\'delta_unit\\': \\'None\\', \\'merge_function\\': \\'None\\', \\'merge_kwargs\\': \\'None\\', \\'samples\\': \\'((-1, 1), (6, 1))\\'}}\\n\\t\\t conversion.ToNumpy {\\'ToNumpy\\': {\\'reference_dataset\\': \\'None\\', \\'run_parallel\\': \\'False\\', \\'saved_records\\': \\'None\\', \\'warn\\': \\'True\\'}}\\n\\t\\t reshape.Rearrange {\\'Rearrange\\': {\\'rearrange\\': \"\\'c t h w -> t c h w\\'\", \\'rearrange_kwargs\\': \\'None\\', \\'reverse_rearrange\\': \\'None\\', \\'skip\\': \\'False\\'}}\\n\\t\\t reshape.Squeeze {\\'Squeeze\\': {\\'axis\\': \\'0\\'}}, Deviation\\n\\tInitialisation Deviation Normalisation\\n\\t\\t deviation PosixPath(\\'stats_folder/std.npy\\')\\n\\t\\t expand False\\n\\t\\t mean PosixPath(\\'stats_folder/mean.npy\\'), Cache\\n\\tInitialisation An `pyearthtools.pipeline` implementation of the `CachingIndex` from `pyearthtools.data`.\\n\\t\\t cache \\'/home/548/tjl548/cache_folder\\'\\n\\t\\t cache_validity \\'warn\\'\\n\\t\\t pattern None\\n\\t\\t pattern_kwargs {\\'extension\\': \"\\'npy\\'\"}\\n\\t\\t save_kwargs None)', 'exceptions_to_ignore': 'None', 'iterator': 'None', 'sampler': 'None'}}\n", - "\t\t train_split {'Randomise': {'iterator': {'DateRange': {'end': \"'2015-01-12T00'\", 'interval': \"'1h'\", 'start': \"'2015-01-01T00'\"}}, 'seed': '42'}}\n", - "\t\t valid_split {'DateRange': {'end': \"'2016-01-12T00'\", 'interval': \"'1h'\", 'start': \"'2016-01-01T00'\"}}" + "\t\t pipelines {'Pipeline': {'__args': '(Pipeline\\n\\tDescription `pyearthtools.pipeline` Data Pipeline\\n\\n\\n\\tInitialisation \\n\\t\\t exceptions_to_ignore None\\n\\t\\t iterator None\\n\\t\\t sampler None\\n\\tSteps \\n\\t\\t weatherbench.WB2ERA5 {\\'WB2ERA5\\': {\\'download_dir\\': \"PosixPath(\\'cnn_training/download\\')\", \\'level\\': \\'[850]\\', \\'license_ok\\': \\'True\\', \\'resolution\\': \"\\'64x32\\'\", \\'variables\\': \"[\\'2m_temperature\\', \\'u\\', \\'v\\', \\'geopotential\\', \\'vorticity\\']\"}}\\n\\t\\t sort.Sort {\\'Sort\\': {\\'order\\': \"[\\'2m_temperature\\', \\'u_component_of_wind\\', \\'v_component_of_wind\\', \\'vorticity\\', \\'geopotential\\']\", \\'strict\\': \\'False\\'}}\\n\\t\\t coordinates.StandardLongitude {\\'StandardLongitude\\': {\\'longitude_name\\': \"\\'longitude\\'\", \\'type\\': \"\\'0-360\\'\"}}\\n\\t\\t reshape.CoordinateFlatten {\\'CoordinateFlatten\\': {\\'__args\\': \\'()\\', \\'coordinate\\': \"\\'level\\'\", \\'skip_missing\\': \\'False\\'}}\\n\\t\\t idx_modification.TemporalRetrieval {\\'TemporalRetrieval\\': {\\'concat\\': \\'True\\', \\'delta_unit\\': \\'None\\', \\'merge_function\\': \\'None\\', \\'merge_kwargs\\': \\'None\\', \\'samples\\': \\'((0, 1), (6, 1))\\'}}\\n\\t\\t conversion.ToNumpy {\\'ToNumpy\\': {\\'reference_dataset\\': \\'None\\', \\'run_parallel\\': \\'False\\', \\'saved_records\\': \\'None\\', \\'warn\\': \\'True\\'}}\\n\\t\\t reshape.Rearrange {\\'Rearrange\\': {\\'rearrange\\': \"\\'c t h w -> t c h w\\'\", \\'rearrange_kwargs\\': \\'None\\', \\'reverse_rearrange\\': \\'None\\', \\'skip\\': \\'False\\'}}\\n\\t\\t reshape.Squeeze {\\'Squeeze\\': {\\'axis\\': \\'0\\'}}, Deviation\\n\\tInitialisation Deviation Normalisation\\n\\t\\t deviation PosixPath(\\'cnn_training/std.npy\\')\\n\\t\\t expand False\\n\\t\\t mean PosixPath(\\'cnn_training/mean.npy\\'), Cache\\n\\tInitialisation An `pyearthtools.pipeline` implementation of the `CachingIndex` from `pyearthtools.data`.\\n\\t\\t cache \\'/var/home/riomaxim/Synced/work/en_cours/PyEarthTools/notebooks/tutorial/cnn_training/cache\\'\\n\\t\\t cache_validity \\'warn\\'\\n\\t\\t pattern None\\n\\t\\t pattern_kwargs {\\'extension\\': \"\\'npy\\'\"}\\n\\t\\t save_kwargs None)', 'exceptions_to_ignore': 'None', 'iterator': 'None', 'sampler': 'None'}}\n", + "\t\t train_split {'DateRandomise': {'iterator': {'DateRange': {'allowlist': 'None', 'blocklist': 'None', 'end': \"'2015-01-12T00'\", 'interval': \"'6h'\", 'start': \"'2013-01-01T00'\"}}, 'seed': '42'}}\n", + "\t\t valid_split {'DateRange': {'allowlist': 'None', 'blocklist': 'None', 'end': \"'2016-01-12T00'\", 'interval': \"'6h'\", 'start': \"'2016-01-01T00'\"}}" ] }, - "execution_count": 23, + "execution_count": 21, "metadata": {}, "output_type": "execute_result" } @@ -3631,22 +3655,7 @@ }, { "cell_type": "code", - "execution_count": 24, - "id": "e2b41f4f-a13d-45e2-bc25-5b8b6bd6555c", - "metadata": { - "tags": [] - }, - "outputs": [], - "source": [ - "\n", - "checkpoint_dir = Path('checkpoint_dir')\n", - "checkpoint_dir.mkdir(parents=True, exist_ok=True)\n", - "chkpt_path = Path(checkpoint_dir) / \"model.ckpt\"" - ] - }, - { - "cell_type": "code", - "execution_count": 25, + "execution_count": 23, "id": "cf979eb0-7966-4a78-b19e-1f434981803f", "metadata": { "tags": [] @@ -3656,10 +3665,11 @@ "name": "stderr", "output_type": "stream", "text": [ - "Using default `ModelCheckpoint`. Consider installing `litmodels` package to enable `LitModelCheckpoint` for automatic upload to the Lightning model registry.\n", - "GPU available: False, used: False\n", + "💡 Tip: For seamless cloud uploads and versioning, try installing [litmodels](https://pypi.org/project/litmodels/) to enable LitModelCheckpoint, which syncs automatically with the Lightning model registry.\n", + "GPU available: True (cuda), used: True\n", "TPU available: False, using: 0 TPU cores\n", - "HPU available: False, using: 0 HPUs\n" + "HPU available: False, using: 0 HPUs\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n" ] }, { @@ -3668,7 +3678,7 @@ "
┏━━━┳━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┓\n",
        "┃   ┃ Name ┃ Type       ┃ Params ┃ Mode  ┃\n",
        "┡━━━╇━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━┩\n",
-       "│ 0 │ cnn  │ Sequential │ 98.1 K │ train │\n",
+       "│ 0 │ cnn  │ Sequential │ 42.8 K │ train │\n",
        "└───┴──────┴────────────┴────────┴───────┘\n",
        "
\n" ], @@ -3676,7 +3686,7 @@ "┏━━━┳━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┓\n", "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\n", "┡━━━╇━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━┩\n", - "│\u001b[2m \u001b[0m\u001b[2m0\u001b[0m\u001b[2m \u001b[0m│ cnn │ Sequential │ 98.1 K │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m0\u001b[0m\u001b[2m \u001b[0m│ cnn │ Sequential │ 42.8 K │ train │\n", "└───┴──────┴────────────┴────────┴───────┘\n" ] }, @@ -3686,18 +3696,18 @@ { "data": { "text/html": [ - "
Trainable params: 98.1 K                                                                                           \n",
+       "
Trainable params: 42.8 K                                                                                           \n",
        "Non-trainable params: 0                                                                                            \n",
-       "Total params: 98.1 K                                                                                               \n",
+       "Total params: 42.8 K                                                                                               \n",
        "Total estimated model params size (MB): 0                                                                          \n",
        "Modules in train mode: 8                                                                                           \n",
        "Modules in eval mode: 0                                                                                            \n",
        "
\n"
       ],
       "text/plain": [
-       "\u001b[1mTrainable params\u001b[0m: 98.1 K                                                                                           \n",
+       "\u001b[1mTrainable params\u001b[0m: 42.8 K                                                                                           \n",
        "\u001b[1mNon-trainable params\u001b[0m: 0                                                                                            \n",
-       "\u001b[1mTotal params\u001b[0m: 98.1 K                                                                                               \n",
+       "\u001b[1mTotal params\u001b[0m: 42.8 K                                                                                               \n",
        "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 0                                                                          \n",
        "\u001b[1mModules in train mode\u001b[0m: 8                                                                                           \n",
        "\u001b[1mModules in eval mode\u001b[0m: 0                                                                                            \n"
@@ -3709,7 +3719,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "a0056d9972644119b82e01d0ed68f021",
+       "model_id": "7f8126acab3448aeb935405d2c851fa1",
        "version_major": 2,
        "version_minor": 0
       },
@@ -3727,6 +3737,14 @@
       "`Trainer.fit` stopped: `max_epochs=10` reached.\n"
      ]
     },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Calculated indexes\n",
+      "Calculated indexes\n"
+     ]
+    },
     {
      "data": {
       "text/html": [
@@ -3741,8 +3759,8 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "CPU times: user 1min 26s, sys: 4.74 s, total: 1min 31s\n",
-      "Wall time: 1min 44s\n"
+      "CPU times: user 2min 6s, sys: 13.1 s, total: 2min 20s\n",
+      "Wall time: 2min 40s\n"
      ]
     }
    ],
@@ -3750,11 +3768,11 @@
     "%%time\n",
     "# Initialise the trainer with the specified parameters\n",
     "trainer = pyearthtools.training.lightning.Train(\n",
-    "    model,                  # The model to be trained\n",
-    "    data_module,            # The data module for training\n",
-    "    checkpoint_dir,       # Directory to save logs and checkpoints\n",
-    "    max_epochs=max_epochs,  # Maximum number of training epochs\n",
-    "    callbacks=[RichProgressBar()]  # Callbacks for training (e.g., progress bar)\n",
+    "    model,                         # The model to be trained\n",
+    "    data_module,                   # The data module for training\n",
+    "    workdir,                       # Directory to save logs and checkpoints\n",
+    "    max_epochs=max_epochs,         # Maximum number of training epochs\n",
+    "    callbacks=[RichProgressBar(refresh_rate=50)]  # Callbacks for training (e.g., progress bar)\n",
     ")\n",
     "\n",
     "# Fit the model\n",
@@ -3771,7 +3789,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 24,
    "id": "b1af821b-0d56-4c05-8d25-04403712f580",
    "metadata": {
     "tags": []
@@ -3791,12 +3809,12 @@
    "id": "d1225d38-c85c-4795-b640-1143213c0208",
    "metadata": {},
    "source": [
-    "The code below initialises a reverse_pipeline by extracting specific steps from an existing data_preparation_pipeline. The data_preparation_pipeline is a sequence of data preprocessing steps applied to the input data before it is fed into the model. These steps might include normalisation, reshaping, and other transformations necessary for preparing the data.
"
+    "The code below initialises a reverse_pipeline by extracting specific steps from an existing data preparation pipeline. The data preparation pipeline is a sequence of data preprocessing steps applied to the input data before it is fed into the model. These steps might include normalisation, reshaping, and other transformations necessary for preparing the data."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 27,
+   "execution_count": 25,
    "id": "5868ae48-4be0-4f87-a429-60cc5a2d5ce1",
    "metadata": {
     "tags": []
@@ -3814,13 +3832,12 @@
    "id": "0807f2f7-5858-4a11-8e31-82468ceda1b1",
    "metadata": {},
    "source": [
-    "\n",
-    "The ReversedPipeline class is designed to reverse the transformations applied by the original pipeline. By passing `data_preparation_pipeline.steps[-5:-1]` to the `ReversedPipeline` constructor, the code extracts the last few steps (from the fifth-to-last to the second-to-last) of the data_preparation_pipeline."
+    "The ReversedPipeline class is designed to reverse the transformations applied by the original pipeline. By passing `data_preparation_pipeline.steps[-5:-1]` to the `ReversedPipeline` constructor, the code extracts the last few steps (from the fifth-to-last to the second-to-last) of the data preparation pipeline."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 28,
+   "execution_count": 26,
    "id": "e32d829b-df99-4f6e-bf14-750ee17e22bb",
    "metadata": {
     "tags": []
@@ -4224,7 +4241,7 @@
        "\t\t conversion.ToNumpy             {'ToNumpy': {'reference_dataset': 'None', 'run_parallel': 'False', 'saved_records': 'None', 'warn': 'True'}}\n",
        "\t\t reshape.Rearrange              {'Rearrange': {'rearrange': "'c t h w -> t c h w'", 'rearrange_kwargs': 'None', 'reverse_rearrange': 'None', 'skip': 'False'}}\n",
        "\t\t reshape.Squeeze                {'Squeeze': {'axis': '0'}}\n",
-       "\t\t normalisation.Deviation        {'Deviation': {'deviation': "PosixPath('stats_folder/std.npy')", 'expand': 'False', 'mean': "PosixPath('stats_folder/mean.npy')"}}
" + "\t\t normalisation.Deviation {'Deviation': {'deviation': "PosixPath('cnn_training/std.npy')", 'expand': 'False', 'mean': "PosixPath('cnn_training/mean.npy')"}}" ], "text/plain": [ "" @@ -4251,61 +4268,60 @@ "\n", "\n", - "\n", - "\n", - "\n", + "\n", + "\n", "\n", - "%3\n", - "\n", - "\n", + "\n", + "\n", "\n", - "ToNumpy_eb58e04c-956f-4844-9368-78f9dec86b7b\n", - "\n", - "conversion.ToNumpy\n", + "ToNumpy_5b6c94f8-8c30-4bf1-8547-62758882cbdc\n", + "\n", + "conversion.ToNumpy\n", "\n", - "\n", + "\n", "\n", - "Rearrange_6fd05d3c-df0a-4235-8836-60fdfc7515b3\n", - "\n", - "reshape.Rearrange\n", + "Rearrange_83209c0f-8950-47fb-9cc5-418fd86d3c87\n", + "\n", + "reshape.Rearrange\n", "\n", - "\n", + "\n", "\n", - "ToNumpy_eb58e04c-956f-4844-9368-78f9dec86b7b->Rearrange_6fd05d3c-df0a-4235-8836-60fdfc7515b3\n", - "\n", - "\n", + "ToNumpy_5b6c94f8-8c30-4bf1-8547-62758882cbdc->Rearrange_83209c0f-8950-47fb-9cc5-418fd86d3c87\n", + "\n", + "\n", "\n", - "\n", + "\n", "\n", - "Squeeze_ac76137d-47c8-4ac5-8334-c53e2486a951\n", - "\n", - "reshape.Squeeze\n", + "Squeeze_83a518b7-af04-469f-886e-1f80824027aa\n", + "\n", + "reshape.Squeeze\n", "\n", - "\n", + "\n", "\n", - "Rearrange_6fd05d3c-df0a-4235-8836-60fdfc7515b3->Squeeze_ac76137d-47c8-4ac5-8334-c53e2486a951\n", - "\n", - "\n", + "Rearrange_83209c0f-8950-47fb-9cc5-418fd86d3c87->Squeeze_83a518b7-af04-469f-886e-1f80824027aa\n", + "\n", + "\n", "\n", - "\n", + "\n", "\n", - "Deviation_a4a9dc8d-3b1f-4668-80ca-d6944f5f2b2a\n", - "\n", - "normalisation.Deviation\n", + "Deviation_167b31cd-6c7d-481e-a711-697ad631684f\n", + "\n", + "normalisation.Deviation\n", "\n", - "\n", + "\n", "\n", - "Squeeze_ac76137d-47c8-4ac5-8334-c53e2486a951->Deviation_a4a9dc8d-3b1f-4668-80ca-d6944f5f2b2a\n", - "\n", - "\n", + "Squeeze_83a518b7-af04-469f-886e-1f80824027aa->Deviation_167b31cd-6c7d-481e-a711-697ad631684f\n", + "\n", + "\n", "\n", "\n", "\n" ], "text/plain": [ - "" + "" ] }, "metadata": {}, @@ -4318,7 +4334,7 @@ }, { "cell_type": "code", - "execution_count": 29, + "execution_count": 27, "id": "072cbd6c-8a82-4213-99fc-811623ad8c41", "metadata": { "tags": [] @@ -4327,7 +4343,7 @@ "source": [ "# Wrap the trained model with the data preparation pipeline\n", "model_wrapper = pyearthtools.training.lightning.Predict(\n", - " model, # The trained CNN model\n", + " model, # The trained CNN model\n", " data_preparation_normed # The data preparation pipeline\n", ")\n", "\n", @@ -4340,101 +4356,85 @@ }, { "cell_type": "code", - "execution_count": 30, + "execution_count": 28, "id": "b4be6a52", "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "array([[[ 0.24508621, 0.11092616, -0.03467894, ..., 0.4291155 ,\n", - " 0.41032445, 0.34905007],\n", - " [ 0.06729264, 0.22444546, 0.47754756, ..., 0.1731117 ,\n", - " 0.1314369 , 0.07036693],\n", - " [ 0.44198117, 0.65568805, 0.72863257, ..., -0.30857596,\n", - " 0.20914252, 0.7002841 ],\n", - " ...,\n", - " [ 0.01560672, -0.1894289 , -0.48232144, ..., 1.0082207 ,\n", - " 0.60520154, 0.27669346],\n", - " [-0.3044927 , -0.37066957, -0.27879533, ..., 0.48623183,\n", - " 0.23022135, -0.04774872],\n", - " [ 0.51261675, 0.46222374, 0.43611524, ..., 0.784465 ,\n", - " 0.67638767, 0.585016 ]],\n", - "\n", - " [[ 0.1739989 , 0.14216088, 0.1003792 , ..., 0.23476335,\n", - " 0.21560968, 0.19656503],\n", - " [-0.010908 , 0.13098194, 0.16181199, ..., 0.07339582,\n", - " -0.10057873, -0.16229634],\n", - " [ 1.0853866 , 1.1272728 , 1.2927105 , ..., -0.4901738 ,\n", - " 0.47894996, 0.6378232 ],\n", + "array([[[ 1.6265482 , 1.6642686 , 1.7807953 , ..., -0.41998312,\n", + " -0.05053011, -0.62035227],\n", + " [ 1.6282165 , 1.6466169 , 1.755078 , ..., -0.86967456,\n", + " -0.06659667, -0.6387649 ],\n", + " [ 1.6345234 , 1.6240315 , 1.6426649 , ..., -1.0400431 ,\n", + " -0.24917367, -0.6859972 ],\n", " ...,\n", - " [ 0.32347783, 0.1530604 , -0.36704504, ..., 2.2871614 ,\n", - " 1.4820675 , 0.7104936 ],\n", - " [-0.18428366, -0.5963913 , -0.93383473, ..., 0.6886066 ,\n", - " 0.3982786 , 0.12022203],\n", - " [-0.26889893, -0.4153922 , -0.5274983 , ..., 0.32514077,\n", - " 0.10366473, -0.09469063]],\n", - "\n", - " [[ 0.29989573, 0.26873863, 0.23308583, ..., 0.35678092,\n", - " 0.3471158 , 0.32710022],\n", - " [ 0.7007875 , 0.84441924, 1.0783322 , ..., 0.3950204 ,\n", - " 0.5730443 , 0.6233401 ],\n", - " [ 0.56969124, 0.6025304 , 0.6456148 , ..., -0.22357185,\n", - " -0.42198908, 0.18817033],\n", + " [ 1.642726 , 1.4959918 , 1.295203 , ..., -0.7554755 ,\n", + " -0.57467616, -0.5757325 ],\n", + " [ 1.6350749 , 1.5650128 , 1.506095 , ..., -0.63705313,\n", + " -0.49757466, -0.62132066],\n", + " [ 1.6291701 , 1.644258 , 1.6520036 , ..., -0.3815208 ,\n", + " -0.30504227, -0.6222776 ]],\n", + "\n", + " [[ 0.37331304, 0.04814345, -0.97122186, ..., -0.3308119 ,\n", + " -1.1623396 , 0.06890745],\n", + " [ 0.39981708, 0.11974857, -0.916135 , ..., -0.6844403 ,\n", + " -1.2961906 , 0.08652025],\n", + " [ 0.42725852, 0.1846136 , -0.8319019 , ..., -0.95627266,\n", + " -1.4884118 , 0.085261 ],\n", " ...,\n", - " [ 0.5893009 , 0.36362782, 0.04500921, ..., 2.3144917 ,\n", - " 1.386902 , 0.9947691 ],\n", - " [-0.28036693, -0.78888327, -1.3113825 , ..., 0.99616814,\n", - " 0.70593774, 0.20506375],\n", - " [-0.4276789 , -0.54294723, -0.6514161 , ..., 0.04105376,\n", - " -0.15166822, -0.3027542 ]],\n", - "\n", - " ...,\n", - "\n", - " [[ 1.3490297 , 0.8566821 , 1.6260393 , ..., 0.6117243 ,\n", - " 0.5638359 , 1.3058991 ],\n", - " [ 1.9504461 , 0.5837265 , 0.6163757 , ..., -0.5604832 ,\n", - " -0.50448596, 1.9702082 ],\n", - " [ 0.29819992, 1.4200642 , 1.2040739 , ..., -0.32891572,\n", - " -1.4031858 , 1.4697403 ],\n", + " [ 0.24053523, 0.04329998, -0.41543078, ..., -0.07199283,\n", + " -0.6893464 , -0.4049187 ],\n", + " [ 0.28573784, 0.00251288, -0.9453188 , ..., -0.11058147,\n", + " -0.7709441 , -0.21876507],\n", + " [ 0.34219432, -0.00976607, -0.938062 , ..., -0.1951685 ,\n", + " -0.9421096 , -0.02641799]],\n", + "\n", + " [[-0.47082186, -0.44774178, -0.40512496, ..., -0.9397035 ,\n", + " 0.14880559, 1.113687 ],\n", + " [-0.43420443, -0.18625255, -0.3106985 , ..., -1.0107784 ,\n", + " -0.02244223, 0.9295246 ],\n", + " [-0.3905126 , 0.12768012, -0.00546314, ..., -1.2066957 ,\n", + " -0.10597208, 0.83207333],\n", " ...,\n", - " [ 0.45729098, -0.2274404 , 0.56907094, ..., -0.7582452 ,\n", - " -0.25473458, -0.35702682],\n", - " [-0.40157175, -0.6543036 , -0.8019015 , ..., -0.5883408 ,\n", - " 0.42199385, -0.40957326],\n", - " [-0.4185561 , -0.5121959 , -0.53309935, ..., -0.76167035,\n", - " -0.41935343, -0.4870026 ]],\n", - "\n", - " [[ 1.1997573 , 0.91190654, 1.5743611 , ..., 0.49030527,\n", - " 0.96745735, 1.338628 ],\n", - " [ 2.2542176 , 0.42433867, 0.01592403, ..., -1.1286747 ,\n", - " -0.01179639, 1.9916474 ],\n", - " [ 0.4237562 , 1.4438224 , 0.97457874, ..., 0.48495716,\n", - " -1.0389876 , 1.1999533 ],\n", + " [-0.6347259 , -0.38031957, 0.3319075 , ..., 0.6130836 ,\n", + " 0.95907784, 1.4434704 ],\n", + " [-0.58675295, -0.4606492 , -0.00183191, ..., -0.24645995,\n", + " 0.7135204 , 1.4787238 ],\n", + " [-0.5219114 , -0.6370296 , -0.06315067, ..., -0.68712115,\n", + " 0.48995844, 1.3457698 ]],\n", + "\n", + " [[ 0.45943126, 0.90787065, -0.32233453, ..., 0.52636695,\n", + " -0.9199432 , -1.2558699 ],\n", + " [ 0.49145618, 0.91514105, 0.82572055, ..., -0.3147475 ,\n", + " -0.3943931 , -1.0036206 ],\n", + " [ 0.6085306 , 0.9962605 , 1.3351266 , ..., 0.31028575,\n", + " -0.19064371, -0.88718075],\n", " ...,\n", - " [-0.11583887, -0.19046861, -0.4617741 , ..., -0.11557942,\n", - " -0.85613203, 0.17999189],\n", - " [-0.34016842, -0.47557616, -0.25548655, ..., -1.026576 ,\n", - " -1.1687881 , -0.15066133],\n", - " [ 0.2402395 , 0.32311684, 0.11746781, ..., -1.2505002 ,\n", - " -1.0781028 , -0.46903607]],\n", - "\n", - " [[ 0.9869987 , 1.1268078 , 1.5579988 , ..., 0.6903477 ,\n", - " 0.74076974, 1.5238316 ],\n", - " [ 1.8661613 , 0.6105358 , 0.14305058, ..., -1.1973902 ,\n", - " 0.00862824, 2.0179327 ],\n", - " [ 0.3577485 , 1.557207 , 0.8968994 , ..., 0.13660753,\n", - " -0.71716374, 1.4018275 ],\n", + " [ 0.4116845 , 0.7484452 , 0.22491366, ..., -1.1199399 ,\n", + " -0.6356092 , -0.08859143],\n", + " [ 0.51986814, -0.24747247, 0.14966637, ..., -0.40968487,\n", + " 0.02281968, -0.42017704],\n", + " [ 0.50627023, 0.5339893 , -0.06993005, ..., -0.6905567 ,\n", + " -0.8807853 , -1.1061283 ]],\n", + "\n", + " [[ 0.93798494, 1.0052907 , 0.90717006, ..., 0.5913818 ,\n", + " 1.0622437 , 1.2788589 ],\n", + " [ 0.94440717, 1.0617036 , 1.03835 , ..., 0.42319804,\n", + " 1.0339918 , 1.3107482 ],\n", + " [ 0.9468403 , 1.0707257 , 1.136673 , ..., 0.2485732 ,\n", + " 1.0005698 , 1.3366479 ],\n", " ...,\n", - " [-0.5810135 , -0.36615527, -0.8015445 , ..., -0.42343837,\n", - " -0.4228332 , 2.3804533 ],\n", - " [-0.5192234 , 0.70658493, -0.32903892, ..., -0.5793631 ,\n", - " -0.7078126 , -0.4471825 ],\n", - " [-0.13518474, -0.30073836, -0.21317519, ..., -0.9163058 ,\n", - " -0.874373 , -0.25766635]]], shape=(53, 32, 64), dtype=float32)" + " [ 0.8905333 , 0.81879306, 0.60550255, ..., 0.8694485 ,\n", + " 0.93078625, 1.1228474 ],\n", + " [ 0.91156334, 0.88583666, 0.7065313 , ..., 0.8718242 ,\n", + " 1.00816 , 1.181128 ],\n", + " [ 0.9283387 , 0.9496423 , 0.861162 , ..., 0.76870537,\n", + " 1.0587218 , 1.2356416 ]]], shape=(5, 64, 32), dtype=float32)" ] }, - "execution_count": 30, + "execution_count": 28, "metadata": {}, "output_type": "execute_result" } @@ -4445,7 +4445,7 @@ }, { "cell_type": "code", - "execution_count": 31, + "execution_count": 29, "id": "8f03e1ed-51a5-4f42-9d18-dfae1f6c6a1c", "metadata": { "tags": [] @@ -4454,7 +4454,7 @@ { "data": { "application/vnd.jupyter.widget-view+json": { - "model_id": "04d4dd37136741d3ad9e255f2018680b", + "model_id": "724878007368402ab8c446fc06e7ce62", "version_major": 2, "version_minor": 0 }, @@ -4469,10 +4469,54 @@ "name": "stderr", "output_type": "stream", "text": [ - "Using default `ModelCheckpoint`. Consider installing `litmodels` package to enable `LitModelCheckpoint` for automatic upload to the Lightning model registry.\n", - "GPU available: False, used: False\n", + "💡 Tip: For seamless cloud uploads and versioning, try installing [litmodels](https://pypi.org/project/litmodels/) to enable LitModelCheckpoint, which syncs automatically with the Lightning model registry.\n", + "GPU available: True (cuda), used: True\n", "TPU available: False, using: 0 TPU cores\n", - "HPU available: False, using: 0 HPUs\n" + "HPU available: False, using: 0 HPUs\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n" ] }, { @@ -4489,8 +4533,8 @@ "name": "stdout", "output_type": "stream", "text": [ - "CPU times: user 5min 53s, sys: 30.2 s, total: 6min 23s\n", - "Wall time: 6min 26s\n" + "CPU times: user 2.7 s, sys: 156 ms, total: 2.86 s\n", + "Wall time: 2.86 s\n" ] } ], @@ -4498,7 +4542,7 @@ "%%time \n", "\n", "# Define the test split using a date range with 1-hour intervals\n", - "test_split = pyearthtools.pipeline.iterators.DateRange(test_start, test_end, interval=\"1h\")\n", + "test_split = pyearthtools.pipeline.iterators.DateRange(test_start, test_end, interval=\"6h\")\n", "\n", "# Initialise lists to store true values and predictions\n", "y_true = []\n", @@ -4524,7 +4568,7 @@ }, { "cell_type": "code", - "execution_count": 32, + "execution_count": 30, "id": "b60a847e-1717-4537-a376-cd295337b652", "metadata": { "tags": [] @@ -4553,28 +4597,76 @@ " */\n", "\n", ":root {\n", - " --xr-font-color0: var(--jp-content-font-color0, rgba(0, 0, 0, 1));\n", - " --xr-font-color2: var(--jp-content-font-color2, rgba(0, 0, 0, 0.54));\n", - " --xr-font-color3: var(--jp-content-font-color3, rgba(0, 0, 0, 0.38));\n", - " --xr-border-color: var(--jp-border-color2, #e0e0e0);\n", - " --xr-disabled-color: var(--jp-layout-color3, #bdbdbd);\n", - " --xr-background-color: var(--jp-layout-color0, white);\n", - " --xr-background-color-row-even: var(--jp-layout-color1, white);\n", - " --xr-background-color-row-odd: var(--jp-layout-color2, #eeeeee);\n", + " --xr-font-color0: var(\n", + " --jp-content-font-color0,\n", + " var(--pst-color-text-base rgba(0, 0, 0, 1))\n", + " );\n", + " --xr-font-color2: var(\n", + " --jp-content-font-color2,\n", + " var(--pst-color-text-base, rgba(0, 0, 0, 0.54))\n", + " );\n", + " --xr-font-color3: var(\n", + " --jp-content-font-color3,\n", + " var(--pst-color-text-base, rgba(0, 0, 0, 0.38))\n", + " );\n", + " --xr-border-color: var(\n", + " --jp-border-color2,\n", + " hsl(from var(--pst-color-on-background, white) h s calc(l - 10))\n", + " );\n", + " --xr-disabled-color: var(\n", + " --jp-layout-color3,\n", + " hsl(from var(--pst-color-on-background, white) h s calc(l - 40))\n", + " );\n", + " --xr-background-color: var(\n", + " --jp-layout-color0,\n", + " var(--pst-color-on-background, white)\n", + " );\n", + " --xr-background-color-row-even: var(\n", + " --jp-layout-color1,\n", + " hsl(from var(--pst-color-on-background, white) h s calc(l - 5))\n", + " );\n", + " --xr-background-color-row-odd: var(\n", + " --jp-layout-color2,\n", + " hsl(from var(--pst-color-on-background, white) h s calc(l - 15))\n", + " );\n", "}\n", "\n", "html[theme=\"dark\"],\n", "html[data-theme=\"dark\"],\n", "body[data-theme=\"dark\"],\n", "body.vscode-dark {\n", - " --xr-font-color0: rgba(255, 255, 255, 1);\n", - " --xr-font-color2: rgba(255, 255, 255, 0.54);\n", - " --xr-font-color3: rgba(255, 255, 255, 0.38);\n", - " --xr-border-color: #1f1f1f;\n", - " --xr-disabled-color: #515151;\n", - " --xr-background-color: #111111;\n", - " --xr-background-color-row-even: #111111;\n", - " --xr-background-color-row-odd: #313131;\n", + " --xr-font-color0: var(\n", + " --jp-content-font-color0,\n", + " var(--pst-color-text-base, rgba(255, 255, 255, 1))\n", + " );\n", + " --xr-font-color2: var(\n", + " --jp-content-font-color2,\n", + " var(--pst-color-text-base, rgba(255, 255, 255, 0.54))\n", + " );\n", + " --xr-font-color3: var(\n", + " --jp-content-font-color3,\n", + " var(--pst-color-text-base, rgba(255, 255, 255, 0.38))\n", + " );\n", + " --xr-border-color: var(\n", + " --jp-border-color2,\n", + " hsl(from var(--pst-color-on-background, #111111) h s calc(l + 10))\n", + " );\n", + " --xr-disabled-color: var(\n", + " --jp-layout-color3,\n", + " hsl(from var(--pst-color-on-background, #111111) h s calc(l + 40))\n", + " );\n", + " --xr-background-color: var(\n", + " --jp-layout-color0,\n", + " var(--pst-color-on-background, #111111)\n", + " );\n", + " --xr-background-color-row-even: var(\n", + " --jp-layout-color1,\n", + " hsl(from var(--pst-color-on-background, #111111) h s calc(l + 5))\n", + " );\n", + " --xr-background-color-row-odd: var(\n", + " --jp-layout-color2,\n", + " hsl(from var(--pst-color-on-background, #111111) h s calc(l + 15))\n", + " );\n", "}\n", "\n", ".xr-wrap {\n", @@ -4630,6 +4722,7 @@ "\n", ".xr-section-item input + label {\n", " color: var(--xr-disabled-color);\n", + " border: 2px solid transparent !important;\n", "}\n", "\n", ".xr-section-item input:enabled + label {\n", @@ -4638,7 +4731,7 @@ "}\n", "\n", ".xr-section-item input:focus + label {\n", - " border: 2px solid var(--xr-font-color0);\n", + " border: 2px solid var(--xr-font-color0) !important;\n", "}\n", "\n", ".xr-section-item input:enabled + label:hover {\n", @@ -4770,7 +4863,9 @@ ".xr-var-item label,\n", ".xr-var-item > .xr-var-name span {\n", " background-color: var(--xr-background-color-row-even);\n", + " border-color: var(--xr-background-color-row-odd);\n", " margin-bottom: 0;\n", + " padding-top: 2px;\n", "}\n", "\n", ".xr-var-item > .xr-var-name:hover span {\n", @@ -4781,6 +4876,7 @@ ".xr-var-list > li:nth-child(odd) > label,\n", ".xr-var-list > li:nth-child(odd) > .xr-var-name span {\n", " background-color: var(--xr-background-color-row-odd);\n", + " border-color: var(--xr-background-color-row-even);\n", "}\n", "\n", ".xr-var-name {\n", @@ -4830,8 +4926,15 @@ ".xr-var-data,\n", ".xr-index-data {\n", " display: none;\n", - " background-color: var(--xr-background-color) !important;\n", - " padding-bottom: 5px !important;\n", + " border-top: 2px dotted var(--xr-background-color);\n", + " padding-bottom: 20px !important;\n", + " padding-top: 10px !important;\n", + "}\n", + "\n", + ".xr-var-attrs-in + label,\n", + ".xr-var-data-in + label,\n", + ".xr-index-data-in + label {\n", + " padding: 0 1px;\n", "}\n", "\n", ".xr-var-attrs-in:checked ~ .xr-var-attrs,\n", @@ -4844,6 +4947,12 @@ " float: right;\n", "}\n", "\n", + ".xr-var-data > pre,\n", + ".xr-index-data > pre,\n", + ".xr-var-data > table > tbody > tr {\n", + " background-color: transparent !important;\n", + "}\n", + "\n", ".xr-var-name span,\n", ".xr-var-data,\n", ".xr-index-name div,\n", @@ -4903,2213 +5012,335 @@ " stroke: currentColor;\n", " fill: currentColor;\n", "}\n", - "
<xarray.Dataset> Size: 115MB\n",
-       "Dimensions:    (time: 264, latitude: 32, longitude: 64)\n",
+       "\n",
+       ".xr-var-attrs-in:checked + label > .xr-icon-file-text2,\n",
+       ".xr-var-data-in:checked + label > .xr-icon-database,\n",
+       ".xr-index-data-in:checked + label > .xr-icon-database {\n",
+       "  color: var(--xr-font-color0);\n",
+       "  filter: drop-shadow(1px 1px 5px var(--xr-font-color2));\n",
+       "  stroke-width: 0.8px;\n",
+       "}\n",
+       "
<xarray.Dataset> Size: 2MB\n",
+       "Dimensions:                 (time: 44, longitude: 64, latitude: 32)\n",
        "Coordinates:\n",
-       "  * time       (time) datetime64[ns] 2kB 2017-01-01 ... 2017-01-11T23:00:00\n",
-       "  * latitude   (latitude) float64 256B -87.19 -81.56 -75.94 ... 81.56 87.19\n",
-       "  * longitude  (longitude) float64 512B 0.0 5.625 11.25 ... 343.1 348.8 354.4\n",
-       "Data variables: (12/53)\n",
-       "    u50        (time, latitude, longitude) float32 2MB -4.516 -4.181 ... -7.626\n",
-       "    u100       (time, latitude, longitude) float32 2MB -4.504 -3.801 ... -1.783\n",
-       "    u150       (time, latitude, longitude) float32 2MB -4.168 -3.161 ... -0.1854\n",
-       "    u200       (time, latitude, longitude) float32 2MB -4.082 -3.166 ... 3.008\n",
-       "    u250       (time, latitude, longitude) float32 2MB -3.733 -3.103 ... 2.488\n",
-       "    u300       (time, latitude, longitude) float32 2MB -3.452 -2.377 ... 3.883\n",
-       "    ...         ...\n",
-       "    vo500      (time, latitude, longitude) float32 2MB -5.498e-06 ... 4.115e-06\n",
-       "    vo600      (time, latitude, longitude) float32 2MB -4.675e-06 ... -2.075e-05\n",
-       "    vo700      (time, latitude, longitude) float32 2MB 1.607e-05 ... -1.521e-05\n",
-       "    vo850      (time, latitude, longitude) float32 2MB -5.099e-06 ... -7.604e-06\n",
-       "    vo925      (time, latitude, longitude) float32 2MB -1.709e-06 ... -1.359e-05\n",
-       "    vo1000     (time, latitude, longitude) float32 2MB -2.693e-06 ... -1.537e-05\n",
+       "  * time                    (time) datetime64[ns] 352B 2017-01-01 ... 2017-01...\n",
+       "  * longitude               (longitude) float64 512B 0.0 5.625 ... 348.8 354.4\n",
+       "  * latitude                (latitude) float64 256B -87.19 -81.56 ... 87.19\n",
+       "Data variables:\n",
+       "    2m_temperature          (time, longitude, latitude) float32 360kB 236.5 ....\n",
+       "    u_component_of_wind850  (time, longitude, latitude) float32 360kB -4.186 ...\n",
+       "    v_component_of_wind850  (time, longitude, latitude) float32 360kB -4.593 ...\n",
+       "    vorticity850            (time, longitude, latitude) float32 360kB -4.668e...\n",
+       "    geopotential850         (time, longitude, latitude) float32 360kB 1.223e+...\n",
        "Attributes:\n",
-       "    level-dtype:  int32
    • time
      PandasIndex
      PandasIndex(DatetimeIndex(['2017-01-01 00:00:00', '2017-01-01 06:00:00',\n",
+       "               '2017-01-01 12:00:00', '2017-01-01 18:00:00',\n",
+       "               '2017-01-02 00:00:00', '2017-01-02 06:00:00',\n",
+       "               '2017-01-02 12:00:00', '2017-01-02 18:00:00',\n",
+       "               '2017-01-03 00:00:00', '2017-01-03 06:00:00',\n",
+       "               '2017-01-03 12:00:00', '2017-01-03 18:00:00',\n",
+       "               '2017-01-04 00:00:00', '2017-01-04 06:00:00',\n",
+       "               '2017-01-04 12:00:00', '2017-01-04 18:00:00',\n",
+       "               '2017-01-05 00:00:00', '2017-01-05 06:00:00',\n",
+       "               '2017-01-05 12:00:00', '2017-01-05 18:00:00',\n",
+       "               '2017-01-06 00:00:00', '2017-01-06 06:00:00',\n",
+       "               '2017-01-06 12:00:00', '2017-01-06 18:00:00',\n",
+       "               '2017-01-07 00:00:00', '2017-01-07 06:00:00',\n",
+       "               '2017-01-07 12:00:00', '2017-01-07 18:00:00',\n",
+       "               '2017-01-08 00:00:00', '2017-01-08 06:00:00',\n",
+       "               '2017-01-08 12:00:00', '2017-01-08 18:00:00',\n",
+       "               '2017-01-09 00:00:00', '2017-01-09 06:00:00',\n",
+       "               '2017-01-09 12:00:00', '2017-01-09 18:00:00',\n",
+       "               '2017-01-10 00:00:00', '2017-01-10 06:00:00',\n",
+       "               '2017-01-10 12:00:00', '2017-01-10 18:00:00',\n",
+       "               '2017-01-11 00:00:00', '2017-01-11 06:00:00',\n",
+       "               '2017-01-11 12:00:00', '2017-01-11 18:00:00'],\n",
+       "              dtype='datetime64[ns]', name='time', freq=None))
    • longitude
      PandasIndex
      PandasIndex(Index([               0.0,              5.625,              11.25,\n",
+       "                   16.875,               22.5,             28.125,\n",
+       "                    33.75,             39.375,               45.0,\n",
+       "                   50.625,              56.25,  61.87499999999999,\n",
+       "                     67.5,             73.125,              78.75,\n",
+       "                   84.375,               90.0,             95.625,\n",
+       "                   101.25,            106.875,              112.5,\n",
+       "                  118.125, 123.74999999999999,            129.375,\n",
+       "                    135.0,            140.625,             146.25,\n",
+       "                  151.875,              157.5,            163.125,\n",
+       "                   168.75,            174.375,              180.0,\n",
+       "                  185.625,             191.25,            196.875,\n",
+       "                    202.5,            208.125,             213.75,\n",
+       "                  219.375,              225.0, 230.62499999999997,\n",
+       "                   236.25,            241.875, 247.49999999999997,\n",
+       "                  253.125,             258.75,            264.375,\n",
+       "                    270.0,            275.625,             281.25,\n",
+       "                  286.875,              292.5,            298.125,\n",
+       "                   303.75,            309.375,              315.0,\n",
+       "                  320.625,             326.25,            331.875,\n",
+       "                    337.5,            343.125,             348.75,\n",
+       "                  354.375],\n",
+       "      dtype='float64', name='longitude'))
    • latitude
      PandasIndex
      PandasIndex(Index([ -87.18750000000003,  -81.56250000000001,            -75.9375,\n",
+       "        -70.31249999999999,  -64.68750000000001,            -59.0625,\n",
+       "                  -53.4375,            -47.8125,            -42.1875,\n",
+       "                  -36.5625, -30.937499999999996, -25.312500000000004,\n",
+       "       -19.687499999999996, -14.062499999999991,  -8.437499999999996,\n",
+       "        -2.812500000000003,   2.812500000000003,   8.437500000000009,\n",
+       "        14.062500000000004,  19.687499999999996,  25.312500000000004,\n",
+       "         30.93750000000001,  36.562499999999986,             42.1875,\n",
+       "                   47.8125,             53.4375,  59.062500000000014,\n",
+       "         64.68750000000001,             70.3125,             75.9375,\n",
+       "         81.56249999999997,   87.18750000000003],\n",
+       "      dtype='float64', name='latitude'))
  • level-dtype :
    int64
    • " ], "text/plain": [ - " Size: 115MB\n", - "Dimensions: (time: 264, latitude: 32, longitude: 64)\n", + " Size: 2MB\n", + "Dimensions: (time: 44, longitude: 64, latitude: 32)\n", "Coordinates:\n", - " * time (time) datetime64[ns] 2kB 2017-01-01 ... 2017-01-11T23:00:00\n", - " * latitude (latitude) float64 256B -87.19 -81.56 -75.94 ... 81.56 87.19\n", - " * longitude (longitude) float64 512B 0.0 5.625 11.25 ... 343.1 348.8 354.4\n", - "Data variables: (12/53)\n", - " u50 (time, latitude, longitude) float32 2MB -4.516 -4.181 ... -7.626\n", - " u100 (time, latitude, longitude) float32 2MB -4.504 -3.801 ... -1.783\n", - " u150 (time, latitude, longitude) float32 2MB -4.168 -3.161 ... -0.1854\n", - " u200 (time, latitude, longitude) float32 2MB -4.082 -3.166 ... 3.008\n", - " u250 (time, latitude, longitude) float32 2MB -3.733 -3.103 ... 2.488\n", - " u300 (time, latitude, longitude) float32 2MB -3.452 -2.377 ... 3.883\n", - " ... ...\n", - " vo500 (time, latitude, longitude) float32 2MB -5.498e-06 ... 4.115e-06\n", - " vo600 (time, latitude, longitude) float32 2MB -4.675e-06 ... -2.075e-05\n", - " vo700 (time, latitude, longitude) float32 2MB 1.607e-05 ... -1.521e-05\n", - " vo850 (time, latitude, longitude) float32 2MB -5.099e-06 ... -7.604e-06\n", - " vo925 (time, latitude, longitude) float32 2MB -1.709e-06 ... -1.359e-05\n", - " vo1000 (time, latitude, longitude) float32 2MB -2.693e-06 ... -1.537e-05\n", + " * time (time) datetime64[ns] 352B 2017-01-01 ... 2017-01...\n", + " * longitude (longitude) float64 512B 0.0 5.625 ... 348.8 354.4\n", + " * latitude (latitude) float64 256B -87.19 -81.56 ... 87.19\n", + "Data variables:\n", + " 2m_temperature (time, longitude, latitude) float32 360kB 236.5 ....\n", + " u_component_of_wind850 (time, longitude, latitude) float32 360kB -4.186 ...\n", + " v_component_of_wind850 (time, longitude, latitude) float32 360kB -4.593 ...\n", + " vorticity850 (time, longitude, latitude) float32 360kB -4.668e...\n", + " geopotential850 (time, longitude, latitude) float32 360kB 1.223e+...\n", "Attributes:\n", - " level-dtype: int32" + " level-dtype: int64" ] }, - "execution_count": 32, + "execution_count": 30, "metadata": {}, "output_type": "execute_result" } @@ -7120,7 +5351,7 @@ }, { "cell_type": "code", - "execution_count": 33, + "execution_count": 31, "id": "3fd145ba-4761-4d41-aa14-7dbf8042bb9f", "metadata": { "tags": [] @@ -7132,13 +5363,13 @@ "Text(0.5, 1.05, 'Predictions')" ] }, - "execution_count": 33, + "execution_count": 31, "metadata": {}, "output_type": "execute_result" }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
    " ] @@ -7148,13 +5379,13 @@ } ], "source": [ - "grid = y_preds.isel(time=slice(3))[\"z850\"].plot(col=\"time\")\n", + "grid = y_preds.isel(time=slice(3))[\"geopotential850\"].plot(x=\"longitude\", y=\"latitude\", col=\"time\")\n", "grid.fig.suptitle(\"Predictions\", y=1.05)" ] }, { "cell_type": "code", - "execution_count": 34, + "execution_count": 32, "id": "ee675ea9-62a0-4c46-acbb-a99f71f9113e", "metadata": { "tags": [] @@ -7166,13 +5397,13 @@ "Text(0.5, 1.05, 'Ground truth')" ] }, - "execution_count": 34, + "execution_count": 32, "metadata": {}, "output_type": "execute_result" }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
    " ] @@ -7182,7 +5413,7 @@ } ], "source": [ - "grid = y_true.isel(time=slice(3))[\"z850\"].plot(col=\"time\")\n", + "grid = y_true.isel(time=slice(3))[\"geopotential850\"].plot(x=\"longitude\", y=\"latitude\", col=\"time\")\n", "grid.fig.suptitle(\"Ground truth\", y=1.05)" ] }, @@ -7196,7 +5427,7 @@ }, { "cell_type": "code", - "execution_count": 35, + "execution_count": 33, "id": "8c923d43-c76b-478c-a6c2-280aa7cb50df", "metadata": { "tags": [] @@ -7210,7 +5441,7 @@ }, { "cell_type": "code", - "execution_count": 36, + "execution_count": 34, "id": "0697141f-d144-47e5-922a-5f292b6a5bd8", "metadata": { "tags": [] @@ -7239,28 +5470,76 @@ " */\n", "\n", ":root {\n", - " --xr-font-color0: var(--jp-content-font-color0, rgba(0, 0, 0, 1));\n", - " --xr-font-color2: var(--jp-content-font-color2, rgba(0, 0, 0, 0.54));\n", - " --xr-font-color3: var(--jp-content-font-color3, rgba(0, 0, 0, 0.38));\n", - " --xr-border-color: var(--jp-border-color2, #e0e0e0);\n", - " --xr-disabled-color: var(--jp-layout-color3, #bdbdbd);\n", - " --xr-background-color: var(--jp-layout-color0, white);\n", - " --xr-background-color-row-even: var(--jp-layout-color1, white);\n", - " --xr-background-color-row-odd: var(--jp-layout-color2, #eeeeee);\n", + " --xr-font-color0: var(\n", + " --jp-content-font-color0,\n", + " var(--pst-color-text-base rgba(0, 0, 0, 1))\n", + " );\n", + " --xr-font-color2: var(\n", + " --jp-content-font-color2,\n", + " var(--pst-color-text-base, rgba(0, 0, 0, 0.54))\n", + " );\n", + " --xr-font-color3: var(\n", + " --jp-content-font-color3,\n", + " var(--pst-color-text-base, rgba(0, 0, 0, 0.38))\n", + " );\n", + " --xr-border-color: var(\n", + " --jp-border-color2,\n", + " hsl(from var(--pst-color-on-background, white) h s calc(l - 10))\n", + " );\n", + " --xr-disabled-color: var(\n", + " --jp-layout-color3,\n", + " hsl(from var(--pst-color-on-background, white) h s calc(l - 40))\n", + " );\n", + " --xr-background-color: var(\n", + " --jp-layout-color0,\n", + " var(--pst-color-on-background, white)\n", + " );\n", + " --xr-background-color-row-even: var(\n", + " --jp-layout-color1,\n", + " hsl(from var(--pst-color-on-background, white) h s calc(l - 5))\n", + " );\n", + " --xr-background-color-row-odd: var(\n", + " --jp-layout-color2,\n", + " hsl(from var(--pst-color-on-background, white) h s calc(l - 15))\n", + " );\n", "}\n", "\n", "html[theme=\"dark\"],\n", "html[data-theme=\"dark\"],\n", "body[data-theme=\"dark\"],\n", "body.vscode-dark {\n", - " --xr-font-color0: rgba(255, 255, 255, 1);\n", - " --xr-font-color2: rgba(255, 255, 255, 0.54);\n", - " --xr-font-color3: rgba(255, 255, 255, 0.38);\n", - " --xr-border-color: #1f1f1f;\n", - " --xr-disabled-color: #515151;\n", - " --xr-background-color: #111111;\n", - " --xr-background-color-row-even: #111111;\n", - " --xr-background-color-row-odd: #313131;\n", + " --xr-font-color0: var(\n", + " --jp-content-font-color0,\n", + " var(--pst-color-text-base, rgba(255, 255, 255, 1))\n", + " );\n", + " --xr-font-color2: var(\n", + " --jp-content-font-color2,\n", + " var(--pst-color-text-base, rgba(255, 255, 255, 0.54))\n", + " );\n", + " --xr-font-color3: var(\n", + " --jp-content-font-color3,\n", + " var(--pst-color-text-base, rgba(255, 255, 255, 0.38))\n", + " );\n", + " --xr-border-color: var(\n", + " --jp-border-color2,\n", + " hsl(from var(--pst-color-on-background, #111111) h s calc(l + 10))\n", + " );\n", + " --xr-disabled-color: var(\n", + " --jp-layout-color3,\n", + " hsl(from var(--pst-color-on-background, #111111) h s calc(l + 40))\n", + " );\n", + " --xr-background-color: var(\n", + " --jp-layout-color0,\n", + " var(--pst-color-on-background, #111111)\n", + " );\n", + " --xr-background-color-row-even: var(\n", + " --jp-layout-color1,\n", + " hsl(from var(--pst-color-on-background, #111111) h s calc(l + 5))\n", + " );\n", + " --xr-background-color-row-odd: var(\n", + " --jp-layout-color2,\n", + " hsl(from var(--pst-color-on-background, #111111) h s calc(l + 15))\n", + " );\n", "}\n", "\n", ".xr-wrap {\n", @@ -7316,6 +5595,7 @@ "\n", ".xr-section-item input + label {\n", " color: var(--xr-disabled-color);\n", + " border: 2px solid transparent !important;\n", "}\n", "\n", ".xr-section-item input:enabled + label {\n", @@ -7324,7 +5604,7 @@ "}\n", "\n", ".xr-section-item input:focus + label {\n", - " border: 2px solid var(--xr-font-color0);\n", + " border: 2px solid var(--xr-font-color0) !important;\n", "}\n", "\n", ".xr-section-item input:enabled + label:hover {\n", @@ -7456,7 +5736,9 @@ ".xr-var-item label,\n", ".xr-var-item > .xr-var-name span {\n", " background-color: var(--xr-background-color-row-even);\n", + " border-color: var(--xr-background-color-row-odd);\n", " margin-bottom: 0;\n", + " padding-top: 2px;\n", "}\n", "\n", ".xr-var-item > .xr-var-name:hover span {\n", @@ -7467,6 +5749,7 @@ ".xr-var-list > li:nth-child(odd) > label,\n", ".xr-var-list > li:nth-child(odd) > .xr-var-name span {\n", " background-color: var(--xr-background-color-row-odd);\n", + " border-color: var(--xr-background-color-row-even);\n", "}\n", "\n", ".xr-var-name {\n", @@ -7516,8 +5799,15 @@ ".xr-var-data,\n", ".xr-index-data {\n", " display: none;\n", - " background-color: var(--xr-background-color) !important;\n", - " padding-bottom: 5px !important;\n", + " border-top: 2px dotted var(--xr-background-color);\n", + " padding-bottom: 20px !important;\n", + " padding-top: 10px !important;\n", + "}\n", + "\n", + ".xr-var-attrs-in + label,\n", + ".xr-var-data-in + label,\n", + ".xr-index-data-in + label {\n", + " padding: 0 1px;\n", "}\n", "\n", ".xr-var-attrs-in:checked ~ .xr-var-attrs,\n", @@ -7530,6 +5820,12 @@ " float: right;\n", "}\n", "\n", + ".xr-var-data > pre,\n", + ".xr-index-data > pre,\n", + ".xr-var-data > table > tbody > tr {\n", + " background-color: transparent !important;\n", + "}\n", + "\n", ".xr-var-name span,\n", ".xr-var-data,\n", ".xr-index-name div,\n", @@ -7589,715 +5885,149 @@ " stroke: currentColor;\n", " fill: currentColor;\n", "}\n", - "
    <xarray.Dataset> Size: 435kB\n",
-       "Dimensions:    (latitude: 32, longitude: 64)\n",
+       "\n",
+       ".xr-var-attrs-in:checked + label > .xr-icon-file-text2,\n",
+       ".xr-var-data-in:checked + label > .xr-icon-database,\n",
+       ".xr-index-data-in:checked + label > .xr-icon-database {\n",
+       "  color: var(--xr-font-color0);\n",
+       "  filter: drop-shadow(1px 1px 5px var(--xr-font-color2));\n",
+       "  stroke-width: 0.8px;\n",
+       "}\n",
+       "
    <xarray.Dataset> Size: 42kB\n",
+       "Dimensions:                 (longitude: 64, latitude: 32)\n",
        "Coordinates:\n",
-       "  * latitude   (latitude) float64 256B -87.19 -81.56 -75.94 ... 81.56 87.19\n",
-       "  * longitude  (longitude) float64 512B 0.0 5.625 11.25 ... 343.1 348.8 354.4\n",
-       "Data variables: (12/53)\n",
-       "    u50        (latitude, longitude) float32 8kB 1.133 1.178 ... 11.24 13.31\n",
-       "    u100       (latitude, longitude) float32 8kB 1.086 1.166 ... 5.481 5.843\n",
-       "    u150       (latitude, longitude) float32 8kB 1.702 1.628 ... 3.452 3.287\n",
-       "    u200       (latitude, longitude) float32 8kB 1.687 1.597 ... 1.659 1.818\n",
-       "    u250       (latitude, longitude) float32 8kB 1.986 1.623 ... 2.243 3.913\n",
-       "    u300       (latitude, longitude) float32 8kB 3.648 3.353 ... 3.681 5.765\n",
-       "    ...         ...\n",
-       "    vo500      (latitude, longitude) float32 8kB 1.949e-05 ... 2.765e-05\n",
-       "    vo600      (latitude, longitude) float32 8kB 1.826e-05 ... 3.03e-05\n",
-       "    vo700      (latitude, longitude) float32 8kB 2.584e-05 ... 3.02e-05\n",
-       "    vo850      (latitude, longitude) float32 8kB 9.394e-06 ... 2.363e-05\n",
-       "    vo925      (latitude, longitude) float32 8kB 9.061e-06 ... 2.945e-05\n",
-       "    vo1000     (latitude, longitude) float32 8kB 9.061e-06 ... 1.884e-05\n",
+       "  * longitude               (longitude) float64 512B 0.0 5.625 ... 348.8 354.4\n",
+       "  * latitude                (latitude) float64 256B -87.19 -81.56 ... 87.19\n",
+       "Data variables:\n",
+       "    2m_temperature          (longitude, latitude) float32 8kB 11.67 ... 5.455\n",
+       "    u_component_of_wind850  (longitude, latitude) float32 8kB 0.8308 ... 3.061\n",
+       "    v_component_of_wind850  (longitude, latitude) float32 8kB 0.5541 ... 5.419\n",
+       "    vorticity850            (longitude, latitude) float32 8kB 4.153e-06 ... 1...\n",
+       "    geopotential850         (longitude, latitude) float32 8kB 525.4 ... 472.6\n",
        "Attributes:\n",
-       "    level-dtype:  int32
    • longitude
      PandasIndex
      PandasIndex(Index([               0.0,              5.625,              11.25,\n",
+       "                   16.875,               22.5,             28.125,\n",
+       "                    33.75,             39.375,               45.0,\n",
+       "                   50.625,              56.25,  61.87499999999999,\n",
+       "                     67.5,             73.125,              78.75,\n",
+       "                   84.375,               90.0,             95.625,\n",
+       "                   101.25,            106.875,              112.5,\n",
+       "                  118.125, 123.74999999999999,            129.375,\n",
+       "                    135.0,            140.625,             146.25,\n",
+       "                  151.875,              157.5,            163.125,\n",
+       "                   168.75,            174.375,              180.0,\n",
+       "                  185.625,             191.25,            196.875,\n",
+       "                    202.5,            208.125,             213.75,\n",
+       "                  219.375,              225.0, 230.62499999999997,\n",
+       "                   236.25,            241.875, 247.49999999999997,\n",
+       "                  253.125,             258.75,            264.375,\n",
+       "                    270.0,            275.625,             281.25,\n",
+       "                  286.875,              292.5,            298.125,\n",
+       "                   303.75,            309.375,              315.0,\n",
+       "                  320.625,             326.25,            331.875,\n",
+       "                    337.5,            343.125,             348.75,\n",
+       "                  354.375],\n",
+       "      dtype='float64', name='longitude'))
    • latitude
      PandasIndex
      PandasIndex(Index([ -87.18750000000003,  -81.56250000000001,            -75.9375,\n",
+       "        -70.31249999999999,  -64.68750000000001,            -59.0625,\n",
+       "                  -53.4375,            -47.8125,            -42.1875,\n",
+       "                  -36.5625, -30.937499999999996, -25.312500000000004,\n",
+       "       -19.687499999999996, -14.062499999999991,  -8.437499999999996,\n",
+       "        -2.812500000000003,   2.812500000000003,   8.437500000000009,\n",
+       "        14.062500000000004,  19.687499999999996,  25.312500000000004,\n",
+       "         30.93750000000001,  36.562499999999986,             42.1875,\n",
+       "                   47.8125,             53.4375,  59.062500000000014,\n",
+       "         64.68750000000001,             70.3125,             75.9375,\n",
+       "         81.56249999999997,   87.18750000000003],\n",
+       "      dtype='float64', name='latitude'))
  • level-dtype :
    int64
    • " ], "text/plain": [ - " Size: 435kB\n", - "Dimensions: (latitude: 32, longitude: 64)\n", + " Size: 42kB\n", + "Dimensions: (longitude: 64, latitude: 32)\n", "Coordinates:\n", - " * latitude (latitude) float64 256B -87.19 -81.56 -75.94 ... 81.56 87.19\n", - " * longitude (longitude) float64 512B 0.0 5.625 11.25 ... 343.1 348.8 354.4\n", - "Data variables: (12/53)\n", - " u50 (latitude, longitude) float32 8kB 1.133 1.178 ... 11.24 13.31\n", - " u100 (latitude, longitude) float32 8kB 1.086 1.166 ... 5.481 5.843\n", - " u150 (latitude, longitude) float32 8kB 1.702 1.628 ... 3.452 3.287\n", - " u200 (latitude, longitude) float32 8kB 1.687 1.597 ... 1.659 1.818\n", - " u250 (latitude, longitude) float32 8kB 1.986 1.623 ... 2.243 3.913\n", - " u300 (latitude, longitude) float32 8kB 3.648 3.353 ... 3.681 5.765\n", - " ... ...\n", - " vo500 (latitude, longitude) float32 8kB 1.949e-05 ... 2.765e-05\n", - " vo600 (latitude, longitude) float32 8kB 1.826e-05 ... 3.03e-05\n", - " vo700 (latitude, longitude) float32 8kB 2.584e-05 ... 3.02e-05\n", - " vo850 (latitude, longitude) float32 8kB 9.394e-06 ... 2.363e-05\n", - " vo925 (latitude, longitude) float32 8kB 9.061e-06 ... 2.945e-05\n", - " vo1000 (latitude, longitude) float32 8kB 9.061e-06 ... 1.884e-05\n", + " * longitude (longitude) float64 512B 0.0 5.625 ... 348.8 354.4\n", + " * latitude (latitude) float64 256B -87.19 -81.56 ... 87.19\n", + "Data variables:\n", + " 2m_temperature (longitude, latitude) float32 8kB 11.67 ... 5.455\n", + " u_component_of_wind850 (longitude, latitude) float32 8kB 0.8308 ... 3.061\n", + " v_component_of_wind850 (longitude, latitude) float32 8kB 0.5541 ... 5.419\n", + " vorticity850 (longitude, latitude) float32 8kB 4.153e-06 ... 1...\n", + " geopotential850 (longitude, latitude) float32 8kB 525.4 ... 472.6\n", "Attributes:\n", - " level-dtype: int32" + " level-dtype: int64" ] }, - "execution_count": 36, + "execution_count": 34, "metadata": {}, "output_type": "execute_result" } @@ -8308,7 +6038,7 @@ }, { "cell_type": "code", - "execution_count": 37, + "execution_count": 35, "id": "5e43f0ff-97f2-4a86-9bc9-f419dce1c588", "metadata": { "tags": [] @@ -8317,18 +6047,18 @@ { "data": { "text/plain": [ - "" + "" ] }, - "execution_count": 37, + "execution_count": 35, "metadata": {}, "output_type": "execute_result" }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ - "
    " + "
    " ] }, "metadata": {}, @@ -8336,13 +6066,13 @@ } ], "source": [ - "mae_score[[\"z500\", \"z850\", \"z1000\"]].to_array(dim=\"field\").plot(col=\"field\")" + "mae_score[\"geopotential850\"].plot(x=\"longitude\", y=\"latitude\")" ] }, { "cell_type": "code", "execution_count": null, - "id": "95287070-b739-49c6-9982-08d4dd332a25", + "id": "cbd99564-4260-4271-a26e-6bc29a4fe9d5", "metadata": {}, "outputs": [], "source": [] @@ -8364,7 +6094,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.13.2" + "version": "3.13.5" }, "nbsphinx": { "orphan": true diff --git a/notebooks/tutorial/cnn_training/stats/mean.npy b/notebooks/tutorial/cnn_training/stats/mean.npy deleted file mode 100644 index 7e66889a..00000000 Binary files a/notebooks/tutorial/cnn_training/stats/mean.npy and /dev/null differ diff --git a/notebooks/tutorial/cnn_training/stats/std.npy b/notebooks/tutorial/cnn_training/stats/std.npy deleted file mode 100644 index 387960d7..00000000 Binary files a/notebooks/tutorial/cnn_training/stats/std.npy and /dev/null differ diff --git a/packages/data/src/pyearthtools/data/download/weatherbench.py b/packages/data/src/pyearthtools/data/download/weatherbench.py index 894246ca..05ebeb1d 100644 --- a/packages/data/src/pyearthtools/data/download/weatherbench.py +++ b/packages/data/src/pyearthtools/data/download/weatherbench.py @@ -4,7 +4,6 @@ import hashlib from pathlib import Path from typing import Literal -from abc import ABC, abstractmethod import fsspec import xarray as xr @@ -160,7 +159,7 @@ def open_local_dataset(path: Path, variables: list[str], level: list[int]) -> xr return dset_full -class WeatherBench2(ABC, AdvancedTimeDataIndex): +class WeatherBench2(AdvancedTimeDataIndex): """WeatherBench2 cloud-optimized ground truth and baseline datasets https://github.com/google-research/weatherbench2 @@ -179,7 +178,8 @@ class WeatherBench2(ABC, AdvancedTimeDataIndex): @decorators.variable_modifications("variables") def __init__( self, - url: str, + dataset_url: str, + license_url: str, *, variables: str | list[str] | None = None, level: int | list[int] | None = None, @@ -201,6 +201,10 @@ def __init__( downloaded. Args: + dataset_url (str): + URL of the zarr dataset + license_url (str): + License of the dataset variables (str | list[str] | None, optional): Variables to retrieve, can be either short_name or long_name. Default to None, to retrieve all variables. @@ -216,12 +220,11 @@ def __init__( License has been read. Defaults to False. """ super().__init__(transforms or TransformCollection(), data_interval="1 hour") - self.record_initialisation() # retrieve variables name mapping and levels for the dataset from pyearthtools.data.download._weatherbench import DATASETS_INFOS - long_names, valid_levels = DATASETS_INFOS[url] + long_names, valid_levels = DATASETS_INFOS[dataset_url] # create short variables name mappings short_names = {val: key for key, val in long_names.items() if val is not None} @@ -250,18 +253,18 @@ def __init__( def open_online_dataset(): # skip parsing unused variables, this can make loading much faster drop_variables = [var for var in long_names if var not in set(variables)] - ds = xr.open_zarr(url, chunks=chunks, drop_variables=drop_variables, **kwargs) + ds = xr.open_zarr(dataset_url, chunks=chunks, drop_variables=drop_variables, **kwargs) if level is not None: ds = Select(level=level, ignore_missing=True)(ds) return ds if download_dir is None: ds = open_online_dataset() - license = self.license_url + license = license_url else: # use a hash of the url to identify the dataset subfolder - url_hash = hashlib.sha256(url.encode()).hexdigest() + url_hash = hashlib.sha256(dataset_url.encode()).hexdigest() download_path = Path(download_dir) / url_hash # try to open dataset from download dir if defined @@ -271,11 +274,11 @@ def open_online_dataset(): except MissingVariableFile: ds_remote = open_online_dataset() save_local_dataset(download_path, ds_remote) - (download_path / "dataset_url").write_text(url) + (download_path / "dataset_url").write_text(dataset_url) ds = open_local_dataset(download_path, variables, level) if not (license := download_path / "LICENSE").is_file(): - with fsspec.open(self.license_url, "rt").open() as fd: + with fsspec.open(license_url, "rt").open() as fd: license_txt = fd.read() license.write_text(license_txt) @@ -291,11 +294,6 @@ def open_online_dataset(): self._license = license self._kwargs = kwargs - @property - @abstractmethod - def license_url(self): - pass - @property def _desc_(self) -> dict[str, str]: return { @@ -347,7 +345,7 @@ class WB2ERA5(WeatherBench2): } @decorators.check_arguments(resolution=["raw", "1440x721", "240x121", "64x32"]) - def __init__(self, resolution: str = "64x32", **kwargs): + def __init__(self, *, resolution: str = "64x32", **kwargs): """ See :class:`pyearthtools.data.download.weatherbench.WeatherBench2` for additional parameters. @@ -358,13 +356,11 @@ def __init__(self, resolution: str = "64x32", **kwargs): The "raw" dataset is not subsampled, i.e. is hourly with 36 levels. Defaults to "64x32". """ - url = f"gs://weatherbench2/datasets/era5/{self.DATASETS[resolution]}" - super().__init__(url, **kwargs) + dataset_url = f"gs://weatherbench2/datasets/era5/{self.DATASETS[resolution]}" + license_url = "gs://weatherbench2/datasets/era5/LICENSE" + super().__init__(dataset_url, license_url, **kwargs) self.resolution = resolution - - @property - def license_url(self): - return "gs://weatherbench2/datasets/era5/LICENSE" + self.record_initialisation() @classmethod def sample(cls): @@ -405,7 +401,7 @@ class WB2ERA5Clim(WeatherBench2): @decorators.check_arguments( resolution=["1440x721", "512x256", "240x121", "64x32"], period=["1990-2017", "1990-2019"] ) - def __init__(self, resolution: str = "64x32", period: str = "1990-2017", **kwargs): + def __init__(self, *, resolution: str = "64x32", period: str = "1990-2017", **kwargs): """ See :class:`pyearthtools.data.download.weatherbench.WeatherBench2` for additional parameters. @@ -418,14 +414,12 @@ def __init__(self, resolution: str = "64x32", period: str = "1990-2017", **kwarg Covered time period, either "1990-2017" or "1990-2019". Defaults to "1990-2017". """ - url = f"gs://weatherbench2/datasets/era5-hourly-climatology/{self.DATASETS[(period, resolution)]}" - super().__init__(url, **kwargs) + dataset_url = f"gs://weatherbench2/datasets/era5-hourly-climatology/{self.DATASETS[(period, resolution)]}" + license_url = "gs://weatherbench2/datasets/era5-hourly-climatology/LICENSE" + super().__init__(dataset_url, license_url, **kwargs) self.period = period self.resolution = resolution - - @property - def license_url(self): - return "gs://weatherbench2/datasets/era5-hourly-climatology/LICENSE" + self.record_initialisation() @classmethod def sample(cls):