From d8a9897e8e84cff372d8cbd176c0dcec0e6d6c0b Mon Sep 17 00:00:00 2001
From: Rosheen Naeem <rosheennaeem4@gmail.com>
Date: Fri, 8 Mar 2024 17:59:29 +0100
Subject: [PATCH] using ocf_datapipes in Open-Source-Quartz-Solar-Forecast
 (#76)

* accessing ocf_datapipes in quartz project

* refactoring of code

* moved datapipes to scripts
---
 .gitignore                               |  5 ++-
 scripts/datapipes/__init__.py            |  0
 scripts/datapipes/ocf_datapipe.py        | 56 ++++++++++++++++++++++++
 scripts/datapipes/pv_and_nwp_config.yaml | 23 ++++++++++
 4 files changed, 83 insertions(+), 1 deletion(-)
 create mode 100644 scripts/datapipes/__init__.py
 create mode 100644 scripts/datapipes/ocf_datapipe.py
 create mode 100644 scripts/datapipes/pv_and_nwp_config.yaml

diff --git a/.gitignore b/.gitignore
index 2eea525d..b4ab6c30 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1 +1,4 @@
-.env
\ No newline at end of file
+.env
+scripts/datapipes/test.nc
+scripts/datapipes/UK_PV_metadata.csv
+scripts/datapipes/nwp_data
\ No newline at end of file
diff --git a/scripts/datapipes/__init__.py b/scripts/datapipes/__init__.py
new file mode 100644
index 00000000..e69de29b
diff --git a/scripts/datapipes/ocf_datapipe.py b/scripts/datapipes/ocf_datapipe.py
new file mode 100644
index 00000000..b501eb7d
--- /dev/null
+++ b/scripts/datapipes/ocf_datapipe.py
@@ -0,0 +1,56 @@
+import ocf_datapipes  # noqa
+from ocf_datapipes.training.example.simple_pv import simple_pv_datapipe
+from ocf_datapipes.training.example.pv_nwp import pv_nwp_datapipe
+from ocf_datapipes.batch import BatchKey
+from datetime import datetime
+import os
+import certifi
+import ssl
+import warnings
+
+# Suppress warnings and configure SSL certificate for secure connections
+warnings.filterwarnings("ignore")
+os.environ['SSL_CERT_FILE'] = certifi.where()
+ssl._create_default_https_context = ssl._create_unverified_context
+
+def process_ocf_datapipes(config_file):
+
+    """
+    This function demonstrates the use of Open Climate Fix's datapipes with the Open Source Quartz Solar Forecast Project.
+    It is an exploratory integration and not currently utilized in the Quartz Solar Forecast production pipeline.
+
+    It processes solar power (PV) and numerical weather prediction (NWP) data to prepare it for forecasting tasks.
+
+    :param config_file: The config file that specifies the paths to the data files, data preprocessing parameters, and other configurations necessary for the datapipes to function correctly.
+    -  'pv_and_nwp_config.yaml' specifes the paths to necessary data files (e.g., PV output data, NWP data).
+    -  The YAML file needs to be edited to reflect the correct paths to your data files and any specific preprocessing requirements for your project.
+
+    This script is meant as a starting point for integrating Open Climate Fix datapipes into the Quartz Solar Forecast Project, serving as an example of how to preprocess and load data for solar power forecasting.
+    """
+
+    pv_data_pipe = simple_pv_datapipe(configuration_filename=config_file)
+    pv_nwp_data_pipe = pv_nwp_datapipe(configuration_filename=config_file)
+
+    ## Pv datapipe: store the first batch of processed data for inspection
+    pv_batch = next(iter(pv_data_pipe))
+
+    # Access the batch elements through batch keys: Convert the first set of timestamps to readable dates
+    pv_times_readable = [datetime.utcfromtimestamp(ts).strftime('%Y-%m-%d %H:%M:%S') for ts in pv_batch[BatchKey.pv_time_utc][0]]
+    print("Readable PV Times:", pv_times_readable)
+
+    # Check Observed Capacity (Wp), and Nominal Capacity (Wp) of the PV system
+    observed_capacity = pv_batch[BatchKey.pv_observed_capacity_wp]
+    nominal_capacity = pv_batch[BatchKey.pv_nominal_capacity_wp]
+
+    print("Observed Capacity (Wp):", observed_capacity)
+    print("Nominal Capacity (Wp):", nominal_capacity)
+
+    # NWP_PV DataPipe: Retrieve and print NWP data from the batch
+    pv_nwp_batch = next(iter(pv_nwp_data_pipe))
+    nwp_data = pv_nwp_batch[BatchKey.nwp]
+    print("NWP data", nwp_data)
+
+
+# Load configuration from YAML specifying data sources and processing parameters for datapipes
+config_file = 'pv_and_nwp_config.yaml'
+process_ocf_datapipes(config_file)
\ No newline at end of file
diff --git a/scripts/datapipes/pv_and_nwp_config.yaml b/scripts/datapipes/pv_and_nwp_config.yaml
new file mode 100644
index 00000000..8e4fcb25
--- /dev/null
+++ b/scripts/datapipes/pv_and_nwp_config.yaml
@@ -0,0 +1,23 @@
+input_data:
+  pv:
+    pv_files_groups:
+      - label: solar_sheffield_passiv
+        pv_filename: 'test.nc' # Available at https://github.com/openclimatefix/ocf_datapipes/blob/main/tests/data/pv/passiv/test.nc
+        pv_metadata_filename: 'UK_PV_metadata.csv' # Available at https://github.com/openclimatefix/ocf_datapipes/blob/main/tests/data/pv/passiv/UK_PV_metadata.csv
+        get_center: false
+        pv_image_size_meters_height: 10000000
+        pv_image_size_meters_width: 10000000
+  nwp:
+    ukv:
+      nwp_channels:
+        - t
+      nwp_image_size_pixels_height: 2
+      nwp_image_size_pixels_width: 2
+      nwp_zarr_path: nwp_data/test.zarr # Available at https://github.com/openclimatefix/ocf_datapipes/tree/main/tests/data/nwp_data/test.zarr
+      nwp_provider: "ukv"
+      history_minutes: 60
+      forecast_minutes: 120
+      time_resolution_minutes: 60
+      index_by_id: True
+      dropout_timedeltas_minutes: [-180]
+      dropout_fraction: 1.0
\ No newline at end of file