Release

.github/workflows/test.yml

@@ -23,14 +23,18 @@ jobs:
     - name: Set up Python
       uses: actions/setup-python@v4
       with:
-        python-version: '3.12.5'  # Specify Python version (e.g., '3.9')
+        python-version: '3.11.9'  # Specify Python version (e.g., '3.9')
 
-    # Install dependencies (you can specify requirements.txt or pyproject.toml)
+    # Install Poetry
+    - name: Install Poetry
+      run: |
+          curl -sSL https://install.python-poetry.org | python3 -
+          export PATH="$HOME/.local/bin:$PATH"
+
+      # Install dependencies using Poetry
     - name: Install dependencies
       run: |
-        python -m pip install --upgrade pip
-        pip install pytest pandas numpy entsoe-py redis tensorflow scikit-learn sphinx
-        pip install .
+          poetry install
 
     # Run pytest to execute tests
     - name: Generate .config file inside the test folder
@@ -40,4 +44,4 @@ jobs:
           echo "enable_energy_caching=false" >> .codegreencore.config    
     - name: Run tests
       run: |
-        pytest
+        poetry run pytest

.gitignore

@@ -171,6 +171,6 @@ tests/data1
 
 # temp, will remove later
 codegreen_core/models/files
-codegreen_core/utilities/log.py
-
-.vscode
+Dockerfile
+.vscode
+poetry.lock

codegreen_core/__init__.py

@@ -1,2 +1,3 @@
 from .utilities.config import Config
+
 Config.load_config()

codegreen_core/data/__init__.py

@@ -1,2 +1,3 @@
 from .main import *
-__all__ = ['energy']
+
+__all__ = ["energy"]

codegreen_core/data/main.py

@@ -1,12 +1,13 @@
 import pandas as pd
 from datetime import datetime
 
-from ..utilities.message import Message,CodegreenDataError
-from ..utilities  import metadata as meta  
+from ..utilities.message import Message, CodegreenDataError
+from ..utilities import metadata as meta
 from . import entsoe as et
 
-def energy(country,start_time,end_time,type="generation",interval60=True)-> pd.DataFrame:
-  """
+
+def energy(country, start_time, end_time, type="generation", interval60=True) -> dict:
+    """
     Returns hourly time series of energy production mix for a specified country and time range.
 
     This method fetches the energy data for the specified country between the specified duration.
@@ -15,56 +16,65 @@ def energy(country,start_time,end_time,type="generation",interval60=True)-> pd.D
 
     For example, if the source is ENTSOE, the data contains:
 
-     ========================== ========== ================================================================ 
-      Column                     type       Description                                                     
-     ========================== ========== ================================================================ 
-      startTimeUTC               datetime   Start date in UTC (60 min interval)                             
-      Biomass                    float64                                                                    
-      Fossil Hard coal           float64                                                                    
-      Geothermal                 float64                                                                    
-      ....more energy sources    float64                                                                    
-      **renewableTotal**         float64    The total based on all renewable sources                        
-      renewableTotalWS           float64    The total production using only Wind and Solar energy sources   
-      nonRenewableTotal          float64                                                                    
-      total                      float64    Total using all energy sources                                  
-      percentRenewable           int64                                                                      
-      percentRenewableWS         int64      Percentage of energy produced using only wind and solar energy  
-      Wind_per                   int64      Percentages of individual energy sources                        
-      Solar_per                  int64                                                                      
-      Nuclear_per                int64                                                                      
-      Hydroelectricity_per       int64                                                                      
-      Geothermal_per             int64                                                                      
-      Natural Gas_per            int64                                                                      
-      Petroleum_per              int64                                                                      
-      Coal_per                   int64                                                                      
-      Biomass_per                int64                                                                      
-     ========================== ========== ================================================================ 
+     ========================== ========== ================================================================
+      Column                     type       Description
+     ========================== ========== ================================================================
+      startTimeUTC               datetime   Start date in UTC (60 min interval)
+      Biomass                    float64
+      Fossil Hard coal           float64
+      Geothermal                 float64
+      ....more energy sources    float64
+      **renewableTotal**         float64    The total based on all renewable sources
+      renewableTotalWS           float64    The total production using only Wind and Solar energy sources
+      nonRenewableTotal          float64
+      total                      float64    Total using all energy sources
+      percentRenewable           int64
+      percentRenewableWS         int64      Percentage of energy produced using only wind and solar energy
+      Wind_per                   int64      Percentages of individual energy sources
+      Solar_per                  int64
+      Nuclear_per                int64
+      Hydroelectricity_per       int64
+      Geothermal_per             int64
+      Natural Gas_per            int64
+      Petroleum_per              int64
+      Coal_per                   int64
+      Biomass_per                int64
+     ========================== ========== ================================================================
 
     Note : fields marked bold are calculated based on the data fetched.
 
     :param str country: The 2 alphabet country code.
     :param datetime start_time: The start date for data retrieval. A Datetime object. Note that this date will be rounded to the nearest hour.
     :param datetime end_time: The end date for data retrieval. A datetime object. This date is also rounded to the nearest hour.
-    :param str type: The type of data to retrieve; either 'historical' or 'forecasted'. Defaults to 'historical'.
-    :return: A DataFrame containing the hourly energy production mix.
-    :rtype: pd.DataFrame
+    :param str type: The type of data to retrieve; either 'generation' or 'forecast'. Defaults to 'generation'.
+    :return: A dictionary containing:
+      - `error`: A string with an error message, empty if no errors.
+      - `data_available`: A boolean indicating if data was successfully retrieved.
+      - `data`: A pandas DataFrame containing the energy data if available, empty DataFrame if not.
+      - `time_interval` : the time interval of the DataFrame
+    :rtype: dict
     """
-  if not isinstance(country, str):
-    raise ValueError("Invalid country")
-  if not isinstance(start_time,datetime):
-    raise ValueError("Invalid start date")
-  if not isinstance(end_time, datetime):
-    raise ValueError("Invalid end date")
-  if type not in ['generation', 'forecast']:
-    raise ValueError(Message.INVALID_ENERGY_TYPE)
-  # check start<end and both are not same 
-
-  e_source = meta.get_country_energy_source(country)
-  if e_source=="ENTSOE" :
-    if type == "generation":
-      return et.get_actual_production_percentage(country,start_time,end_time,interval60)
-    elif type == "forecast":
-      return et.get_forecast_percent_renewable(country,start_time,end_time)
-  else:
-    raise CodegreenDataError(Message.NO_ENERGY_SOURCE)
-  return None
+    if not isinstance(country, str):
+        raise ValueError("Invalid country")
+    if not isinstance(start_time, datetime):
+        raise ValueError("Invalid start date")
+    if not isinstance(end_time, datetime):
+        raise ValueError("Invalid end date")
+    if type not in ["generation", "forecast"]:
+        raise ValueError(Message.INVALID_ENERGY_TYPE)
+    # check start<end and both are not same
+
+    if start_time > end_time:
+        raise ValueError("Invalid time.End time should be greater than start time")
+
+    e_source = meta.get_country_energy_source(country)
+    if e_source == "ENTSOE":
+        if type == "generation":
+            return et.get_actual_production_percentage(
+                country, start_time, end_time, interval60
+            )
+        elif type == "forecast":
+            return et.get_forecast_percent_renewable(country, start_time, end_time)
+    else:
+        raise CodegreenDataError(Message.NO_ENERGY_SOURCE)
+    return None

codegreen_core/models/predict.py

@@ -11,50 +11,60 @@
 
 # Path to the models directory
 models_dir = Path(__file__).parent / "files"
-
+
+
+def predicted_energy(country):
+    # do the forecast from now , same return format as data.energy
+    return {"data": None}
+
 
 # Function to load a specific model by name
-def load_prediction_model(country,version=None):
+def _load_prediction_model(country, version=None):
     """Load a model by name"""
-    model_details = get_prediction_model_details(country,version)
+    model_details = get_prediction_model_details(country, version)
     model_path = models_dir / model_details["name"]
     print(model_path)
     if not model_path.exists():
         raise FileNotFoundError(f"Model does not exist.")
-
-    return load_model(model_path,compile=False)
 
+    return load_model(model_path, compile=False)
 
-def run(country,input,model_version=None):
+
+def _run(country, input, model_version=None):
     """Returns the prediction values"""
-    
+
     seq_length = len(input)
-    date = input[['startTimeUTC']].copy()
+    date = input[["startTimeUTC"]].copy()
     # Convert 'startTimeUTC' column to datetime
-    date['startTimeUTC'] = pd.to_datetime(date['startTimeUTC'])
+    date["startTimeUTC"] = pd.to_datetime(date["startTimeUTC"])
     # Get the last date value
-    last_date = date.iloc[-1]['startTimeUTC']
+    last_date = date.iloc[-1]["startTimeUTC"]
     # Calculate the next hour
     next_hour = last_date + timedelta(hours=1)
     # Create a range of 48 hours starting from the next hour
-    next_48_hours = pd.date_range(next_hour, periods=48, freq='h')
+    next_48_hours = pd.date_range(next_hour, periods=48, freq="h")
     # Create a DataFrame with the next 48 hours
     next_48_hours_df = pd.DataFrame(
-        {'startTimeUTC': next_48_hours.strftime('%Y%m%d%H%M')})
-
-    model_details = get_prediction_model_details(country,model_version)
-
-    lstm = load_prediction_model(country,model_version) #load_model(model_path,compile=False)
+        {"startTimeUTC": next_48_hours.strftime("%Y%m%d%H%M")}
+    )
+
+    model_details = get_prediction_model_details(country, model_version)
+
+    lstm = load_prediction_model(
+        country, model_version
+    )  # load_model(model_path,compile=False)
 
     scaler = StandardScaler()
-    percent_renewable = input['percentRenewable']
+    percent_renewable = input["percentRenewable"]
     forecast_values_total = []
     prev_values_total = percent_renewable.values.flatten()
     for _ in range(48):
         scaled_prev_values_total = scaler.fit_transform(
-            prev_values_total.reshape(-1, 1))
-        x_pred_total = scaled_prev_values_total[-(
-            seq_length-1):].reshape(1, (seq_length-1), 1)
+            prev_values_total.reshape(-1, 1)
+        )
+        x_pred_total = scaled_prev_values_total[-(seq_length - 1) :].reshape(
+            1, (seq_length - 1), 1
+        )
         # Make the prediction using the loaded model
         predicted_value_total = lstm.predict(x_pred_total, verbose=0)
         # Inverse transform the predicted value
@@ -64,24 +74,29 @@ def run(country,input,model_version=None):
         prev_values_total = prev_values_total[1:]
     # Create a DataFrame
     forecast_df = pd.DataFrame(
-        {'startTimeUTC': next_48_hours_df['startTimeUTC'], 'percentRenewableForecast': forecast_values_total})
-    forecast_df["percentRenewableForecast"] = forecast_df["percentRenewableForecast"].round(
-    ).astype(int)
-    forecast_df['percentRenewableForecast'] = forecast_df['percentRenewableForecast'].apply(
-        lambda x: 0 if x <= 0 else x)
-
+        {
+            "startTimeUTC": next_48_hours_df["startTimeUTC"],
+            "percentRenewableForecast": forecast_values_total,
+        }
+    )
+    forecast_df["percentRenewableForecast"] = (
+        forecast_df["percentRenewableForecast"].round().astype(int)
+    )
+    forecast_df["percentRenewableForecast"] = forecast_df[
+        "percentRenewableForecast"
+    ].apply(lambda x: 0 if x <= 0 else x)
+
     input_percentage = input["percentRenewable"].tolist()
     input_start = input.iloc[0]["startTimeUTC"]
-    input_end = input.iloc[-1]["startTimeUTC"]    
-    
+    input_end = input.iloc[-1]["startTimeUTC"]
+
     return {
         "input": {
             "country": country,
             "model": model_details["name"],
             "percentRenewable": input_percentage,
             "start": input_start,
-            "end": input_end
+            "end": input_end,
         },
-        "output": forecast_df
+        "output": forecast_df,
     }
-

codegreen_core/models/train.py

@@ -1 +1 @@
-# the code for model training comes here # todo later
+# the code for model training comes here # todo later