merged developed, fixed conflicts

Signed-off-by: Duncan Ragsdale <[email protected]>

NewTaskRequirements.md

@@ -0,0 +1,117 @@
+# Required files for creating a new PV Validation Hub Task
+
+## config.json
+
+Example JSON:
+
+```json
+{
+    "category_name": "Time Shift Analysis",
+    "function_name": "detect_time_shifts",
+    "comparison_type": "time_series",
+    "performance_metrics": [
+        "runtime",
+        "mean_absolute_error"
+    ],
+    "allowable_kwargs": [
+        "latitude",
+        "longitude",
+        "data_sampling_frequency"
+    ],
+    "ground_truth_compare": [
+        "time_series"
+    ],
+    "public_results_table": "time-shift-public-metrics.json",
+    "private_results_columns": [
+        "system_id",
+        "file_name",
+        "run_time",
+        "data_requirements",
+        "mean_absolute_error_time_series",
+        "data_sampling_frequency",
+        "issue"
+    ],
+    "plots": [
+        {
+            "type": "histogram",
+            "x_val": "mean_absolute_error_time_series",
+            "color_code": "issue",
+            "title": "Time Series MAE Distribution by Issue",
+            "save_file_path": "mean_absolute_error_time_series_dist.png"
+        },
+        {
+            "type": "histogram",
+            "x_val": "mean_absolute_error_time_series",
+            "color_code": "data_sampling_frequency",
+            "title": "Time Series MAE Distribution by Sampling Frequency",
+            "save_file_path": "mean_absolute_error_time_series_dist.png"
+        },
+        {
+            "type": "histogram",
+            "x_val": "run_time",
+            "title": "Run Time Distribution",
+            "save_file_path": "run_time_dist.png"
+        }
+    ]
+}
+```
+
+## system_metadata.csv
+
+Required columns:
+
+```csv
+system_id,name,azimuth,tilt,elevation,latitude,longitude,tracking,climate_type,dc_capacity
+```
+
+## file_metadata.csv
+
+Required columns:
+
+```csv
+file_id,system_id,file_name,timezone,data_sampling_frequency,issue
+```
+
+## template.py (Marimo template with cli args input)
+
+Marimo python file will need to input data from `mo.cli_args()` method
+
+Example:
+
+```python
+def create_df_from_cli_args():
+        args = mo.cli_args().to_dict()
+        data = args.get("results_df")
+        rows = []
+        for row in data:
+            rows.append(json.loads(row))
+
+        df = pd.DataFrame.from_records(rows)
+        return df
+```
+
+## csv data files
+
+file names must match what is included in the file_name in the file_metadata.csv
+
+## ground truth csv data files
+
+file names must match what is included in the data files folder
+
+## Markdown files for Task
+
+### description.md
+
+The markdown file used for the description tab in an analysis.
+
+### dataset.md
+
+The markdown file to describe the dataset in the data tab.
+
+### shortdesc.md
+
+The markdown file that is used on the card.
+
+### SubmissionInstructions.md
+
+The markdown file that is used on the Submission Instructions tab in the analysis.

compressions/1/sdt-submission/submission_wrapper.py

@@ -5,7 +5,6 @@
 def detect_time_shifts(
     time_series, latitude=None, longitude=None, data_sampling_frequency=None
 ):
-    raise NotImplementedError("This function is not implemented yet.")
     dh = DataHandler(time_series.to_frame())
     dh.run_pipeline(fix_shifts=True, verbose=False, round_shifts_to_hour=False)
     return dh.time_shift_analysis.correction_estimate

dockerize-workflow/.dockerignore

@@ -0,0 +1,34 @@
+# Ignore files and directories generated by the operating system
+.DS_Store
+Thumbs.db
+
+# Ignore build artifacts
+node_modules
+dist
+build
+
+# Ignore development and debugging files
+npm-debug.log*
+yarn-debug.log*
+yarn-error.log*
+.vscode
+
+# Ignore editor-specific files
+*.swp
+*.swo
+*.bak
+
+# Ignore project-specific files
+.env
+.env.local
+.env.*.local
+.env.development
+.env.test
+.env.production
+
+# Ignore any other files or directories as needed
+__pycache__
+.venv
+.dockerignore
+Dockerfile
+docker-compose.yml

dockerize-workflow/.gitignore

@@ -0,0 +1,3 @@
+results/*
+data/*
+certs/*

dockerize-workflow/Dockerfile

@@ -0,0 +1,20 @@
+# Use an official Python runtime as the base image
+FROM python:3.12-alpine
+
+# Set the working directory to /app
+
+WORKDIR /app
+
+# Copy the current directory contents into the container at /app
+COPY . .
+
+RUN apk add --no-cache docker-cli
+
+RUN apk add gcc python3-dev musl-dev linux-headers
+
+RUN python3 -m venv .venv && \
+    .venv/bin/pip install --upgrade pip && \
+    .venv/bin/pip install --no-cache-dir -r requirements.txt
+
+# run container without closing
+CMD ["tail", "-f", "/dev/null"]

dockerize-workflow/docker-compose.yaml

@@ -0,0 +1,16 @@
+services:
+  dockerize-workflow:
+    build:
+      context: ./
+    env_file:
+      - .env
+    ports:
+      - "8787:8787"
+    volumes:
+      - /var/run/docker.sock:/var/run/docker.sock
+      - ./certs:/app/certs
+      - ./main.py:/app/main.py
+      - ./requirements.txt:/app/requirements.txt
+      - ./results:/app/results
+      - ./data:/app/data
+      - ./environment:/app/environment

dockerize-workflow/environment/Dockerfile

@@ -0,0 +1,33 @@
+# Use an official Python runtime as the base image
+FROM python:3.11-slim
+
+# Set the working directory in the container
+WORKDIR /app
+
+RUN apt-get update 
+
+COPY unzip.py .
+COPY requirements.txt .
+
+# Install the Python dependencies for the submission wrapper
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Copy the submission package into the container
+COPY $zip_file .
+
+# Unzip the submission package
+
+RUN python -m unzip $zip_file submission
+
+WORKDIR /app/submission
+
+# Install the Python dependencies
+RUN pip install --no-cache-dir -r requirements.txt
+
+
+# Set the working directory in the container
+WORKDIR /app
+
+COPY submission_wrapper.py .
+# Command to keep the container running without doing anything
+# CMD tail -f /dev/null

dockerize-workflow/environment/requirements.txt

@@ -0,0 +1,2 @@
+numpy
+solar-data-tools==1.3.2

dockerize-workflow/environment/submission_wrapper.py

@@ -0,0 +1,152 @@
+from importlib import import_module
+import inspect
+import sys
+import pandas as pd
+import numpy as np
+from time import perf_counter
+from functools import wraps
+from typing import Any, ParamSpec, Union, Tuple, TypeVar, Callable, cast
+from logging import Logger
+import logging
+
+T = TypeVar("T")
+
+P = ParamSpec("P")
+
+
+def logger_if_able(
+    message: str, logger: Logger | None = None, level: str = "INFO"
+):
+    if logger is not None:
+        levels_dict = {
+            "DEBUG": logging.DEBUG,
+            "INFO": logging.INFO,
+            "WARNING": logging.WARNING,
+            "ERROR": logging.ERROR,
+            "CRITICAL": logging.CRITICAL,
+        }
+
+        level = level.upper()
+
+        if level not in levels_dict:
+            raise Exception(f"Invalid log level: {level}")
+
+        log_level = levels_dict[level]
+
+        logger.log(log_level, message)
+    else:
+        print(message)
+
+
+def timing(verbose: bool = True, logger: Union[Logger, None] = None):
+    def decorator(func: Callable[P, T]):
+        def wrapper(*args: P.args, **kwargs: P.kwargs) -> Tuple[T, float]:
+            start_time = perf_counter()
+            result = func(*args, **kwargs)
+            end_time = perf_counter()
+            execution_time = end_time - start_time
+            if verbose:
+                msg = (
+                    f"{func.__name__} took {execution_time:.3f} seconds to run"
+                )
+                logger_if_able(msg, logger)
+            return result, execution_time
+
+        return wrapper
+
+    return decorator
+
+
+def format_args_for_submission(data_dir: str, args: list[str]):
+    filename = args[0]
+
+    file_path = f"{data_dir}/{filename}"
+
+    df = pd.read_csv(
+        file_path,
+        index_col=0,
+        parse_dates=True,
+    )
+
+    print(df.head(5))
+
+    series: pd.Series = df.asfreq("60min").squeeze()
+
+    submission_args = [series, *args[1:]]
+
+    return submission_args
+
+
+def import_submission_function(submission_file_name: str, function_name: str):
+    # Dynamically import function from submission.py
+    try:
+        submission_module = import_module(submission_file_name)
+    except ModuleNotFoundError as e:
+        print(f"ModuleNotFoundError: {submission_file_name} not found")
+        raise e
+
+    try:
+        submission_function: Callable[[pd.Series, Any], np.ndarray] = getattr(
+            submission_module, function_name
+        )
+        function_parameters = list(
+            inspect.signature(submission_function).parameters.keys()
+        )
+    except AttributeError as e:
+        print(
+            f"AttributeError: {function_name} not found in submission module"
+        )
+        raise e
+
+    return submission_function, function_parameters
+
+
+def main():
+    args = sys.argv[1:]
+
+    if len(args) < 1:
+        print("Function name not provided")
+        sys.exit(1)
+
+    submission_file_name = args[0]
+    function_name = args[1]
+    data_file_name = args[2]
+
+    print("Getting submission function...")
+
+    submission_function, function_parameters = import_submission_function(
+        submission_file_name, function_name
+    )
+    print("Got submission function")
+
+    print(f"Submission file name: {submission_file_name}")
+    print(f"Function name: {function_name}")
+    print(f"Function: {submission_function}")
+    print(f"Function parameters: {function_parameters}")
+
+    data_dir = "/app/data/"
+    results_dir = "/app/results/"
+
+    submission_args = format_args_for_submission(data_dir, args[2:])
+
+    print(f"Submission args: {submission_args}")
+
+    results, execution_time = timing()(submission_function)(*submission_args)
+
+    print(f"Execution time: {execution_time}")
+
+    print(f"Results: {results}")
+
+    # save results to csv file
+    results_df = pd.DataFrame(results)
+    results_file = f"{results_dir}/{data_file_name}"
+    results_df.to_csv(results_file)
+
+    execution_tuple = (data_file_name, execution_time)
+    execution_file = f"{results_dir}/time.csv"
+    execution_df = pd.DataFrame([execution_tuple])
+    execution_df.to_csv(execution_file, mode="a", header=False)
+
+
+if __name__ == "__main__":
+    main()