import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
import re

# Load and parse raw data from a text file
def parse_raw_data(file_path):
    pattern = re.compile(r"\((.*?)\)\s+([\d.]+)\s+\(([\d.eE+,-]+)\)")
    data = []
    with open(file_path, "r") as f:
        for line in f:
            match = pattern.search(line)
            if match:
                activation_combo = match.group(1)
                learning_rate = float(match.group(2))
                top_accuracies = tuple(map(float, match.group(3).split(",")))
                data.append((activation_combo, learning_rate, top_accuracies))
    return data

# Replace this with your actual path
file_path = "./parsed.txt"
data = parse_raw_data(file_path)

# Convert to DataFrame
df = pd.DataFrame(data, columns=["activation_combo", "learning_rate", "accuracy"])
df[["top1", "top3", "top5"]] = pd.DataFrame(df["accuracy"].tolist(), index=df.index)

# Unique sorted learning rates and activation combos
learning_rates = sorted(df["learning_rate"].unique())
activation_combos = df["activation_combo"].unique()

# Settings for bar positions
bar_width = 0.2
x = np.arange(len(learning_rates))

# Start plotting
fig, (ax1, ax2, ax3) = plt.subplots(3, 1, figsize=(14, 16), sharex=True)

# Function to draw bars
def plot_bars(ax, column, title, ylabel):
    for i, combo in enumerate(activation_combos):
        combo_data = df[df["activation_combo"] == combo].sort_values("learning_rate")
        ax.bar(x + i * bar_width, combo_data[column], width=bar_width, label=combo)
    ax.set_ylabel(ylabel)
    ax.set_title(title)
    ax.legend(title="Activation Combo")

# Plot each accuracy type
plot_bars(ax1, "top1", "Top-1 Accuracy", "Top-1 Accuracy")
plot_bars(ax2, "top3", "Top-3 Accuracy", "Top-3 Accuracy")
plot_bars(ax3, "top5", "Top-5 Accuracy", "Top-5 Accuracy")

# X-axis ticks for learning rates
ax3.set_xticks(x + bar_width * (len(activation_combos) - 1) / 2)
ax3.set_xticklabels([str(lr) for lr in learning_rates])
ax3.set_xlabel("Learning Rate")

# Final layout
plt.tight_layout()
plt.show()