From fddfec33375275e296a136d03a037e41be2b9585 Mon Sep 17 00:00:00 2001
From: Sermet Pekin <sermet.pekin@gmail.com>
Date: Fri, 6 Dec 2024 09:34:52 +0300
Subject: [PATCH] cross entropy

---
 .gitignore                    |   1 +
 micrograd/cross_entropy.py    |  42 ++++++++
 micrograd/engine.py           |  11 ++
 pyproject.toml                |   5 +-
 test/test_crossEntropyLoss.py |  33 ++++++
 torch_e2.py                   | 197 ++++++++++++++++++++++++++++++++++
 torchexample.py               |  65 ++++++++---
 uv.lock                       |  71 +++++++++++-
 8 files changed, 405 insertions(+), 20 deletions(-)
 create mode 100644 micrograd/cross_entropy.py
 create mode 100644 test/test_crossEntropyLoss.py
 create mode 100644 torch_e2.py

diff --git a/.gitignore b/.gitignore
index 561ba984..8624e774 100644
--- a/.gitignore
+++ b/.gitignore
@@ -13,3 +13,4 @@ ignore*/
 data/
 FashionMNIST/
 raw/
+*.pth
diff --git a/micrograd/cross_entropy.py b/micrograd/cross_entropy.py
new file mode 100644
index 00000000..f70a0416
--- /dev/null
+++ b/micrograd/cross_entropy.py
@@ -0,0 +1,42 @@
+from typing import List
+
+from micrograd.engine import Value
+import math
+
+class CrossEntropyLoss:
+    @staticmethod
+    def forward(logits: List[Value], target: int) -> Value:
+        """
+        Computes CrossEntropyLoss for a single example.
+
+        :param logits: List of Value objects, raw outputs (logits) from the model.
+        :param target: Integer index of the true class.
+        :return: Loss Value.
+        """
+        # Step 1: Compute the exponentials of the logits
+        exp_logits = [logit.exp() for logit in logits]
+
+        # Step 2: Compute the sum of the exponentials
+        sum_exp_logits = sum(exp_logits)
+
+        # Step 3: Compute the softmax probabilities
+        probs = [exp_logit / sum_exp_logits for exp_logit in exp_logits]
+
+        # Step 4: Compute the negative log-likelihood loss for the target class
+        loss = -probs[target].log()
+
+        return loss
+
+    @staticmethod
+    def batch_forward(batch_logits: List[List[Value]], batch_targets: List[int]) -> Value:
+        """
+        Computes the average CrossEntropyLoss for a batch.
+
+        :param batch_logits: List of List[Value] for all samples in the batch.
+        :param batch_targets: List of true class indices for the batch.
+        :return: Average loss Value.
+        """
+        batch_loss = sum(
+            CrossEntropyLoss.forward(logits, target) for logits, target in zip(batch_logits, batch_targets)
+        )
+        return batch_loss / len(batch_targets)
diff --git a/micrograd/engine.py b/micrograd/engine.py
index 0b9d304a..b0bfd35e 100644
--- a/micrograd/engine.py
+++ b/micrograd/engine.py
@@ -130,6 +130,17 @@ def _backward():
 
         return out
 
+    def log(self) -> 'Value':
+        """Logarithm is only defined for positive values."""
+        clamped_data = max(self.data, 1e-7)
+        out = Value(math.log(clamped_data), (self,), 'log')
+
+        def _backward():
+            self.grad += (1 / clamped_data) * out.grad
+
+        out._backward = _backward
+        return out
+
     def backward(self) -> None:
 
         # topological order all the children in the graph
diff --git a/pyproject.toml b/pyproject.toml
index c0a96089..998b914b 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -2,7 +2,7 @@
 name = "micrograd"
 version = "0.2.0"
 authors = [
-    { name = "Andrej Karpathy", email = "andrej.karpathy@gmail.com" },
+
     { name = "Sermet Pekin", email = "sermet.pekin@gmail.com" },
 ]
 description = "A tiny scalar-valued autograd engine with a small PyTorch-like neural network library on top."
@@ -13,6 +13,7 @@ dependencies = [
     "black>=24.10.0",
     "graphviz>=0.20.3",
     "matplotlib>=3.7.5",
+    "pandas>=2.2.3",
     "pytest>=8.3.4",
     "ruff>=0.8.1",
     "scikit-learn>=1.3.2",
@@ -23,7 +24,7 @@ dependencies = [
 
 [project.urls]
 Source = "https://github.com/SermetPekin/micrograd"
-Original = "https://github.com/karpathy/micrograd"
+Inspired-by = "https://github.com/karpathy/micrograd"
 
 
 [build-system]
diff --git a/test/test_crossEntropyLoss.py b/test/test_crossEntropyLoss.py
new file mode 100644
index 00000000..6f0b0db5
--- /dev/null
+++ b/test/test_crossEntropyLoss.py
@@ -0,0 +1,33 @@
+import pytest
+import torch
+import torch.nn as nn
+from micrograd.engine import Value
+from micrograd.cross_entropy import CrossEntropyLoss  # Assuming this is your custom implementation
+
+def test_micrograd_vs_torch_cross_entropy():
+    # Define the logits and targets
+    logits_micrograd = [[Value(-1.0), Value(-2.0), Value(-3.0)],
+                        [Value(0.5), Value(-1.5), Value(-0.5)]]
+    targets_micrograd = [0, 2]
+
+    # Torch equivalent tensors
+    logits_torch = torch.tensor([[-1.0, -2.0, -3.0],
+                                 [0.5, -1.5, -0.5]], dtype=torch.float32)
+    targets_torch = torch.tensor([0, 2], dtype=torch.long)
+
+    # Compute micrograd loss
+    loss_micrograd = CrossEntropyLoss.batch_forward(logits_micrograd, targets_micrograd)
+    micrograd_loss_value = loss_micrograd.data
+
+    # Compute torch loss
+    criterion = nn.CrossEntropyLoss()
+    loss_torch = criterion(logits_torch, targets_torch)
+    torch_loss_value = loss_torch.item()
+
+    # Print losses for debugging
+    print(f"Micrograd Loss: {micrograd_loss_value:.4f}")
+    print(f"Torch Loss: {torch_loss_value:.4f}")
+
+    # Assert that the losses are approximately equal
+    assert abs(micrograd_loss_value - torch_loss_value) < 1e-4, \
+        f"Losses do not match: Micrograd={micrograd_loss_value}, Torch={torch_loss_value}"
diff --git a/torch_e2.py b/torch_e2.py
new file mode 100644
index 00000000..909cbbe1
--- /dev/null
+++ b/torch_e2.py
@@ -0,0 +1,197 @@
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+from sklearn.model_selection import train_test_split
+import pandas as pd
+import numpy as np
+
+import matplotlib.pyplot as plt
+from torch.utils.data import DataLoader
+from torchvision import datasets
+from torchvision.transforms import ToTensor
+
+
+def init_data():
+    def fnc(d: str):
+        dict_ = {
+            'Setosa': 0,
+            'Versicolor': 1,
+            'Virginica': 2,
+
+        }
+        return dict_.get(d, d)
+
+    url = "https://gist.githubusercontent.com/netj/8836201/raw/6f9306ad21398ea43cba4f7d537619d0e07d5ae3/iris.csv"
+    df = pd.read_csv(url)
+    df['variety'] = df['variety'].apply(fnc)
+    return df
+
+
+def process_data_for_torch(df):
+    X = df.drop('variety', axis=1)
+    y = df['variety']
+    X = X.values
+    y = y.values
+
+    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)
+    X_train = torch.FloatTensor(X_train)
+    X_test = torch.FloatTensor(X_test)
+    y_train = torch.LongTensor(y_train)
+    y_test = torch.LongTensor(y_test)
+    return X_train, X_test, y_train, y_test
+
+
+class Model(nn.Module):
+    def __init__(self, in_feats: int = 4, out_feats: int = 3, hidden1=7, hidden2=7):
+        super(Model, self).__init__()
+        self.fc1 = nn.Linear(in_feats, hidden1)
+        self.fc2 = nn.Linear(hidden1, hidden2)
+        self.out = nn.Linear(hidden2, out_feats)
+
+    def forward(self, x):
+        x = F.relu(self.fc1(x))
+        x = F.relu(self.fc2(x))
+        x = self.out(x)
+        return x
+
+
+def with_torch(save=True):
+    df = init_data()
+    X_train, X_test, y_train, y_test = process_data_for_torch(df)
+
+    model = Model()
+
+    criterion = nn.CrossEntropyLoss()
+    optimizer = torch.optim.Adam(model.parameters(), lr=0.01)
+
+    # model.parameters
+    epochs = 60
+    losses = []
+    for i in range(epochs):
+        y_pred = model.forward(X_train)
+        loss = criterion(y_pred, y_train)
+
+        losses.append(loss.detach().numpy())
+
+        if i % 10 == 0:
+            print(f'Epoch : {i} and loss : {loss}')
+
+        # backprop
+
+        optimizer.zero_grad()
+        loss.backward()
+        optimizer.step()
+
+    if save:
+        # Save the model
+        torch.save(model.state_dict(), "iris_model.pth")
+
+    with torch.no_grad():
+        y_eval = model.forward(X_test)
+        loss = criterion(y_eval, y_test).item()
+        accuracy = calculate_accuracy(y_eval, y_test)
+
+        print(f"Test Loss: {loss}")
+        print(f"Test Accuracy: {accuracy * 100:.2f}%")
+        return model, losses
+
+
+def load_model():
+    # Load the model
+    loaded_model = Model()
+    loaded_model.load_state_dict(torch.load("iris_model.pth"))
+    loaded_model.eval()
+    return loaded_model
+
+
+def calculate_accuracy(_y_pred, _y_true):
+    y_pred_classes = torch.argmax(_y_pred, axis=1)  # Get the predicted class
+    acc = (y_pred_classes == _y_true).sum().item() / len(_y_true)
+    return acc
+
+
+# with_torch()
+
+def with_micrograd():
+    from micrograd import Value
+    from micrograd import MLP
+
+    # Define the Micrograd model
+    in_feats = 4  # Input features (Iris dataset)
+    hidden1 = 7  # Hidden layer 1
+    hidden2 = 7  # Hidden layer 2
+    out_feats = 3  # Output classes
+
+    model = MLP(in_feats, [hidden1, hidden2, out_feats])  # Equivalent to your PyTorch model
+    df = init_data()
+    X_train, X_test, y_train, y_test = process_data_for_torch(df)
+    # Hyperparameters
+    learning_rate = 0.01
+    epochs = 100
+    losses = []
+
+    # # # One-hot encode y_train
+    # y_train_onehot = np.zeros((y_train.size, out_feats))
+    # y_train_onehot[np.arange(y_train.size), y_train] = 1
+    # One-hot encode y_train
+    y_train_onehot = np.zeros((y_train.shape[0], out_feats))  # Initialize with zeros
+    y_train_onehot[np.arange(y_train.shape[0]), y_train] = 1  # Set the appropriate index to 1
+
+
+    # Training loop
+    for epoch in range(epochs):
+        epoch_loss = 0.0
+
+        for i in range(len(X_train)):
+            # Forward pass
+            inputs = [Value(x) for x in X_train[i]]
+            targets = [Value(y) for y in y_train_onehot[i]]
+            outputs = model(inputs)
+
+            # Calculate Cross-Entropy Loss
+            exp_outputs = [o.exp() for o in outputs]
+            sum_exp_outputs = sum(exp_outputs)
+            probs = [o / sum_exp_outputs for o in exp_outputs]
+            loss = -sum(t * p.log() for t, p in zip(targets, probs))
+
+            epoch_loss += loss.data
+
+            # Backpropagation
+            model.zero_grad()  # Zero gradients
+            loss.backward()
+
+            # Update weights
+            for param in model.parameters():
+                param.data -= learning_rate * param.grad
+
+        losses.append(epoch_loss / len(X_train))
+        if epoch % 10 == 0:
+            print(f"Epoch {epoch}, Loss: {epoch_loss / len(X_train)}")
+
+
+    # Evaluation
+    correct = 0
+    total = len(X_test)
+
+    for i in range(len(X_test)):
+        inputs = [Value(x) for x in X_test[i]]
+        outputs = model(inputs)
+        predicted = np.argmax([o.data for o in outputs])
+        if predicted == y_test[i]:
+            correct += 1
+
+    accuracy = correct / total
+    print(f"Test Accuracy: {accuracy * 100:.2f}%")
+
+    # import matplotlib.pyplot as plt
+
+    # Plot the losses
+    plt.plot(range(epochs), losses, label='Micrograd Loss')
+    plt.xlabel('Epochs')
+    plt.ylabel('Loss')
+    plt.title('Micrograd Loss Curve')
+    plt.legend()
+    plt.show()
+
+
+with_micrograd()
diff --git a/torchexample.py b/torchexample.py
index ec984d74..71327f7e 100644
--- a/torchexample.py
+++ b/torchexample.py
@@ -1,4 +1,3 @@
-# -*- coding: utf-8 -*-
 
 import torch
 from torch import nn
@@ -28,9 +27,9 @@
 train_dataloader = DataLoader(training_data, batch_size=batch_size)
 test_dataloader = DataLoader(test_data, batch_size=batch_size)
 
-for X, y in test_dataloader:
-    print(f"Shape of X [N, C, H, W]: {X.shape}")
-    print(f"Shape of y: {y.shape} {y.dtype}")
+for X_Gl, y_Gl in test_dataloader:
+    print(f"Shape of X [N, C, H, W]: {X_Gl.shape}")
+    print(f"Shape of y: {y_Gl.shape} {y_Gl.dtype}")
     break
 
 # Get cpu, gpu or mps device for training.
@@ -43,13 +42,14 @@
 )
 print(f"Using {device} device")
 
+
 # Define model
 class NeuralNetwork(nn.Module):
     def __init__(self):
         super().__init__()
         self.flatten = nn.Flatten()
         self.linear_relu_stack = nn.Sequential(
-            nn.Linear(28*28, 512),
+            nn.Linear(28 * 28, 512),
             nn.ReLU(),
             nn.Linear(512, 512),
             nn.ReLU(),
@@ -61,11 +61,13 @@ def forward(self, x):
         logits = self.linear_relu_stack(x)
         return logits
 
-model = NeuralNetwork().to(device)
-print(model)
 
-loss_fn = nn.CrossEntropyLoss()
-optimizer = torch.optim.SGD(model.parameters(), lr=1e-3)
+MODEL = NeuralNetwork().to(device)
+print(MODEL)
+
+LOSS_FN = nn.CrossEntropyLoss()
+OPTIMIZER = torch.optim.SGD(MODEL.parameters(), lr=1e-3)
+
 
 def train(dataloader, model, loss_fn, optimizer):
     size = len(dataloader.dataset)
@@ -86,6 +88,7 @@ def train(dataloader, model, loss_fn, optimizer):
             loss, current = loss.item(), (batch + 1) * len(X)
             print(f"loss: {loss:>7f}  [{current:>5d}/{size:>5d}]")
 
+
 def test(dataloader, model, loss_fn):
     size = len(dataloader.dataset)
     num_batches = len(dataloader)
@@ -99,11 +102,39 @@ def test(dataloader, model, loss_fn):
             correct += (pred.argmax(1) == y).type(torch.float).sum().item()
     test_loss /= num_batches
     correct /= size
-    print(f"Test Error: \n Accuracy: {(100*correct):>0.1f}%, Avg loss: {test_loss:>8f} \n")
-
-epochs = 5
-for t in range(epochs):
-    print(f"Epoch {t+1}\n-------------------------------")
-    train(train_dataloader, model, loss_fn, optimizer)
-    test(test_dataloader, model, loss_fn)
-print("Done!")
\ No newline at end of file
+    print(f"Test Error: \n Accuracy: {(100 * correct):>0.1f}%, Avg loss: {test_loss:>8f} \n")
+
+
+EPOCHS = 5  # 5
+for t in range(EPOCHS):
+    print(f"Epoch {t + 1}\n-------------------------------")
+    train(train_dataloader, MODEL, LOSS_FN, OPTIMIZER)
+    test(test_dataloader, MODEL, LOSS_FN)
+print("Done!")
+
+torch.save(MODEL.state_dict(), "model.pth")
+print("Saved PyTorch Model State to model.pth")
+
+MODEL_2 = NeuralNetwork().to(device)
+MODEL_2.load_state_dict(torch.load("model.pth", weights_only=True))
+
+classes = [
+    "T-shirt/top",
+    "Trouser",
+    "Pullover",
+    "Dress",
+    "Coat",
+    "Sandal",
+    "Shirt",
+    "Sneaker",
+    "Bag",
+    "Ankle boot",
+]
+
+MODEL.eval()
+x, y_ = test_data[0][0], test_data[0][1]
+with torch.no_grad():
+    x = x.to(device)
+    pred = MODEL(x)
+    predicted, actual = classes[pred[0].argmax(0)], classes[y_]
+    print(f'Predicted: "{predicted}", Actual: "{actual}"')
diff --git a/uv.lock b/uv.lock
index d2a49ee7..f2c81696 100644
--- a/uv.lock
+++ b/uv.lock
@@ -1,7 +1,8 @@
 version = 1
 requires-python = ">=3.10"
 resolution-markers = [
-    "python_full_version < '3.12'",
+    "python_full_version < '3.11'",
+    "python_full_version == '3.11.*'",
     "python_full_version >= '3.12'",
 ]
 
@@ -460,6 +461,7 @@ dependencies = [
     { name = "black" },
     { name = "graphviz" },
     { name = "matplotlib" },
+    { name = "pandas" },
     { name = "pytest" },
     { name = "ruff" },
     { name = "scikit-learn" },
@@ -473,6 +475,7 @@ requires-dist = [
     { name = "black", specifier = ">=24.10.0" },
     { name = "graphviz", specifier = ">=0.20.3" },
     { name = "matplotlib", specifier = ">=3.7.5" },
+    { name = "pandas", specifier = ">=2.2.3" },
     { name = "pytest", specifier = ">=8.3.4" },
     { name = "ruff", specifier = ">=0.8.1" },
     { name = "scikit-learn", specifier = ">=1.3.2" },
@@ -699,6 +702,54 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/88/ef/eb23f262cca3c0c4eb7ab1933c3b1f03d021f2c48f54763065b6f0e321be/packaging-24.2-py3-none-any.whl", hash = "sha256:09abb1bccd265c01f4a3aa3f7a7db064b36514d2cba19a2f694fe6150451a759", size = 65451 },
 ]
 
+[[package]]
+name = "pandas"
+version = "2.2.3"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "numpy" },
+    { name = "python-dateutil" },
+    { name = "pytz" },
+    { name = "tzdata" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/9c/d6/9f8431bacc2e19dca897724cd097b1bb224a6ad5433784a44b587c7c13af/pandas-2.2.3.tar.gz", hash = "sha256:4f18ba62b61d7e192368b84517265a99b4d7ee8912f8708660fb4a366cc82667", size = 4399213 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/aa/70/c853aec59839bceed032d52010ff5f1b8d87dc3114b762e4ba2727661a3b/pandas-2.2.3-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:1948ddde24197a0f7add2bdc4ca83bf2b1ef84a1bc8ccffd95eda17fd836ecb5", size = 12580827 },
+    { url = "https://files.pythonhosted.org/packages/99/f2/c4527768739ffa4469b2b4fff05aa3768a478aed89a2f271a79a40eee984/pandas-2.2.3-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:381175499d3802cde0eabbaf6324cce0c4f5d52ca6f8c377c29ad442f50f6348", size = 11303897 },
+    { url = "https://files.pythonhosted.org/packages/ed/12/86c1747ea27989d7a4064f806ce2bae2c6d575b950be087837bdfcabacc9/pandas-2.2.3-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:d9c45366def9a3dd85a6454c0e7908f2b3b8e9c138f5dc38fed7ce720d8453ed", size = 66480908 },
+    { url = "https://files.pythonhosted.org/packages/44/50/7db2cd5e6373ae796f0ddad3675268c8d59fb6076e66f0c339d61cea886b/pandas-2.2.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:86976a1c5b25ae3f8ccae3a5306e443569ee3c3faf444dfd0f41cda24667ad57", size = 13064210 },
+    { url = "https://files.pythonhosted.org/packages/61/61/a89015a6d5536cb0d6c3ba02cebed51a95538cf83472975275e28ebf7d0c/pandas-2.2.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:b8661b0238a69d7aafe156b7fa86c44b881387509653fdf857bebc5e4008ad42", size = 16754292 },
+    { url = "https://files.pythonhosted.org/packages/ce/0d/4cc7b69ce37fac07645a94e1d4b0880b15999494372c1523508511b09e40/pandas-2.2.3-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:37e0aced3e8f539eccf2e099f65cdb9c8aa85109b0be6e93e2baff94264bdc6f", size = 14416379 },
+    { url = "https://files.pythonhosted.org/packages/31/9e/6ebb433de864a6cd45716af52a4d7a8c3c9aaf3a98368e61db9e69e69a9c/pandas-2.2.3-cp310-cp310-win_amd64.whl", hash = "sha256:56534ce0746a58afaf7942ba4863e0ef81c9c50d3f0ae93e9497d6a41a057645", size = 11598471 },
+    { url = "https://files.pythonhosted.org/packages/a8/44/d9502bf0ed197ba9bf1103c9867d5904ddcaf869e52329787fc54ed70cc8/pandas-2.2.3-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:66108071e1b935240e74525006034333f98bcdb87ea116de573a6a0dccb6c039", size = 12602222 },
+    { url = "https://files.pythonhosted.org/packages/52/11/9eac327a38834f162b8250aab32a6781339c69afe7574368fffe46387edf/pandas-2.2.3-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:7c2875855b0ff77b2a64a0365e24455d9990730d6431b9e0ee18ad8acee13dbd", size = 11321274 },
+    { url = "https://files.pythonhosted.org/packages/45/fb/c4beeb084718598ba19aa9f5abbc8aed8b42f90930da861fcb1acdb54c3a/pandas-2.2.3-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:cd8d0c3be0515c12fed0bdbae072551c8b54b7192c7b1fda0ba56059a0179698", size = 15579836 },
+    { url = "https://files.pythonhosted.org/packages/cd/5f/4dba1d39bb9c38d574a9a22548c540177f78ea47b32f99c0ff2ec499fac5/pandas-2.2.3-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c124333816c3a9b03fbeef3a9f230ba9a737e9e5bb4060aa2107a86cc0a497fc", size = 13058505 },
+    { url = "https://files.pythonhosted.org/packages/b9/57/708135b90391995361636634df1f1130d03ba456e95bcf576fada459115a/pandas-2.2.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:63cc132e40a2e084cf01adf0775b15ac515ba905d7dcca47e9a251819c575ef3", size = 16744420 },
+    { url = "https://files.pythonhosted.org/packages/86/4a/03ed6b7ee323cf30404265c284cee9c65c56a212e0a08d9ee06984ba2240/pandas-2.2.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:29401dbfa9ad77319367d36940cd8a0b3a11aba16063e39632d98b0e931ddf32", size = 14440457 },
+    { url = "https://files.pythonhosted.org/packages/ed/8c/87ddf1fcb55d11f9f847e3c69bb1c6f8e46e2f40ab1a2d2abadb2401b007/pandas-2.2.3-cp311-cp311-win_amd64.whl", hash = "sha256:3fc6873a41186404dad67245896a6e440baacc92f5b716ccd1bc9ed2995ab2c5", size = 11617166 },
+    { url = "https://files.pythonhosted.org/packages/17/a3/fb2734118db0af37ea7433f57f722c0a56687e14b14690edff0cdb4b7e58/pandas-2.2.3-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:b1d432e8d08679a40e2a6d8b2f9770a5c21793a6f9f47fdd52c5ce1948a5a8a9", size = 12529893 },
+    { url = "https://files.pythonhosted.org/packages/e1/0c/ad295fd74bfac85358fd579e271cded3ac969de81f62dd0142c426b9da91/pandas-2.2.3-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:a5a1595fe639f5988ba6a8e5bc9649af3baf26df3998a0abe56c02609392e0a4", size = 11363475 },
+    { url = "https://files.pythonhosted.org/packages/c6/2a/4bba3f03f7d07207481fed47f5b35f556c7441acddc368ec43d6643c5777/pandas-2.2.3-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:5de54125a92bb4d1c051c0659e6fcb75256bf799a732a87184e5ea503965bce3", size = 15188645 },
+    { url = "https://files.pythonhosted.org/packages/38/f8/d8fddee9ed0d0c0f4a2132c1dfcf0e3e53265055da8df952a53e7eaf178c/pandas-2.2.3-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fffb8ae78d8af97f849404f21411c95062db1496aeb3e56f146f0355c9989319", size = 12739445 },
+    { url = "https://files.pythonhosted.org/packages/20/e8/45a05d9c39d2cea61ab175dbe6a2de1d05b679e8de2011da4ee190d7e748/pandas-2.2.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:6dfcb5ee8d4d50c06a51c2fffa6cff6272098ad6540aed1a76d15fb9318194d8", size = 16359235 },
+    { url = "https://files.pythonhosted.org/packages/1d/99/617d07a6a5e429ff90c90da64d428516605a1ec7d7bea494235e1c3882de/pandas-2.2.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:062309c1b9ea12a50e8ce661145c6aab431b1e99530d3cd60640e255778bd43a", size = 14056756 },
+    { url = "https://files.pythonhosted.org/packages/29/d4/1244ab8edf173a10fd601f7e13b9566c1b525c4f365d6bee918e68381889/pandas-2.2.3-cp312-cp312-win_amd64.whl", hash = "sha256:59ef3764d0fe818125a5097d2ae867ca3fa64df032331b7e0917cf5d7bf66b13", size = 11504248 },
+    { url = "https://files.pythonhosted.org/packages/64/22/3b8f4e0ed70644e85cfdcd57454686b9057c6c38d2f74fe4b8bc2527214a/pandas-2.2.3-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:f00d1345d84d8c86a63e476bb4955e46458b304b9575dcf71102b5c705320015", size = 12477643 },
+    { url = "https://files.pythonhosted.org/packages/e4/93/b3f5d1838500e22c8d793625da672f3eec046b1a99257666c94446969282/pandas-2.2.3-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:3508d914817e153ad359d7e069d752cdd736a247c322d932eb89e6bc84217f28", size = 11281573 },
+    { url = "https://files.pythonhosted.org/packages/f5/94/6c79b07f0e5aab1dcfa35a75f4817f5c4f677931d4234afcd75f0e6a66ca/pandas-2.2.3-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:22a9d949bfc9a502d320aa04e5d02feab689d61da4e7764b62c30b991c42c5f0", size = 15196085 },
+    { url = "https://files.pythonhosted.org/packages/e8/31/aa8da88ca0eadbabd0a639788a6da13bb2ff6edbbb9f29aa786450a30a91/pandas-2.2.3-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f3a255b2c19987fbbe62a9dfd6cff7ff2aa9ccab3fc75218fd4b7530f01efa24", size = 12711809 },
+    { url = "https://files.pythonhosted.org/packages/ee/7c/c6dbdb0cb2a4344cacfb8de1c5808ca885b2e4dcfde8008266608f9372af/pandas-2.2.3-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:800250ecdadb6d9c78eae4990da62743b857b470883fa27f652db8bdde7f6659", size = 16356316 },
+    { url = "https://files.pythonhosted.org/packages/57/b7/8b757e7d92023b832869fa8881a992696a0bfe2e26f72c9ae9f255988d42/pandas-2.2.3-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:6374c452ff3ec675a8f46fd9ab25c4ad0ba590b71cf0656f8b6daa5202bca3fb", size = 14022055 },
+    { url = "https://files.pythonhosted.org/packages/3b/bc/4b18e2b8c002572c5a441a64826252ce5da2aa738855747247a971988043/pandas-2.2.3-cp313-cp313-win_amd64.whl", hash = "sha256:61c5ad4043f791b61dd4752191d9f07f0ae412515d59ba8f005832a532f8736d", size = 11481175 },
+    { url = "https://files.pythonhosted.org/packages/76/a3/a5d88146815e972d40d19247b2c162e88213ef51c7c25993942c39dbf41d/pandas-2.2.3-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:3b71f27954685ee685317063bf13c7709a7ba74fc996b84fc6821c59b0f06468", size = 12615650 },
+    { url = "https://files.pythonhosted.org/packages/9c/8c/f0fd18f6140ddafc0c24122c8a964e48294acc579d47def376fef12bcb4a/pandas-2.2.3-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:38cf8125c40dae9d5acc10fa66af8ea6fdf760b2714ee482ca691fc66e6fcb18", size = 11290177 },
+    { url = "https://files.pythonhosted.org/packages/ed/f9/e995754eab9c0f14c6777401f7eece0943840b7a9fc932221c19d1abee9f/pandas-2.2.3-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:ba96630bc17c875161df3818780af30e43be9b166ce51c9a18c1feae342906c2", size = 14651526 },
+    { url = "https://files.pythonhosted.org/packages/25/b0/98d6ae2e1abac4f35230aa756005e8654649d305df9a28b16b9ae4353bff/pandas-2.2.3-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1db71525a1538b30142094edb9adc10be3f3e176748cd7acc2240c2f2e5aa3a4", size = 11871013 },
+    { url = "https://files.pythonhosted.org/packages/cc/57/0f72a10f9db6a4628744c8e8f0df4e6e21de01212c7c981d31e50ffc8328/pandas-2.2.3-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:15c0e1e02e93116177d29ff83e8b1619c93ddc9c49083f237d4312337a61165d", size = 15711620 },
+    { url = "https://files.pythonhosted.org/packages/ab/5f/b38085618b950b79d2d9164a711c52b10aefc0ae6833b96f626b7021b2ed/pandas-2.2.3-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:ad5b65698ab28ed8d7f18790a0dc58005c7629f227be9ecc1072aa74c0c1d43a", size = 13098436 },
+]
+
 [[package]]
 name = "pathspec"
 version = "0.12.1"
@@ -844,6 +895,15 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/ec/57/56b9bcc3c9c6a792fcbaf139543cee77261f3651ca9da0c93f5c1221264b/python_dateutil-2.9.0.post0-py2.py3-none-any.whl", hash = "sha256:a8b2bc7bffae282281c8140a97d3aa9c14da0b136dfe83f850eea9a5f7470427", size = 229892 },
 ]
 
+[[package]]
+name = "pytz"
+version = "2024.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/3a/31/3c70bf7603cc2dca0f19bdc53b4537a797747a58875b552c8c413d963a3f/pytz-2024.2.tar.gz", hash = "sha256:2aa355083c50a0f93fa581709deac0c9ad65cca8a9e9beac660adcbd493c798a", size = 319692 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/11/c3/005fcca25ce078d2cc29fd559379817424e94885510568bc1bc53d7d5846/pytz-2024.2-py2.py3-none-any.whl", hash = "sha256:31c7c1817eb7fae7ca4b8c7ee50c72f93aa2dd863de768e1ef4245d426aa0725", size = 508002 },
+]
+
 [[package]]
 name = "ruff"
 version = "0.8.1"
@@ -1134,6 +1194,15 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/26/9f/ad63fc0248c5379346306f8668cda6e2e2e9c95e01216d2b8ffd9ff037d0/typing_extensions-4.12.2-py3-none-any.whl", hash = "sha256:04e5ca0351e0f3f85c6853954072df659d0d13fac324d0072316b67d7794700d", size = 37438 },
 ]
 
+[[package]]
+name = "tzdata"
+version = "2024.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/e1/34/943888654477a574a86a98e9896bae89c7aa15078ec29f490fef2f1e5384/tzdata-2024.2.tar.gz", hash = "sha256:7d85cc416e9382e69095b7bdf4afd9e3880418a2413feec7069d533d6b4e31cc", size = 193282 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a6/ab/7e5f53c3b9d14972843a647d8d7a853969a58aecc7559cb3267302c94774/tzdata-2024.2-py2.py3-none-any.whl", hash = "sha256:a48093786cdcde33cad18c2555e8532f34422074448fbc874186f0abd79565cd", size = 346586 },
+]
+
 [[package]]
 name = "virtualenv"
 version = "20.28.0"