Your goal is to train and evaluate a simple neural network using the palmer penguins dataset, which contains information about penguin species and various physical measurements.

1. Load the Dataset:

   • Read the penguins.csv file into a Pandas DataFrame.
   • Handle missing values by dropping rows with any NaN values.
   • Encode the species column as integers (target variable).

2. Prepare Features and Target:

   • Select bill_length_mm, bill_depth_mm, flipper_length_mm, and body_mass_g as features.
   • Encode the species column as the target.

3. Split the Data:

   • Use an 80-20 split for train-test data.

4. Create PyTorch Tensors:

   • Convert the features into FloatTensor and the target into LongTensor.

5. Define a Neural Network:

   • Create a model class with one hidden layer using PyTorch’s nn.Module.

6. Train the Model:

   • Use the CrossEntropyLoss function and the Adam optimizer.
   • Train for 100 epochs and print the loss every 10 epochs.

7. Evaluate the Model:

   • Calculate test accuracy.
   • Display the confusion matrix to analyze the predictions.