📌 EDA in Public (Part 1): Cleaning and Exploring Sales Data with Pandas

🗂 Category: PROGRAMMING

🕒 Date: 2025-12-12 | ⏱️ Read time: 11 min read

Hey everyone! Welcome to the start of a major data journey that I’m calling “EDA…

#DataScience #AI #Python

🤖🧠 S3PRL Toolkit: Advancing Self-Supervised Speech Representation Learning

🗓️ 13 Dec 2025
📚 AI News & Trends

The field of speech technology has witnessed a transformative shift in recent years, powered by the rise of self-supervised learning (SSL). Instead of relying on large amounts of labeled data, self-supervised models learn from the patterns and structures inherent in raw audio, enabling powerful and general-purpose speech representations. At the forefront of this innovation stands ...

#S3PRL #SelfSupervisedLearning #SpeechTechnology #SSL #SpeechRepresentationLearning #AI

📌 Spectral Community Detection in Clinical Knowledge Graphs

🗂 Category: GRAPH THEORY

🕒 Date: 2025-12-12 | ⏱️ Read time: 22 min read

Introduction How do we identify latent groups of patients in a large cohort? How can…

#DataScience #AI #Python

💡 Cons & Pros of Naive Bayes Algorithm

Naive Bayes is a #classification algorithm that is widely used in #machinelearning and #naturallanguageprocessing tasks. It is based on Bayes’ theorem, which describes the probability of an event based on prior knowledge of conditions related to that event. While Naive Bayes has its advantages, it also has some limitations.

💡 Pros of Naive Bayes:

1️⃣ Simplicity and efficiency
Naive Bayes is a simple and computationally efficient algorithm that is easy to understand and implement. It requires a relatively small amount of training data to estimate the parameters needed for classification.

2️⃣ Fast training and prediction
Due to its simplicity, Naive Bayes has fast training and inference compared to more complex algorithms, which makes it suitable for large-scale and real-time applications.

3️⃣ Handles high-dimensional data
Naive Bayes performs well even when the number of features is large compared to the number of samples. It scales effectively in high-dimensional spaces, which is why it is popular in text classification and spam filtering.

4️⃣ Works well with categorical data
Naive Bayes naturally supports categorical or discrete features, and variants like Multinomial and Bernoulli Naive Bayes are especially effective for text and count data. Continuous features can be handled with Gaussian Naive Bayes or by discretization.

5️⃣ Robust to many irrelevant features
Because each feature contributes independently to the final probability, many irrelevant features tend not to hurt performance severely, especially when there is enough data.

💡 Cons of Naive Bayes:

1️⃣ Strong independence assumption
The core limitation is the assumption that features are conditionally independent given the class, which is rarely true in real-world data and can degrade performance when strong feature interactions exist.

2️⃣ Lack of feature interactions
Naive Bayes cannot model complex relationships or interactions between features. Each feature influences the prediction on its own, which limits the model’s expressiveness compared to methods like trees, SVMs, or neural networks.

3️⃣ Sensitivity to imbalanced data
With highly imbalanced class distributions, posterior probabilities can become dominated by the majority class, causing poor performance on minority classes unless you rebalance or adjust priors.

4️⃣ Limited representation power
Naive Bayes works best when class boundaries are relatively simple. For complex, non-linear decision boundaries, more flexible models (e.g., SVMs, ensembles, neural networks) usually achieve higher accuracy.

5️⃣ Reliance on good-quality data
The algorithm is sensitive to noisy data, missing values, and rare events. Zero-frequency problems (unseen feature–class combinations) can cause zero probabilities unless techniques like Laplace smoothing are used.

📌 How to Increase Coding Iteration Speed

🗂 Category: LLM APPLICATIONS

🕒 Date: 2025-12-13 | ⏱️ Read time: 8 min read

Learn how to become a more efficient programmer with local testing

#DataScience #AI #Python

📌 NeurIPS 2025 Best Paper Review: Qwen’s Systematic Exploration of Attention Gating

🗂 Category: LARGE LANGUAGE MODELS

🕒 Date: 2025-12-13 | ⏱️ Read time: 27 min read

This one little trick can bring about enhanced training stability, the use of larger learning…

#DataScience #AI #Python

📌 The Machine Learning “Advent Calendar” Day 13: LASSO and Ridge Regression in Excel

🗂 Category: MACHINE LEARNING

🕒 Date: 2025-12-13 | ⏱️ Read time: 7 min read

Ridge and Lasso regression are often perceived as more complex versions of linear regression. In…

#DataScience #AI #Python

📌 The Skills That Bridge Technical Work and Business Impact

🗂 Category: AUTHOR SPOTLIGHTS

🕒 Date: 2025-12-14 | ⏱️ Read time: 10 min read

In the Author Spotlight series, TDS Editors chat with members of our community about their…

#DataScience #AI #Python

📌 Stop Writing Spaghetti if-else Chains: Parsing JSON with Python’s match-case

🗂 Category: PROGRAMMING

🕒 Date: 2025-12-14 | ⏱️ Read time: 6 min read

Introduction If you work in data science, data engineering, or as as a frontend/backend developer,…

#DataScience #AI #Python

📌 The Machine Learning “Advent Calendar” Day 14: Softmax Regression in Excel

🗂 Category: MACHINE LEARNING

🕒 Date: 2025-12-14 | ⏱️ Read time: 7 min read

Softmax Regression is simply Logistic Regression extended to multiple classes. By computing one linear score…

#DataScience #AI #Python

📌 6 Technical Skills That Make You a Senior Data Scientist

🗂 Category: DATA SCIENCE

🕒 Date: 2025-12-15 | ⏱️ Read time: 11 min read

Beyond writing code, these are the design-level decisions, trade-offs, and habits that quietly separate senior…

#DataScience #AI #Python

📌 Geospatial exploratory data analysis with GeoPandas and DuckDB

🗂 Category: PROGRAMMING

🕒 Date: 2025-12-15 | ⏱️ Read time: 13 min read

In this article, I’ll show you how to use two popular Python libraries to carry…

#DataScience #AI #Python

📌 Lessons Learned from Upgrading to LangChain 1.0 in Production

🗂 Category: AGENTIC AI

🕒 Date: 2025-12-15 | ⏱️ Read time: 5 min read

What worked, what broke, and why I did it

#DataScience #AI #Python

Machine Learning Fundamentals

A structured Machine Learning Fundamentals guide covering core concepts, intuition, math basics, ML algorithms, deep learning, and real-world workflows.

https://news.1rj.ru/str/DataScienceM 🩷

Try the bot with a large search database within Petligram

Tip: Optimize PyTorch Model Performance with torch.compile

Explanation:
torch.compile (introduced in PyTorch 2.0) is a powerful JIT (Just-In-Time) compiler that automatically transforms your PyTorch model into highly optimized, high-performance code. It works by analyzing your model's computation graph, fusing operations, eliminating redundant computations, and compiling them into efficient kernels (e.g., using Triton for GPU acceleration). This significantly reduces Python overhead and improves memory locality, leading to substantial speedups (often 30-50% or more) during training and inference, especially on GPUs and for larger models, without requiring changes to your model architecture or training loop. The primary dynamic mode intelligently compiles subgraphs as they are encountered, providing a balance of performance and flexibility.

Example:

import torch
import torch.nn as nn
import time

# Define a simple neural network
class SimpleNet(nn.Module):
    def __init__(self):
        super().__init__()
        self.fc1 = nn.Linear(1024, 2048)
        self.relu = nn.ReLU()
        self.fc2 = nn.Linear(2048, 1024)
        self.dropout = nn.Dropout(0.2)

    def forward(self, x):
        x = self.fc1(x)
        x = self.relu(x)
        x = self.dropout(x)
        x = self.fc2(x)
        return x

# Prepare model and dummy data
device = "cuda" if torch.cuda.is_available() else "cpu"
model = SimpleNet().to(device)
dummy_input = torch.randn(128, 1024).to(device)
dummy_target = torch.randn(128, 1024).to(device)
optimizer = torch.optim.Adam(model.parameters(), lr=0.001)
criterion = nn.MSELoss()
num_iterations = 50

# --- Benchmark without torch.compile ---
print(f"--- Running without torch.compile on {device} ---")
start_time = time.time()
for _ in range(num_iterations):
    optimizer.zero_grad()
    output = model(dummy_input)
    loss = criterion(output, dummy_target)
    loss.backward()
    optimizer.step()
if device == "cuda":
    torch.cuda.synchronize() # Wait for GPU ops to complete
time_uncompiled = time.time() - start_time
print(f"Time without compile: {time_uncompiled:.4f} seconds\n")

# --- Benchmark with torch.compile ---
# Apply torch.compile to the model. This happens once upfront.
# The default backend 'inductor' is typically the best performing.
compiled_model = torch.compile(model)
# Ensure optimizer is correctly set up for the compiled model's parameters
# (in this case, `compiled_model` shares parameters with `model`, so no re-init needed if parameters are the same object)

print(f"--- Running with torch.compile on {device} ---")
start_time = time.time()
for _ in range(num_iterations):
    optimizer.zero_grad()
    output = compiled_model(dummy_input) # Use the compiled model
    loss = criterion(output, dummy_target)
    loss.backward()
    optimizer.step()
if device == "cuda":
    torch.cuda.synchronize() # Wait for GPU ops to complete
time_compiled = time.time() - start_time
print(f"Time with compile: {time_compiled:.4f} seconds")

if time_uncompiled > 0:
    print(f"\nSpeedup: {time_uncompiled / time_compiled:.2f}x")

━━━━━━━━━━━━━━━
By: @DataScienceM ✨

📌 The Machine Learning “Advent Calendar” Day 15: SVM in Excel

🗂 Category: MACHINE LEARNING

🕒 Date: 2025-12-15 | ⏱️ Read time: 12 min read

Instead of starting with margins and geometry, this article builds the Support Vector Machine step…

#DataScience #AI #Python

The bot is more mm great for finding anything in Telegram

📌 When (Not) to Use Vector DB

🗂 Category: LARGE LANGUAGE MODELS

🕒 Date: 2025-12-16 | ⏱️ Read time: 8 min read

When indexing hurts more than it helps: how we realized our RAG use case needed…

#DataScience #AI #Python