TensorFlow GPU Installation: EASIEST METHOD YET!
Вставка
- Опубліковано 4 жов 2024
- In this video, I'll show you the simplest and fastest way to install TensorFlow 2.17.0 with GPU support on WSL (Windows Subsystem for Linux) using Ubuntu 24.04. You won't need to go through the hassle of manually installing or configuring tools like CUDA Toolkit or cuDNN. Just follow these easy steps to get your machine up and running with TensorFlow for deep learning and AI development in no time!
💻 What You'll Learn in This Video:
Installing TensorFlow 2.17.0 GPU version on WSL Ubuntu 24.04
Skipping manual CUDA and cuDNN configurations
Simplified setup for developers and AI enthusiasts
🔧 No additional setup needed! This method is perfect if you want to quickly dive into TensorFlow without worrying about configuring complex environments. Whether you're new to WSL or a seasoned developer, this guide is designed to save you time and effort.
⚡ Why Watch?
Easy TensorFlow 2.17 GPU setup in just a few steps
No need for multiple tool installations
Great for AI, deep learning, and machine learning workflows
👍 Like this video if it helped you set up TensorFlow quickly! Don't forget to subscribe for more tutorials on AI tools, machine learning, and development workflows.
###### Commands #######
wsl --install -d Ubuntu-24.04
python3 --version
nvidia-smi
apt install python3.12-venv
python3.12 -m venv tf217
source tf217/bin/activate
pip install --upgrade pip
pip install tensorflow[and-cuda]
pip install ipykernel
python3.12 -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))"
Helped me a lot. Thank you. It installed fine on Windows 10 with intel CPU and nvidia GPU.
hello sir, can you please make a video on how to install chrome remote desktop on Ubuntu 24.04 for remote access from other OS (e.g. Windows)? I found out there are very little information for Chrome Remote Desktop installation in Ubuntu 24.04, and alot people are encountering issues while installing it. It would be helpful if you can make a video about it. Thank you.
sir plz github source link
Hi, I posted all the commands and the codes on the description
@@TechJotters24
sir i want this code
# Define a simple CNN model
def create_model():
model = tf.keras.models.Sequential([
tf.keras.layers.Conv2D(32, (3, 3), activation='relu', input_shape=(28, 28, 1)),
tf.keras.layers.MaxPooling2D((2, 2)),
tf.keras.layers.Conv2D (64, (3, 3), activation='relu'),
tf.keras.layers.MaxPooling2D((2, 2)),
tf.keras.layers. Flatten(),
tf.keras.layers. Dense (64, activation='relu'),
tf.keras.layers. Dense(10, activation='softmax'),
])
model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])
return model
# Function to train and time the mode 3.6x)
def train_model(device_name):
with tf.device(device_name):
model = create_model()
start_time = time.time()
model.fit(x_train, y_train, epochs=5, batch_size=64, verbose=0)
end_time = time.time()
return end_time - start_time
# Train on CPU
cpu_time = train_model('/CPU:0')
print(f'Time taken to train on 3xJ: {cpu_time:.2f} seconds')
# Train on GPU (if available)
if tf.config.list_physical_devices('GPU'):
gpu_time = train_model('/GPU:0')
print(f'Time taken to train on GPU: {gpu_time:.2f} seconds')
else:
print("No GPU found.")