Claude Code install shell script

 install Claude Code using this shell script

.

#!/bin/bash

# install-claude.sh

# Installs Node.js 18+, npm, and Claude Code CLI, handling common conflicts.

set -e

echo "=== [1/5] Removing conflicting Node.js and libnode-dev packages ==="

apt-get remove -y nodejs nodejs-doc libnode-dev || true

echo "=== [2/5] Cleaning up package manager ==="

apt-get autoremove -y

apt-get clean

echo "=== [3/5] Installing Node.js 18.x and npm ==="

curl -fsSL https://deb.nodesource.com/setup_18.x | bash -

apt-get install -y nodejs

# Ensure npm is installed (should be bundled with Node.js 18.x)

if ! command -v npm &>/dev/null; then

echo "npm not found, installing..."

apt-get install -y npm

fi

echo "=== [4/5] Verifying Node.js and npm installation ==="

node --version

npm --version

echo "=== [5/5] Installing Claude Code CLI ==="

npm install -g @anthropic-ai/claude-code

echo "=== Claude Code CLI installed successfully! ==="

echo "Run 'claude' to get started."

..

Thank you!

What is "GPU_MAX_HW_QUEUES"?

 GPU_MAX_HW_QUEUES controls the maximum number of hardware queues that can be used for GPU operations in ROCm/HIP environment (AMD GPUs).

Specifically:

1. Each hardware queue can handle GPU operations (kernels) independently
2. Default value is typically 8
3. In your script, it's set to 4: export GPU_MAX_HW_QUEUES=4

This setting affects:

• Parallel kernel execution
• Memory transfers
• Overall GPU utilization
• Resource allocation

Lower values (like 4) can:

• Reduce overhead
• Improve memory usage
• Lead to more predictable performance
• Be better for large models like LLaMA 70B

Higher values can:

• Allow more parallel operations
• Potentially improve throughput for smaller models
• Use more GPU resources

In your case, setting GPU_MAX_HW_QUEUES=4 is a conservative choice that helps manage memory and scheduling overhead when training the large LLaMA 70B model.

How many GPUs do I need to train a LLM?

How many GPUs do I need to train a LLM?

This is a complicated question in general, but if we assume that you are using FSDP with 

FULL_SHARD, activation checkpointing, and DecoupledLionW, then a good rule of thumb is:

Your total cluster memory in GB should be larger than 12 * N (# billions of params).

E.g. To train a GPT-13B model which has ~13 billion params, 

have at least 12 * 13 = 156 GB of total memory across your GPUs. 

You can accomplish this with 4xA100-40GB, or 2xA100-80GB, etc.

If you run into OOM errors when using small device counts, 

reduce device_train_microbatch_size until it succeeds.

Keep in mind: even though training will work in these minimalist settings, 

you will get much better throughput_per_device 

if you use a larger cluster or devices with higher memory capacity, 

because this will enable you to use larger microbatch sizes.

Unknown parameter in retrievalConfiguration.vectorSearchConfiguration: "overrideSearchType", must be one of: numberOfResults

 Error in AWS bedrock like:

Unknown parameter in retrievalConfiguration.vectorSearchConfiguration: "overrideSearchType", must be one of: numberOfResults

Solution 

Update boto3 sdk as latest one.

The parameters changed on 2024-03-27.

refer to here: https://awsapichanges.com/archive/changes/cd42c1-bedrock-agent-runtime.html

Thank you!