meikuam

# basic stuff
sudo apt install \
    curl \
    wget \
    vim \
    git \
    cmake \
    unzip \

meikuam

1. STOP THE DOCKER DAEMON

sudo service docker stop

2. ADD A CONFIGURATION FILE TO TELL THE DOCKER DAEMON WHAT IS THE LOCATION OF THE DATA DIRECTORY Using your preferred text editor add a file named daemon.json under the directory /etc/docker. The file should have this content:

{
  "data-root": "/path/to/your/docker"
}

meikuam

int led_pins[6] = {3, 5, 6, 9, 10, 11};
byte led_values[6] = {128, 128, 128, 128, 128, 128};

int button0_pin = 2;
int button0_state = 0;


void setup_leds() {
  // вроде как это не обязательно при использовании analogWrite
  for (int i = 0; i < 6; i++) {

meikuam

"""
this script takes `image.png` as input, 
resize it to 63pix height, 
then binarize it, 
to create `_Bootscreen.h` file

`_Bootscreen.h` can be used as part of https://github.com/MarlinFirmware/Marlin

I used it with: https://github.com/bigtreetech/BIGTREETECH-SKR-mini-E3/tree/master/firmware/V2.0/Marlin-2.0.7.2-SKR-mini-E3-V2.0

meikuam

>>> import graphviz
>>> import torch
>>> import torch.nn as nn
>>> x = torch.randn([1, 1, 16])
>>> x.shape
torch.Size([1, 1, 16])
>>> rnn = nn.LSTM(
...             input_size=16,
...             hidden_size=8,
...             num_layers=2,

meikuam

# build tensorflow2.2.0
sudo apt-get install -y \
    protobuf-compiler python3-pil python3-lxml python3-tk cython \
    autoconf automake libtool curl make g++ unzip wget git \
    libgflags-dev libgoogle-glog-dev liblmdb-dev libleveldb-dev \
    libhdf5-serial-dev libhdf5-dev python3-opencv\
    python3-dev python3-numpy python3-skimage gfortran libturbojpeg \
    libboost-all-dev libopenblas-dev libsnappy-dev software-properties-common \
    libfreetype6-dev pkg-config \
    libpng-dev libhdf5-103 libhdf5-cpp-103 libc-ares-dev libblas-dev \

meikuam

# openrazer-meta from repo
sudo apt install python3-setproctitle xautomation python3-pyudev -y

git clone --depth=50 https://github.com/openrazer/openrazer.git && cd openrazer

./scripts/build_debs.sh

sudo dpkg -i openrazer-driver-dkms_2.7.0-0_all.deb
sudo gpasswd -a $USER plugdev
sudo dpkg -i openrazer-daemon_2.7.0-0_all.deb

meikuam

ml roadmap

курсы

• курс ods
• cs231n site
• cs231n video
• cs224n

полезная штука

• можно начать отсюда

meikuam

import numpy as np

# SGD
while True:

    #....
    dx = network.backward()
    x = x - learning_rate * dx

meikuam

List of desktop Nvidia GPUS ordered by CUDA core count

Compute Capability from (https://developer.nvidia.com/cuda-gpus)

GPU	CUDA cores	Memory	Processor frequency	Compute Capability
GeForce GTX TITAN Z	5760	12 GB	705 / 876	3.5
NVIDIA TITAN Xp	3840	12 GB	1582	6.1
GeForce GTX 1080 Ti	3584	11 GB	1582	6.1
GeForce GTX TITAN X	3072	12 GB	1000 / 1075	6.1