Skip to content

Instantly share code, notes, and snippets.

Alexander Veysov snakers4

Block or report user

Report or block snakers4

Hide content and notifications from this user.

Learn more about blocking users

Contact Support about this user’s behavior.

Learn more about reporting abuse

Report abuse
View GitHub Profile
@snakers4
snakers4 / modeling.py
Created Mar 1, 2019
Best pretraining for Russian language - embedding bag interfaces
View modeling.py
class BertEmbeddingBag(nn.Module):
"""Construct the embeddings from word, position and token_type embeddings.
"""
def __init__(self, config):
super(BertEmbeddingBag, self).__init__()
# self.word_embeddings = nn.Embedding(config.vocab_size, config.hidden_size)
ngram_matrix=np.load(config.ngram_matrix_path)
self.old_bag = config.old_bag
@snakers4
snakers4 / send_test_email.sh
Last active Feb 24, 2019
Plain temperature monitoring in Ubuntu 18.04
View send_test_email.sh
{
echo To: YOUR_EMAIL@gmail.com
echo From: YOUR_EMAIL@gmail.com
echo Subject: Temperature warning! $TIMESTAMP
echo Current CPU temperature is $TEMP
} | ssmtp YOUR_EMAIL@gmail.com
@snakers4
snakers4 / pandas_multiprocessing_wrappers.py
Created Dec 14, 2018
Pandas multiprocessing wrappers
View pandas_multiprocessing_wrappers.py
from tqdm import tqdm
import numpy as np
import pandas as pd
from multiprocessing import Pool
def _apply_df(args):
df, func, num, kwargs = args
return num, df.apply(func, **kwargs)
def apply_by_multiprocessing(df,func,**kwargs):
@snakers4
snakers4 / parse_cc_index.py
Last active Jul 1, 2019
Plain common crawl pre-processing
View parse_cc_index.py
import gc
import gzip
import time
import json
import shutil
import os,sys
import tldextract
import collections
import pandas as pd
from tqdm import tqdm
@snakers4
snakers4 / parse_cc_index.py
Last active May 23, 2019
Plain scripts to parse Common Crawl
View parse_cc_index.py
import gc
import gzip
import time
import json
import shutil
import os,sys
import tldextract
import collections
import pandas as pd
from tqdm import tqdm
@snakers4
snakers4 / process_wikipedia.py
Last active Jul 17, 2019
Post process wikipedia files produced by wikiextractor
View process_wikipedia.py
import os
import re
import sys
import glob
import nltk
import gensim
import numpy as np
import pandas as pd
from tqdm import tqdm
from uuid import uuid4
@snakers4
snakers4 / calculate_knn.py
Created Aug 26, 2018
Use faiss to calculate a KNN graph on data
View calculate_knn.py
import gc
import tqdm
import faiss
import bcolz
import os,sys
import numpy as np
from tqdm import tqdm
# open the stored bcolz array
# note that these vectors have to be 280 dimensional
@snakers4
snakers4 / Loss.py
Created Jul 21, 2018
Multi class classification focal loss
View Loss.py
import torch
import torch.nn as nn
import torch.nn.functional as F
# Focal loss implementation inspired by
# https://github.com/c0nn3r/RetinaNet/blob/master/focal_loss.py
# https://github.com/doiken23/pytorch_toolbox/blob/master/focalloss2d.py
class MultiClassBCELoss(nn.Module):
def __init__(self,
use_weight_mask=False,
@snakers4
snakers4 / train.sh
Created Jul 7, 2018
VAE explanation bits
View train.sh
python3 train.py \
--epochs 30 --batch-size 512 --seed 42 \
--model_type fc_conv --dataset_type fmnist --latent_space_size 10 \
--do_augs False \
--lr 1e-3 --m1 40 --m2 50 \
--optimizer adam \
--do_running_mean False --img_loss_weight 1.0 --kl_loss_weight 1.0 \
--image_loss_type bce --ssim_window_size 5 \
--print-freq 10 \
--lognumber fmnist_fc_conv_l10_rebalance_no_norm \
@snakers4
snakers4 / Dockerfile
Created Jul 3, 2018
My PyTorch 0.4 Dockerfile
View Dockerfile
# add 7z tar and zip archivers
FROM nvidia/cuda:9.0-cudnn7-devel
# https://docs.docker.com/engine/examples/running_ssh_service/
RUN apt-get update && apt-get install -y openssh-server
RUN mkdir /var/run/sshd
RUN echo 'root:Ubuntu@41' | chpasswd
RUN sed -i 's/PermitRootLogin prohibit-password/PermitRootLogin yes/' /etc/ssh/sshd_config
RUN sed -i 's/#PasswordAuthentication yes/PasswordAuthentication no/' /etc/ssh/sshd_config
RUN mkdir ~/.ssh/
You can’t perform that action at this time.