Answeror

FROM dmlc/mxnet:cuda
MAINTAINER answeror <answeror@gmail.com>

ENV LD_LIBRARY_PATH /usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64

RUN cd /mxnet && git pull origin master && git submodule update
ADD src/io/image_aug_default.cc /mxnet/src/io/image_aug_default.cc
RUN cd /mxnet && make -j8 ADD_LDFLAGS=-L/usr/local/cuda/lib64/stubs
ADD example/image-classification/symbol_resnet.py /mxnet/example/image-classification/symbol_resnet.py
ADD example/image-classification/train_cifar10_resnet.py /mxnet/example/image-classification/train_cifar10_resnet.py

Answeror

from __future__ import absolute_import
import numpy as np


class Accumulators(object):

    _accumulator_library = {}

    def __init__(self):
        self.accumulator_library = {}

Answeror

{
    "string": "{{ string }}",
    "int": {{ int }}
}

Answeror

<?xml version='1.0' encoding='UTF-8'?>
<xsl:stylesheet version='1.0' xmlns:xsl='http://www.w3.org/1999/XSL/Transform'>
    <xsl:output method='xml' encoding='UTF-8'/>
    <xsl:template match='/'>
        <feed xml:lang='en-US' xmlns='http://www.w3.org/2005/Atom'>
            <id>v2ex.com/member/{{ id }}</id>
            <link rel='alternate' type='text/html' href='http://v2ex.com/member/{{ id }}'/>
            <title>V2EX user replies of {{ id }}</title>
            <xsl:for-each select='.//div[@class="dock_area"]'>
                <entry>

Answeror

<?xml version='1.0' encoding='UTF-8'?>
<xsl:stylesheet
    version='1.0'
    xmlns:xsl='http://www.w3.org/1999/XSL/Transform'
>
    <xsl:output method='xml' encoding='UTF-8'/>
    <xsl:template match='/'>
        <feed xml:lang='en-US' xmlns='http://www.w3.org/2005/Atom'>
            <id>nyc</id>
            <link rel='alternate' type='text/html' href='http://www.nyc.gov/html/dot/html/home/home.shtml'/>

Answeror

基于全页面md5 hash的更新通知

这个gist可以用来订阅某个页面, 当该页面全文的md5 hash改变时, 输出的Atom feed会相应地更新. 例如订阅萝莉盟: http://rss.moe/api/source/f742582061a9b028987e?uri=http://www.loli.com/

你可以使用torabot的feed模块或者你喜欢的阅读器订阅上述链接.

Answeror

<feed xml:lang='zh-CN' xmlns='http://www.w3.org/2005/Atom'>
    <id>v2ex.com/t/{{ id }}</id>
    <link rel='alternate' type='text/html' href='http://v2ex.com/t/{{ id }}'/>
    <title>{{ id }}</title>
    {% for r in replies %}
    {% if r.member.username != 'answeror' %}
    <entry>
        <id>{{ r.id }}</id>
        <published>{{ datetime.fromtimestamp(r.created).strftime('%Y-%m-%dT%H:%M:%SZ') }}</published>
        <updated>{{ datetime.fromtimestamp(r.last_modified).strftime('%Y-%m-%dT%H:%M:%SZ') }}</updated>

Answeror

{
    "@jinja2": {
        "template": {"text<": "replies.xml"},
        "kargs": {
            "id": "{{ id }}",
            "replies": {
                "@json_decode": {
                    "@base64_decode": [{
                        "[body]": {"@request": "http://www.v2ex.com/api/replies/show.json?topic_id={{ id }}"}
                    }]

Answeror

Bangumi评论订阅

例子: http://torabot.com/api/source/f4b337c06bf746969244.atom?path=/group/topic/32268

把/group/topic/32268换成你想要订阅的地址, 例如/subject/topic/4369.

Answeror

Bangumi私信订阅

这个gist由torabot自动抓取并生成Atom feed.

使用方法

注意: 公开你的cookie是一件危险的事情. 别人可能利用它来滥用你的Bangumi账户! Torabot的作者保证不会使用你的cookie做这个gist功能以外的事情. 如果你不信任torabot, 请忽略这个gist.

例子: http://torabot.com/api/source/34fd6a535090ca85a2fc.atom?chii_auth=<chii_auth_in_your_cookie>