SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1907.09595 · 2019

MixConv: Mixed Depthwise Convolutional Kernels

arXiv · PDF · Open in the Atlas

Code that ran

2 of 3 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 0 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/1907.09595.json · badge: /api/ran/1907.09595.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
neeraj-j/MixNet community (classifier label)
classifier basis not recorded on the link
2 ran · 1 no recorded run
FunctionStatusWhere it lives
mixnet_m Ran
ran on a synthesized input
neeraj-j/MixNet/mxnet.py
code served (permissive licence) · get_code("67513021328c7f77")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import collections
import re

GlobalParams = collections.namedtuple('GlobalParams', ['batch_norm_momentum', 'batch_norm_epsilon', 'dropout_rate', 'data_format', 'num_classes', 'depth_multiplier', 'depth_divisor', 'min_depth', 'use_keras', 'stem_size', 'feature_size'])

BlockArgs = collections.namedtuple('BlockArgs', ['dw_ksize', 'expand_ksize', 'project_ksize', 'num_repeat', 'input_filters', 'output_filters', 'expand_ratio', 'id_skip', 'strides', 'se_ratio', 'swish', 'dilated'])

class MixnetDecoder(object):
    """A class of Mixnet decoder to get model configuration."""

    def _decode_block_string(self, block_string):
        """Gets a mixnet block through a string notation of arguments."""
        assert isinstance(block_string, str)
        ops = block_string.split('_')
        options = {}
        for op in ops:
            splits = re.split('(\\d.*)', op)
            if len(splits) >= 2:
                key, value = splits[:2]
                options[key] = value
        if 's' not in options or len(options['s']) != 2:
            raise ValueError('Strides options should be a pair of integers.')

        def _parse_ksize(ss):
            return [int(k) for k in ss.split('.')]
        return BlockArgs(expand_ksize=_parse_ksize(options['a']), dw_ksize=_parse_ksize(options['k']), project_ksize=_parse_ksize(options['p']), num_repeat=int(options['r']), input_filters=int(options['i']), output_filters=int(options['o']), expand_ratio=int(options['e']), id_skip='noskip' not in block_string, se_ratio=float(options['se']) if 'se' in options else None, strides=[int(options['s'][0]), int(options['s'][1])], swish='sw' in block_string, dilated='dilated' in block_string)

    def _encode_block_string(self, block):
        """Encodes a Mixnet block to a string."""

        def _encode_ksize(arr):
            return '.'.join([str(k) for k in arr])
        args = ['r%d' % block.num_repeat, 'k%s' % _encode_ksize(block.dw_ksize), 'a%s' % _encode_ksize(block.expand_ksize), 'p%s' % _encode_ksize(block.project_ksize), 's%d%d' % (block.strides[0], block.strides[1]), 'e%s' % block.expand_ratio, 'i%d' % block.input_filters, 'o%d' % block.output_filters]
        if block.se_ratio is not None and block.se_ratio > 0 and (block.se_ratio <= 1):
            args.append('se%s' % block.se_ratio)
        if block.id_skip is False:
            args.append('noskip')
        if block.swish:
            args.append('sw')
        if block.dilated:
            args.append('dilated')
        return '_'.join(args)

    def decode(self, string_list):
        """Decodes a list of string notations to specify blocks inside the network."""
        assert isinstance(string_list, list)
        blocks_args = []
        for block_string in string_list:
            blocks_args.append(self._decode_block_string(block_string))
        return blocks_args

    def encode(self, blocks_args):
        """Encodes a list of Mixnet Blocks to a list of strings."""
        block_strings = []
        for block in blocks_args:
            block_strings.append(self._encode_block_string(block))
        return block_strings

ARGS = {
    'depth_multiplier': 1.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
round_filters Ran
ran, honoured the contract we drafted
neeraj-j/MixNet/mxnet.py
pointer only (licence: MIT); the stored copy get_code serves, from nsarang/MnasNet, is pointer only (licence: NONE) · get_code("285e4d1d9a3faeba")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from nsarang/MnasNet, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

mixnet_s Not yet run
unverified (harvested, no recorded run)
neeraj-j/MixNet/mxnet.py
code served (permissive licence) · get_code("850e6e44d945a586")
How it ran

no invocation recorded.

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 later) and 2 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
round_filters 3 other papers
2 dated earlier, 1 later
1807.11626 (2018-07)
MnasNet: Platform-Aware Neural Architecture Search for Mobile
1904.02877 (2019-04)
Single-Path NAS: Designing Hardware-Efficient ConvNets in less than 4 Hours
2004.01961 (2020-04)
Neural Architecture Search for Lightweight Non-Local Networks

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Depthwise convolution is becoming increasingly popular in modern efficient ConvNets, but its kernel size is often overlooked. In this paper, we systematically study the impact of different kernel sizes, and observe that combining the benefits of multiple kernel sizes can lead to better accuracy and efficiency. Based on this observation, we propose a new mixed depthwise convolution (MixConv), which naturally mixes up multiple kernel sizes in a single convolution. As a simple drop-in replacement of vanilla depthwise convolution, our MixConv improves the accuracy and efficiency for existing MobileNets on both ImageNet classification and COCO object detection. To demonstrate the effectiveness of MixConv, we integrate it into AutoML search space and develop a new family of models, named as MixNets, which outperform previous mobile models including MobileNetV2 [20] (ImageNet top-1 accuracy +4.2%), ShuffleNetV2 [16] (+3.5%), MnasNet [26] (+1.3%), ProxylessNAS [2] (+2.2%), and FBNet [27] (+2.0%). In particular, our MixNet-L achieves a new state-of-the-art 78.9% ImageNet top-1 accuracy under typical mobile settings (&lt;600M FLOPS). Code is at https://github.com/ tensorflow/tpu/tree/master/models/official/mnasnet/mixnet

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("1907.09595")
get_code_for_paper("1907.09595")
have("1907.09595")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/1907.09595.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/1907.09595.svg)](https://syntology.ai/paper/1907.09595)

Connect an agent — have() is free.