search

KamitaniLab / bdpy

Python package for brain decoding analysis (BrainDecoderToolbox2 data format, machine learning analysis, functional MRI)
MIT License
33 stars 22 forks source link

Feature inversion pipeline for modular iCNN construction #81

Closed ganow closed 1 month ago

ganow commented 8 months ago

77

TODO:

github-actions[bot] commented 8 months ago

Coverage

Coverage Report
FileStmtsMissCoverMissing
bdpy/bdata
   bdata.py40032020%73–79, 86, 90, 95, 99, 104, 109, 113, 118, 122, 132–134, 155–172, 190, 206–207, 232–248, 252–262, 276–277, 293, 310, 314, 318–356, 388–477, 508, 539, 547–551, 560, 577–584, 589–598, 604–614, 618, 622–625, 628, 632–653, 656–665, 668–677, 683–691, 696–729, 735–744, 750–757, 761–767, 771–799, 803–824, 828–862, 866–868, 872–874, 878–887
   featureselector.py645711%43–47, 52–93, 98–124
   metadata.py674828%21–30, 34, 38, 42, 46, 50, 54, 74–115, 135–144, 149, 154
   utils.py1131039%44–110, 127–173, 201, 217–253, 258, 263
bdpy/dataform
   datastore.py1078521%59–75, 90–93, 97–98, 102–113, 116–119, 122–127, 131–132, 137–158, 190–197, 222–259, 262–265
   features.py29823322%29–32, 41–47, 74–103, 107, 111, 115, 119, 137–163, 168–197, 213–238, 241–269, 272–275, 278–289, 305–319, 323, 327, 331, 335, 339, 343, 347, 351, 355, 359, 364–394, 398–418, 422–462, 465, 470–477, 491–493, 496–499, 502–505, 508–512, 515–516, 536–549
   kvs.py1401400%4–272
   pd.py9544%25–27, 43–44
   sparse.py675124%19–29, 35–46, 52–58, 65–74, 78, 81–87, 90–93, 96–98, 101–126
   utils.py12120%3–18
bdpy/dataset
   utils.py45450%3–98
bdpy/distcomp
   distcomp.py927815%20–29, 32–49, 52–70, 73–93, 96–107, 111, 114–117, 121–127
bdpy/dl
   caffe.py60600%4–129
bdpy/dl/torch
   base.py432444%31–41, 48, 54, 60, 63, 73–83, 90, 96, 102, 105
   dataset.py74740%1–195
   models.py33329811%28–84, 114–140, 148–169, 175–238, 249–253, 259–279, 288–292, 297–316, 327–331, 340–405, 427–431, 442–494, 515–517, 528–587, 611–614, 625–684, 708–711, 722–771, 790–793, 804–853, 872–875
   torch.py1219224%43–60, 63, 80–95, 105–111, 116, 123, 126, 131, 138, 141, 188–225, 228, 231–243, 246–281
bdpy/dl/torch/domain
   core.py462057%47, 63, 74, 77, 89–90, 99, 102, 137–138, 141–143, 146–148, 164–165, 168, 173
   feature_domain.py241346%14, 18, 27–38, 41, 44
   image_domain.py643742%26, 31, 36, 41, 84–108, 111, 114, 121, 124, 137, 140–150, 179, 211–216, 223, 229
bdpy/evals
   metrics.py95878%12–34, 40–73, 82–112, 118–159, 172–179
bdpy/feature
   feature.py30287%33–74
bdpy/fig
   __init__.py440%6–9
   draw_group_image_set.py88880%3–182
   fig.py88880%16–164
   makeplots.py3363360%1–729
   tile_images.py59590%1–193
bdpy/ml
   crossvalidation.py59548%34–61, 104–128, 138, 164–196
   ensemble.py13931%33–46
   learning.py30926514%43–44, 48, 52, 59, 91–104, 109–125, 128, 158–170, 184–209, 260–284, 290–433, 436–461, 465–504, 507–508, 524–536, 541–613
   model.py14012014%29–39, 54–70, 86–144, 156–169, 184–222, 225, 230–250, 254–258, 271–285
   regress.py11827%29–38
   searchlight.py161319%32–51
bdpy/mri
   fmriprep.py4974529%25–34, 38, 44–62, 65–75, 78–89, 92–160, 163–194, 230–360, 367–380, 384, 388–390, 394, 398–400, 410–434, 437–454, 457–464, 471–472, 475–491, 494, 498, 502–815, 819–831, 842–862, 866
   glm.py403610%46–95
   image.py241921%29–54
   load_epi.py281836%36–50, 56–63, 82–88
   load_mri.py191616%16–36
   roi.py2482346%37–100, 122–148, 165–235, 241–314, 320–387, 399–466, 473–499
   spm.py15813912%26–155, 162–166, 170, 174–179, 183–300
bdpy/opendata
   __init__.py110%1
   openneuro.py2102100%1–329
bdpy/pipeline
   config.py362919%15–64
bdpy/preproc
   interface.py524415%31–40, 60–69, 96–105, 111–123, 148–157, 208–217
   preprocessor.py12910717%35, 43–65, 74–78, 85–97, 104–132, 138–189, 196–227, 234–239
   select_top.py231822%35–60
   util.py6267%14, 22
bdpy/recon
   utils.py55550%4–146
bdpy/recon/torch
   icnn.py1611610%15–478
bdpy/recon/torch/modules
   critic.py442055%19, 36, 58, 64–65, 68, 71, 96–110, 132, 157, 184–185
   encoder.py291162%29, 44, 63, 66, 104–109, 124–125, 175
   generator.py723650%15–22, 28–38, 47, 52, 68, 83, 90, 93, 96, 122–124, 128, 143, 186–189, 193, 208, 247–248, 252, 256, 306–309
   latent.py341362%16, 21, 32, 42, 49, 52, 55, 80–83, 87, 97
bdpy/recon/torch/task
   inversion.py834941%19, 32, 37, 42, 47, 54, 59, 64, 69, 86–89, 92–95, 98, 101–102, 159–167, 185–214, 218–220
bdpy/stats
   corr.py433812%29–77, 96–112
bdpy/task
   callback.py714438%15–20, 53, 102–116, 154–156, 161, 166, 212–218, 233–247, 264–265
   core.py16475%41, 45, 50, 60
bdpy/util
   info.py473623%19–79
   math.py131023%23–38
   utils.py362628%48–66, 93–96, 116–121, 137–145
TOTAL5644478215% 

Tests Skipped Failures Errors Time
1 0 :zzz: 0 :x: 1 :fire: 7.334s :stopwatch:
ShuntaroAoki commented 8 months ago

@ganow ありがとうございます。 現状の Base クラスの実装では、__call__() を abstractmethod として具象クラス側で実装するよう強制していますが、この部分をそれぞれの基底クラスで (各々の役割に応じた名前の) パブリックメソッドとして定義しておき、__call__() でそれを呼ぶ、とするのはどう思われますか? 以下のようなイメージです。

class BaseGenerator(ABC):
    """Generator module."""

    @abstractmethod
    def reset_states(self) -> None:
        """Reset the state of the generator."""
        pass

    @abstractmethod
    def parameters(self, recurse: bool = True) -> Iterator[nn.Parameter]:
        """Return the parameters of the generator."""
        pass

    @abstractmethod
    def generate(self, latent: torch.Tensor) -> torch.Tensor:
        """Forward pass through the generator network.

        Parameters
        ----------
        latent : torch.Tensor
            Latent vector.

        Returns
        -------
        torch.Tensor
            Generated image. The generated images must be in the range [0, 1].
        """

    def __call__(self, latent: torch.Tensor) -> torch.Tensor:
        """Call self.generate().

        Parameters
        ----------
        latent : torch.Tensor
            Latent vector.

        Returns
        -------
        torch.Tensor
            Generated image. The generated images must be in the range [0, 1].
        """
        return self.generate(latent)

意図は、メソッドに名前をあたえることでそれぞれのメソッドの役割を明確化することです。あとこれはほとんど私情ですが、親クラスのプライベートメソッドをオーバーライドすることに気持ち悪さがあります。

ganow commented 8 months ago

なるほど。実装するメソッド名はモジュールごとに変えずにforwardで共通化するのはどう思いますか？nn.moduleと透過的に扱うことを意識したAPIです

ShuntaroAoki commented 8 months ago

なるほど。実装するメソッド名はモジュールごとに変えずにforwardで共通化するのはどう思いますか？nn.moduleと透過的に扱うことを意識したAPIです

BaseGenerator 等が必ずしも nn.module の子クラスではないので、かえって confusing かなと感じます。またモジュールクラスのインスタンスをユーザが直接扱うことは基本的にない (ユーザは外で定義するだけ) ため、呼び出す側の使い易さよりも、定義時のリーダビリティを優先したほうがいいかなと思います。なので若干煩雑ですが、クラスごとに固有の、役割を明示した名前のほうがいいかなと今は思います (とはいえ、クラス名からほとんど自明といえば自明ですが)。

myaokai commented 6 months ago

sample codeを作成してのコメント