プログラムエラーがわかりません

Question

深層学習における学習過程の可視化を中間層出力からやってみるを実行したのですが、実際に学習させるところで

acc, val_acc, loss, val_loss = train(model, X_train, y_train, epoch=100, batch=100)

UMAP(dens_frac=0.0, dens_lambda=0.0, n_neighbors=40, verbose=1)
Construct fuzzy simplicial set
Tue Jul 27 12:35:22 2021 Finding Nearest Neighbors
Tue Jul 27 12:35:22 2021 Building RP forest with 17 trees
Tue Jul 27 12:35:23 2021 NN descent for 16 iterations
	 1  /  16
	 2  /  16
	 3  /  16
	Stopping threshold met -- exiting after 3 iterations
Tue Jul 27 12:35:28 2021 Finished Nearest Neighbor Search
Tue Jul 27 12:35:28 2021 Construct embedding
	completed  0  /  200 epochs
	completed  20  /  200 epochs
	completed  40  /  200 epochs
	completed  60  /  200 epochs
	completed  80  /  200 epochs
	completed  100  /  200 epochs
	completed  120  /  200 epochs
	completed  140  /  200 epochs
	completed  160  /  200 epochs
	completed  180  /  200 epochs
Tue Jul 27 12:35:49 2021 Finished embedding
---------------------------------------------------------------------------
FileNotFoundError                         Traceback (most recent call last)
/var/folders/72/fl6l29ds4xvbv0zcgdzj4wc00000gn/T/ipykernel_3975/2994000927.py in <module>
----> 1 acc, val_acc, loss, val_loss = train(model, X_train, y_train, epoch=100, batch=100)

/var/folders/72/fl6l29ds4xvbv0zcgdzj4wc00000gn/T/ipykernel_3975/3045578850.py in train(model, X_train, y_train, epoch, batch)
     17         plt.colorbar()
     18         plt.title('Epoch  ' + str(i))
---> 19         plt.savefig('./fig/epoch_' + '{0:03d}'.format(i) + '.png', bbox_inches='tight')
     20         plt.close()
     21         # 1epochだけ学習

/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/matplotlib/pyplot.py in savefig(*args, **kwargs)
    964 def savefig(*args, **kwargs):
    965     fig = gcf()
--> 966     res = fig.savefig(*args, **kwargs)
    967     fig.canvas.draw_idle()   # need this if 'transparent=True' to reset colors
    968     return res

/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/matplotlib/figure.py in savefig(self, fname, transparent, **kwargs)
   3003                 patch.set_edgecolor('none')
   3004 
-> 3005         self.canvas.print_figure(fname, **kwargs)
   3006 
   3007         if transparent:

/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/matplotlib/backend_bases.py in print_figure(self, filename, dpi, facecolor, edgecolor, orientation, format, bbox_inches, pad_inches, bbox_extra_artists, backend, **kwargs)
   2259                         orientation=orientation,
   2260                         bbox_inches_restore=_bbox_inches_restore,
-> 2261                         **kwargs)
   2262             finally:
   2263                 if bbox_inches and restore_bbox:

/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/matplotlib/backend_bases.py in wrapper(*args, **kwargs)
   1667             kwargs.pop(arg)
   1668 
-> 1669         return func(*args, **kwargs)
   1670 
   1671     return wrapper

/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/matplotlib/backends/backend_agg.py in print_png(self, filename_or_obj, metadata, pil_kwargs, *args)
    509         mpl.image.imsave(
    510             filename_or_obj, self.buffer_rgba(), format="png", origin="upper",
--> 511             dpi=self.figure.dpi, metadata=metadata, pil_kwargs=pil_kwargs)
    512 
    513     def print_to_buffer(self):

/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/matplotlib/image.py in imsave(fname, arr, vmin, vmax, cmap, format, origin, dpi, metadata, pil_kwargs)
   1614         pil_kwargs.setdefault("format", format)
   1615         pil_kwargs.setdefault("dpi", (dpi, dpi))
-> 1616         image.save(fname, **pil_kwargs)
   1617 
   1618 

/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/PIL/Image.py in save(self, fp, format, **params)
   2167                 fp = builtins.open(filename, "r+b")
   2168             else:
-> 2169                 fp = builtins.open(filename, "w+b")
   2170 
   2171         try:

FileNotFoundError: [Errno 2] No such file or directory: './fig/epoch_000.png'

となってしまいます。
どうしたらいいでしょうか？

Comments

[jbpb0]

> RuntimeError: You must compile a model before training/testing. Use `model.compile(optimizer, loss)`.

書いてある通り、「model.compile()」が必要です

[jbpb0]

https://teratail.com/questions/351312
で参考にした
https://qiita.com/DataSkywalker/items/6f5a8fafee82c195fa0b
のコードには、「model.compile(...」がありますよね

それを参考にして、今回のコードにも「model.compile()」を追加してみてください

Answer 1

plt.savefig('./fig/epoch_' + ...

ここでエラーになっているということは保存だと思われますので、カレントディレクトリに「fig」というディレクトリが無いのが原因ではないかと思います。

うまく解決できないようであれば、この 「fig」を外して、

plt.savefig('./epoch_' + ...

こうしてみると、カレントディレクトリに大量のファイルができているのがわかるはずです。

Comments

[退会済みユーザー]

UMAP(dens_frac=0.0, dens_lambda=0.0, n_neighbors=40, verbose=1)
Construct fuzzy simplicial set
Tue Jul 27 15:18:54 2021 Finding Nearest Neighbors
Tue Jul 27 15:18:54 2021 Building RP forest with 17 trees
Tue Jul 27 15:18:55 2021 NN descent for 16 iterations
1 / 16
2 / 16
3 / 16
Stopping threshold met -- exiting after 3 iterations
Tue Jul 27 15:19:06 2021 Finished Nearest Neighbor Search
Tue Jul 27 15:19:09 2021 Construct embedding
completed 0 / 200 epochs
completed 20 / 200 epochs
completed 40 / 200 epochs
completed 60 / 200 epochs
completed 80 / 200 epochs
completed 100 / 200 epochs
completed 120 / 200 epochs
completed 140 / 200 epochs
completed 160 / 200 epochs
completed 180 / 200 epochs
Tue Jul 27 15:19:29 2021 Finished embedding
---------------------------------------------------------------------------
RuntimeError Traceback (most recent call last)
/var/folders/72/fl6l29ds4xvbv0zcgdzj4wc00000gn/T/ipykernel_4512/2994000927.py in <module>
----> 1 acc, val_acc, loss, val_loss = train(model, X_train, y_train, epoch=100, batch=100)

/var/folders/72/fl6l29ds4xvbv0zcgdzj4wc00000gn/T/ipykernel_4512/3760558164.py in train(model, X_train, y_train, epoch, batch)
20 plt.close()
21 # 1epochだけ学習
---> 22 history = model.fit(X_train, y_train, epochs=1, batch_size=batch, validation_split=0.2, verbose=1)
23
24 # Lossなど保存

/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/keras/engine/training.py in fit(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_freq, max_queue_size, workers, use_multiprocessing, **kwargs)
1152 sample_weight=sample_weight,
1153 class_weight=class_weight,
-> 1154 batch_size=batch_size)
1155
1156 # Prepare validation data.

/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/keras/engine/training.py in _standardize_user_data(self, x, y, sample_weight, class_weight, check_array_lengths, batch_size)
506 if y is not None:
507 if not self.optimizer:
--> 508 raise RuntimeError('You must compile a model before '
509 'training/testing. '
510 'Use `model.compile(optimizer, loss)`.')

RuntimeError: You must compile a model before training/testing. Use `model.compile(optimizer, loss)`.

外して実行したところ上のようなエラーに変わりました。
と言うようなエラーが出てしまいました。

[TakaiY]

質問の内容はもう変っていますね。

読めばわかるように、modelをtraining/testingする前にcompileする必要があるとのこと。
これについては、参照している記事の作者に問い合わせるのがいいのでは？

Answer 2

カレントディレクトリにfig/epoch_000.pngというファイルがありません。

import os
print(os.getcwd())

で表示されるディレクトリにfigというサブディレクトリを作成し、その中にepoch_000.pngを置いてください。

Comments

[退会済みユーザー]

「ディレクトリにfigというサブディレクトリを作成し、その中にepoch_000.pngを置いてください。」
figというディレクトリは表示されたディレクトリに作成したのですが、epoch_000.pngの画像はどうしたらいいのでしょうか？

[退会済みユーザー]

分からなかったので、emacs で　epoch_000.png　というのを作成して実行したのですが、以下のようなエラーが出てきました。
何がいけないのでしょうか？

UMAP(dens_frac=0.0, dens_lambda=0.0, n_neighbors=40, verbose=1)
Construct fuzzy simplicial set
Tue Jul 27 15:40:51 2021 Finding Nearest Neighbors
Tue Jul 27 15:40:51 2021 Building RP forest with 17 trees
Tue Jul 27 15:40:52 2021 NN descent for 16 iterations
1 / 16
2 / 16
3 / 16
Stopping threshold met -- exiting after 3 iterations
Tue Jul 27 15:41:04 2021 Finished Nearest Neighbor Search
Tue Jul 27 15:41:07 2021 Construct embedding
completed 0 / 200 epochs
completed 20 / 200 epochs
completed 40 / 200 epochs
completed 60 / 200 epochs
completed 80 / 200 epochs
completed 100 / 200 epochs
completed 120 / 200 epochs
completed 140 / 200 epochs
completed 160 / 200 epochs
completed 180 / 200 epochs
Tue Jul 27 15:41:27 2021 Finished embedding
---------------------------------------------------------------------------
RuntimeError Traceback (most recent call last)
/var/folders/72/fl6l29ds4xvbv0zcgdzj4wc00000gn/T/ipykernel_4639/2994000927.py in <module>
----> 1 acc, val_acc, loss, val_loss = train(model, X_train, y_train, epoch=100, batch=100)

/var/folders/72/fl6l29ds4xvbv0zcgdzj4wc00000gn/T/ipykernel_4639/3045578850.py in train(model, X_train, y_train, epoch, batch)
20 plt.close()
21 # 1epochだけ学習
---> 22 history = model.fit(X_train, y_train, epochs=1, batch_size=batch, validation_split=0.2, verbose=1)
23
24 # Lossなど保存

/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/keras/engine/training.py in fit(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_freq, max_queue_size, workers, use_multiprocessing, **kwargs)
1152 sample_weight=sample_weight,
1153 class_weight=class_weight,
-> 1154 batch_size=batch_size)
1155
1156 # Prepare validation data.

/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/keras/engine/training.py in _standardize_user_data(self, x, y, sample_weight, class_weight, check_array_lengths, batch_size)
506 if y is not None:
507 if not self.optimizer:
--> 508 raise RuntimeError('You must compile a model before '
509 'training/testing. '
510 'Use `model.compile(optimizer, loss)`.')

RuntimeError: You must compile a model before training/testing. Use `model.compile(optimizer, loss)`.