accusleepy 0.8.1__py3-none-any.whl → 0.9.2__py3-none-any.whl

accusleepy/signal_processing.py

@@ -1,3 +1,4 @@
+import logging
 import os
 import warnings
 
@@ -16,6 +17,7 @@ from accusleepy.constants import (
     EMG_COPIES,
     FILENAME_COL,
     LABEL_COL,
+    MIN_EPOCHS_PER_STATE,
     MIN_WINDOW_LEN,
     UPPER_FREQ,
     SPECTROGRAM_UPPER_FREQ,
@@ -25,10 +27,15 @@ from accusleepy.multitaper import spectrogram
 
 # note: scipy is lazily imported
 
+logger = logging.getLogger(__name__)
+
 
 def resample(
-    eeg: np.array, emg: np.array, sampling_rate: int | float, epoch_length: int | float
-) -> (np.array, np.array, float):
+    eeg: np.ndarray,
+    emg: np.ndarray,
+    sampling_rate: int | float,
+    epoch_length: int | float,
+) -> tuple[np.ndarray, np.ndarray, float]:
     """Resample recording so that epochs contain equal numbers of samples
 
     If the number of samples per epoch is not an integer, epoch-level calculations
@@ -62,8 +69,11 @@ def resample(
 
 
 def standardize_signal_length(
-    eeg: np.array, emg: np.array, sampling_rate: int | float, epoch_length: int | float
-) -> (np.array, np.array):
+    eeg: np.ndarray,
+    emg: np.ndarray,
+    sampling_rate: int | float,
+    epoch_length: int | float,
+) -> tuple[np.ndarray, np.ndarray]:
     """Truncate or pad EEG/EMG signals to have an integer number of epochs
 
     :param eeg: EEG signal
@@ -93,8 +103,11 @@ def standardize_signal_length(
 
 
 def resample_and_standardize(
-    eeg: np.array, emg: np.array, sampling_rate: int | float, epoch_length: int | float
-) -> (np.array, np.array, float):
+    eeg: np.ndarray,
+    emg: np.ndarray,
+    sampling_rate: int | float,
+    epoch_length: int | float,
+) -> tuple[np.ndarray, np.ndarray, float]:
     """Preprocess EEG and EMG signals
 
     Adjust the length and sampling rate of the EEG and EMG signals so that
@@ -117,12 +130,12 @@ def resample_and_standardize(
 
 
 def create_spectrogram(
-    eeg: np.array,
+    eeg: np.ndarray,
     sampling_rate: int | float,
     epoch_length: int | float,
-    time_bandwidth=2,
-    n_tapers=3,
-) -> (np.array, np.array):
+    time_bandwidth: int = 2,
+    n_tapers: int = 3,
+) -> tuple[np.ndarray, np.ndarray]:
     """Create an EEG spectrogram image
 
     :param eeg: EEG signal
@@ -168,11 +181,11 @@ def create_spectrogram(
 
 
 def get_emg_power(
-    emg: np.array,
+    emg: np.ndarray,
     sampling_rate: int | float,
     epoch_length: int | float,
     emg_filter: EMGFilter,
-) -> np.array:
+) -> np.ndarray:
     """Calculate EMG power for each epoch
 
     This applies a 20-50 Hz bandpass filter to the EMG,  calculates the RMS
@@ -202,17 +215,18 @@ def get_emg_power(
         [round(len(emg) / samples_per_epoch), samples_per_epoch],
     )
     rms = np.sqrt(np.mean(np.power(reshaped, 2), axis=1))
-
-    return np.log(rms)
+    log_rms = np.log(rms)
+    log_rms[np.isinf(log_rms)] = 0
+    return log_rms
 
 
 def create_eeg_emg_image(
-    eeg: np.array,
-    emg: np.array,
+    eeg: np.ndarray,
+    emg: np.ndarray,
     sampling_rate: int | float,
     epoch_length: int | float,
     emg_filter: EMGFilter,
-) -> np.array:
+) -> np.ndarray:
     """Stack EEG spectrogram and EMG power into an image
 
     This assumes that each epoch contains an integer number of samples and
@@ -247,8 +261,8 @@ def create_eeg_emg_image(
 
 
 def get_mixture_values(
-    img: np.array, labels: np.array, brain_state_set: BrainStateSet
-) -> (np.array, np.array):
+    img: np.ndarray, labels: np.ndarray, brain_state_set: BrainStateSet
+) -> tuple[np.ndarray, np.ndarray]:
     """Compute weighted feature means and SDs for mixture z-scoring
 
     The outputs of this function can be used to standardize features
@@ -289,12 +303,12 @@ def get_mixture_values(
 
 
 def mixture_z_score_img(
-    img: np.array,
+    img: np.ndarray,
     brain_state_set: BrainStateSet,
-    labels: np.array = None,
-    mixture_means: np.array = None,
-    mixture_sds: np.array = None,
-) -> np.array:
+    labels: np.ndarray | None = None,
+    mixture_means: np.ndarray | None = None,
+    mixture_sds: np.ndarray | None = None,
+) -> tuple[np.ndarray, bool]:
     """Perform mixture z-scoring on a combined EEG+EMG image
 
     If brain state labels are provided, they will be used to calculate
@@ -307,10 +321,10 @@ def mixture_z_score_img(
     :param labels: labels, in "class" format
     :param mixture_means: mixture means
     :param mixture_sds: mixture standard deviations
-    :return:
+    :return: tuple of (z-scored image, whether zero-variance features were detected)
     """
     if labels is None and (mixture_means is None or mixture_sds is None):
-        raise Exception("must provide either labels or mixture means+SDs")
+        raise ValueError("must provide either labels or mixture means+SDs")
     if labels is not None and ((mixture_means is not None) ^ (mixture_sds is not None)):
         warnings.warn("labels were given, mixture means / SDs will be ignored")
 
@@ -319,14 +333,27 @@ def mixture_z_score_img(
             img=img, labels=labels, brain_state_set=brain_state_set
         )
 
+    # replace zero SDs with epsilon to avoid division by zero
+    # This can occur when a feature has no variance (e.g., no EMG signal)
+    zero_sd_mask = mixture_sds == 0
+    had_zero_variance = np.any(zero_sd_mask)
+    if had_zero_variance:
+        n_zero = np.sum(zero_sd_mask)
+        logger.warning(
+            "%s feature(s) have zero variance and will be mapped to neutral values",
+            n_zero,
+        )
+        mixture_sds = mixture_sds.copy()
+        mixture_sds[zero_sd_mask] = 1e-10
+
     img = ((img.T - mixture_means) / mixture_sds).T
     img = (img + ABS_MAX_Z_SCORE) / (2 * ABS_MAX_Z_SCORE)
     img = np.clip(img, 0, 1)
 
-    return img
+    return img, had_zero_variance
 
 
-def format_img(img: np.array, epochs_per_img: int, add_padding: bool) -> np.array:
+def format_img(img: np.ndarray, epochs_per_img: int, add_padding: bool) -> np.ndarray:
     """Adjust the format of an EEG+EMG image
 
     This function converts the values in a combined EEG+EMG image to uint8.
@@ -369,9 +396,18 @@ def create_training_images(
     model_type: str,
     calibration_fraction: float,
     emg_filter: EMGFilter,
-) -> list[int]:
-    """Create training dataset
-
+) -> tuple[list[int], np.ndarray, bool]:
+    """Create training dataset and calculate class balance
+
+    This function creates images that can be used to train the
+    SSANN model, as well as files that describe the training data
+    (optionally split into training and calibration sets).
+    It returns a list of recordings that could not be processed,
+    the class balance of the usable training data, and a flag if any
+    recordings had features with 0 variance.
+
+    For each epoch, the model expects an image containing the
+    EEG spectrogram and EMG power for several surrounding epochs.
     By default, the current epoch is located in the central column
     of pixels in each image. For real-time scoring applications,
     the current epoch is at the right edge of each image.
@@ -384,8 +420,7 @@ def create_training_images(
     :param model_type: default or real-time
     :param calibration_fraction: fraction of training data to use for calibration
     :param emg_filter: EMG filter parameters
-    :return: list of the names of any recordings that could not
-            be used to create training images.
+    :return: tuple of (failed recording names, training class balance, had zero-variance)
     """
     # recordings that had to be skipped
     failed_recordings = list()
@@ -393,9 +428,36 @@ def create_training_images(
     filenames = list()
     # all valid labels from all valid recordings
     all_labels = list()
+    # track if any recording had zero-variance features
+    any_zero_variance = False
     # try to load each recording and create training images
     for i in trange(len(recordings)):
         recording = recordings[i]
+        try:
+            labels, _ = load_labels(recording.label_file)
+        except Exception:
+            logger.exception("Could not load labels for recording %s", recording.name)
+            failed_recordings.append(recording.name)
+            continue
+
+        # Check that each scored brain state has sufficient observations
+        # Ideally, we could use mixture means/SDs from another recording...
+        insufficient_labels = False
+        for brain_state in brain_state_set.brain_states:
+            if brain_state.is_scored:
+                count = np.sum(labels == brain_state.digit)
+                if count < MIN_EPOCHS_PER_STATE:
+                    logger.warning(
+                        "Recording %s can't be used: insufficient labels for class '%s'",
+                        recording.name,
+                        brain_state.name,
+                    )
+                    failed_recordings.append(recording.name)
+                    insufficient_labels = True
+                    break
+        if insufficient_labels:
+            continue
+
         try:
             eeg, emg = load_recording(recording.recording_file)
             sampling_rate = recording.sampling_rate
@@ -406,14 +468,15 @@ def create_training_images(
                 epoch_length=epoch_length,
             )
 
-            labels, _ = load_labels(recording.label_file)
             labels = brain_state_set.convert_digit_to_class(labels)
             img = create_eeg_emg_image(
                 eeg, emg, sampling_rate, epoch_length, emg_filter
             )
-            img = mixture_z_score_img(
+            img, had_zero_variance = mixture_z_score_img(
                 img=img, brain_state_set=brain_state_set, labels=labels
             )
+            if had_zero_variance:
+                any_zero_variance = True
             img = format_img(img=img, epochs_per_img=epochs_per_img, add_padding=True)
 
             # the model type determines which epochs are used in each image
@@ -442,8 +505,10 @@ def create_training_images(
                     all_labels.append(labels[j])
                     Image.fromarray(im).save(os.path.join(output_path, filename))
 
-        except Exception as e:
-            print(e)
+        except Exception:
+            logger.exception(
+                "Failed to create training images for recording %s", recording.name
+            )
             failed_recordings.append(recording.name)
 
     annotations = pd.DataFrame({FILENAME_COL: filenames, LABEL_COL: all_labels})
@@ -460,11 +525,18 @@ def create_training_images(
             os.path.join(output_path, CALIBRATION_ANNOTATION_FILENAME),
             index=False,
         )
+        training_labels = training_set[LABEL_COL].values
     else:
         # annotation file contains info on all training images
         annotations.to_csv(
             os.path.join(output_path, ANNOTATIONS_FILENAME),
             index=False,
         )
+        training_labels = np.array(all_labels)
+
+    # compute class balance from training set
+    class_counts = np.bincount(training_labels, minlength=brain_state_set.n_classes)
+    training_class_balance = class_counts / class_counts.sum()
+    logger.info("Training set class balance: %s", training_class_balance)
 
-    return failed_recordings
+    return failed_recordings, training_class_balance, any_zero_variance

accusleepy/temperature_scaling.py

@@ -1,8 +1,12 @@
+import logging
+
 import numpy as np
 import torch
 from torch import nn, optim
 from torch.nn import functional as F
 
+logger = logging.getLogger(__name__)
+
 
 class ModelWithTemperature(nn.Module):
     """
@@ -72,9 +76,10 @@ class ModelWithTemperature(nn.Module):
         # Calculate NLL and ECE before temperature scaling
         before_temperature_nll = nll_criterion(logits, labels).item()
         before_temperature_ece = ece_criterion(logits, labels).item()
-        print(
-            "Before temperature - NLL: %.3f, ECE: %.3f"
-            % (before_temperature_nll, before_temperature_ece)
+        logger.info(
+            "Before temperature - NLL: %.3f, ECE: %.3f",
+            before_temperature_nll,
+            before_temperature_ece,
         )
 
         # Next: optimize the temperature w.r.t. NLL
@@ -96,16 +101,17 @@ class ModelWithTemperature(nn.Module):
         after_temperature_ece = ece_criterion(
             self.temperature_scale(logits), labels
         ).item()
-        print("Optimal temperature: %.3f" % self.temperature.item())
-        print(
-            "After temperature - NLL: %.3f, ECE: %.3f"
-            % (after_temperature_nll, after_temperature_ece)
+        logger.info("Optimal temperature: %.3f", self.temperature.item())
+        logger.info(
+            "After temperature - NLL: %.3f, ECE: %.3f",
+            after_temperature_nll,
+            after_temperature_ece,
         )
 
         val_acc = round(
             100 * np.mean(labels.cpu().numpy() == predictions.cpu().numpy()), 2
         )
-        print(f"Validation accuracy: {val_acc}%")
+        logger.info("Validation accuracy: %s%%", val_acc)
 
         return self
 

accusleepy/validation.py

@@ -7,8 +7,8 @@ LABEL_LENGTH_ERROR = "label file length does not match recording length"
 
 
 def check_label_validity(
-    labels: np.array,
-    confidence_scores: np.array,
+    labels: np.ndarray,
+    confidence_scores: np.ndarray | None,
     samples_in_recording: int,
     sampling_rate: int | float,
     epoch_length: int | float,
@@ -46,6 +46,71 @@ def check_label_validity(
     return None
 
 
+def validate_and_correct_labels(
+    labels: np.ndarray,
+    confidence_scores: np.ndarray | None,
+    samples_in_recording: int,
+    sampling_rate: int | float,
+    epoch_length: int | float,
+    brain_state_set: BrainStateSet,
+) -> tuple[np.ndarray | None, np.ndarray | None, str | None]:
+    """Validate labels and attempt to correct minor length mismatches.
+
+    If the label array is off by exactly one epoch, it will be padded or
+    truncated and a warning will be returned.
+
+    :param labels: brain state labels
+    :param confidence_scores: optional confidence scores
+    :param samples_in_recording: number of samples in the recording
+    :param sampling_rate: sampling rate in Hz
+    :param epoch_length: epoch length in seconds
+    :param brain_state_set: set of brain state options
+    :return: (labels, confidence_scores, message) - if labels is None,
+        message is an error. Otherwise, any message is a warning
+    """
+    label_error = check_label_validity(
+        labels=labels,
+        confidence_scores=confidence_scores,
+        samples_in_recording=samples_in_recording,
+        sampling_rate=sampling_rate,
+        epoch_length=epoch_length,
+        brain_state_set=brain_state_set,
+    )
+
+    if not label_error:
+        return labels, confidence_scores, None
+
+    # If length is off by one, try to correct it
+    if label_error == LABEL_LENGTH_ERROR:
+        samples_per_epoch = round(sampling_rate * epoch_length)
+        epochs_in_recording = round(samples_in_recording / samples_per_epoch)
+
+        if epochs_in_recording - labels.size == 1:
+            # Pad with one undefined label
+            labels = np.concatenate((labels, np.array([UNDEFINED_LABEL])))
+            if confidence_scores is not None:
+                confidence_scores = np.concatenate((confidence_scores, np.array([0])))
+            return (
+                labels,
+                confidence_scores,
+                "An undefined epoch was added to the label file to correct its length.",
+            )
+
+        if labels.size - epochs_in_recording == 1:
+            # Truncate by one label
+            labels = labels[:-1]
+            if confidence_scores is not None:
+                confidence_scores = confidence_scores[:-1]
+            return (
+                labels,
+                confidence_scores,
+                "The last epoch was removed from the label file to correct its length.",
+            )
+
+    # error could not be fixed
+    return None, None, label_error
+
+
 def check_config_consistency(
     current_brain_states: dict,
     model_brain_states: dict,

{accusleepy-0.8.1.dist-info → accusleepy-0.9.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: accusleepy
-Version: 0.8.1
+Version: 0.9.2
 Summary: Python implementation of AccuSleep
 License: GPL-3.0-only
 Author: Zeke Barger
@@ -80,7 +80,7 @@ please consult the [developer guide](accusleepy/gui/text/dev_guide.md).
 
 ## Changelog
 
-- 0.8.1: Increased test coverage
+- 0.8.1-0.9.2: Improved error handling and code quality
 - 0.8.0: More configurable settings, visual improvements
 - 0.7.1-0.7.3: Bugfixes, code cleanup
 - 0.7.0: More settings can be configured in the UI

{accusleepy-0.8.1.dist-info → accusleepy-0.9.2.dist-info}/RECORD

@@ -1,12 +1,13 @@
 accusleepy/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 accusleepy/__main__.py,sha256=dKzl2N2Hg9lD264CWYNxThRyDKzWwyMwHRXmJxOmMis,104
-accusleepy/bouts.py,sha256=biJEzCjMQR-RTlQcfeomkp6NxcA5FXkhE8Kp2XyL4P4,6039
-accusleepy/brain_state_set.py,sha256=fRkrArHLIbEKimub804yt_mUXoyfsjJEfiJnTjeCMkY,3233
-accusleepy/classification.py,sha256=mF35xMrD9QXGldSnl3vkdHbm7CAptPUNjHxUA_agOTA,9778
+accusleepy/bouts.py,sha256=GVYOSdsQjHzYirRRSFlie8w5iJq74umyGNMRVsQlpK0,6045
+accusleepy/brain_state_set.py,sha256=i9RUwSMKG21AjromjOJYpsHkip3elTFhHYntMguBF6Q,3273
+accusleepy/classification.py,sha256=UJ48QamTvqwqBwu-r4weSno-9TQyPFCk62nbUc5_iKA,8672
 accusleepy/config.json,sha256=VmUFsiGD1ymEyjdzqeM5nTp8jWvDI-DIxLy1_92nueo,875
-accusleepy/constants.py,sha256=62mmsr1NKzF-psS-9esuAE65kAcPL6o8v9UXQEvb5yc,2983
-accusleepy/fileio.py,sha256=-aEgp2oonjQKRizVI1uI5CGknnBOxE6HG0DzxDYbSkM,8512
+accusleepy/constants.py,sha256=y3tvYxJy3M1zycOPVfXLpvhLVdyd7s0dSlokgKG7mUE,3127
+accusleepy/fileio.py,sha256=A-ttN-N3wsEHrU89NvUYb5osuMK1HMmPecCDgtPTiqo,9076
 accusleepy/gui/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+accusleepy/gui/dialogs.py,sha256=DqkarJKgRovTuDDJhqPaw_o8ZfopuR7R6M8C49QWZu0,1230
 accusleepy/gui/icons/brightness_down.png,sha256=PLT1fb83RHIhSRuU7MMMx0G7oJAY7o9wUcnqM8veZfM,12432
 accusleepy/gui/icons/brightness_up.png,sha256=64GnUqgPvN5xZ6Um3wOzwqvUmdAWYZT6eFmWpBsHyks,12989
 accusleepy/gui/icons/double_down_arrow.png,sha256=fGiPIP7_RJ3UAonNhORFVX0emXEmtzRlHI3Tfjai064,4964
@@ -18,26 +19,29 @@ accusleepy/gui/icons/save.png,sha256=J3EA8iU1BqLYRSsrq_OdoZlqrv2yfL7oV54DklTy_DI
 accusleepy/gui/icons/up_arrow.png,sha256=V9yF9t1WgjPaUu-mF1YGe_DfaRHg2dUpR_sUVVcvVvY,3329
 accusleepy/gui/icons/zoom_in.png,sha256=MFWnKZp7Rvh4bLPq4Cqo4sB_jQYedUUtT8-ZO8tNYyc,13589
 accusleepy/gui/icons/zoom_out.png,sha256=IB8Jecb3i0U4qjWRR46ridjLpvLCSe7PozBaLqQqYSw,13055
-accusleepy/gui/images/primary_window.png,sha256=yRGUdIcByF8DtPtDWqjmnQ8J3IHKtmijBsT7N9e1Ij0,595764
+accusleepy/gui/images/primary_window.png,sha256=hO28hOOoQmBljED1UDQES1ZxaAq8431scOzHemKUpCA,168073
 accusleepy/gui/images/viewer_window.png,sha256=b_B7m9WSLMAOzNjctq76SyekO1WfC6qYZVNnYfhjPe8,977197
 accusleepy/gui/images/viewer_window_annotated.png,sha256=uMNUmsZIdzDlQpyoiS3lJGoWlg_T325Oj5hDZhM3Y14,146817
-accusleepy/gui/main.py,sha256=cZQH0qJ_mykeQP7eadc1yzFHYeDwbRngrR4_H1wMxWY,58438
-accusleepy/gui/manual_scoring.py,sha256=QCFF-CIU6nh26NB8Z9ktohCk8p7ola2hgjazD-OrK_o,40697
+accusleepy/gui/main.py,sha256=jtbeF1IXwCBDzpLTc_fFamkfDuNDc9ElotSLYPedWew,25094
+accusleepy/gui/manual_scoring.py,sha256=9Lom83nPqSr_nbudgPQ92rEOUBt-s9dh2ExcL9-dHRo,40698
 accusleepy/gui/mplwidget.py,sha256=rJSTtWmLjHn8r3c9Kb23Rc4XzXl3i9B-JrjNjjlNnmQ,13492
-accusleepy/gui/primary_window.py,sha256=qqV-GAC3BgOus1lJMOg2U4AGQbhy35zLMyN8Q7ouTd8,141479
-accusleepy/gui/primary_window.ui,sha256=2mD0G6b8bsyyUvMCUosQ0K-nAzMA2iiP3yMpeh9INcQ,208209
+accusleepy/gui/primary_window.py,sha256=-CbQ_dn3svE7J1VCGjTFDBCC4Q1DcV7Q7kz3UX3Kwhw,141613
+accusleepy/gui/primary_window.ui,sha256=WmeY7L9VyuiNRx3kSM56kPv5t6RDBVXlte8kT5wH-2s,208461
+accusleepy/gui/recording_manager.py,sha256=0lYrGu-7gX0xpLxBBcn4-TcuSy8CsmiPBJ3l_fZa158,3916
 accusleepy/gui/resources.qrc,sha256=wqPendnTLAuKfVI6v2lKHiRqAWM0oaz2ZuF5cucJdS4,803
 accusleepy/gui/resources_rc.py,sha256=Z2e34h30U4snJjnYdZVV9B6yjATKxxfvgTRt5uXtQdo,329727
+accusleepy/gui/settings_widget.py,sha256=OG9DojTnbHVqJjE0Nk-OVU5c4UlHv7vA94H_dFJB-Jc,16824
 accusleepy/gui/text/dev_guide.md,sha256=PgOXfGvN17fCtnsfGvPhrhK4FUWFGP_TsHA6t9skP3U,2060
-accusleepy/gui/text/main_guide.md,sha256=iZDRp5OWyQX9LV7CMeUFIYv2ryKlIcGALRLXjxR8HpI,8288
+accusleepy/gui/text/main_guide.md,sha256=QO0isKO-fqI-m062NY7k56fmBMNAV_ZFFNvE6ugCMi8,8289
 accusleepy/gui/text/manual_scoring_guide.md,sha256=ow_RMSjFy05NupEDSCuJtu-V65-BPnIkrZqtssFoZCQ,999
 accusleepy/gui/viewer_window.py,sha256=jysFw7C_Tr7mtK1XNWhIpHblBvatwduE3RF2GP4lrro,24479
 accusleepy/gui/viewer_window.ui,sha256=a89iVLk1sJg9N6ZvWAV6YNPStb2Tm4-rs-W7TmIDkb4,31658
-accusleepy/models.py,sha256=kqkcQJoKi7gpnM8gZ7nZbWGvpv7ruNnFLaB7ED1X6Dc,3493
+accusleepy/models.py,sha256=MHMsC9sYfR3fTOQLh4COyjCZo9q2vaX1FmpTHaaVtM4,3516
 accusleepy/multitaper.py,sha256=D5-iglwkFBRciL5tKSNcunMtcq0rM3zHwRHUVPgem1U,25679
-accusleepy/signal_processing.py,sha256=47fEAx8Aqqkiqix1ai2YEK9Fhq6UHoQcwAcOi-a8ewo,16834
-accusleepy/temperature_scaling.py,sha256=glvPcvxHpBdFjwjGfZdNku9L_BozycEmdqZhKKUCCNg,5749
-accusleepy/validation.py,sha256=VpLWK-wD5tCU6lTBG3KYgTi3PWGuYh6NitMgMoMH8JM,4434
-accusleepy-0.8.1.dist-info/METADATA,sha256=29kPvzKr2ykosWwidIt6FKrjvIY-m88Ui9GQPoQPN-g,4710
-accusleepy-0.8.1.dist-info/WHEEL,sha256=zp0Cn7JsFoX2ATtOhtaFYIiE2rmFAD4OcMhtUki8W3U,88
-accusleepy-0.8.1.dist-info/RECORD,,
+accusleepy/services.py,sha256=4xl9uWDaJXIpJTYH08mSgipkkEYM9fDLIcMj7-k9qf0,20228
+accusleepy/signal_processing.py,sha256=k3a2IKchok3-38P7T0XHNpGTiWKBqWtQm7SHwFEkTNA,19973
+accusleepy/temperature_scaling.py,sha256=RXsbDLT5JDrjPbz5ZdFXKj6NxpQCHpq5BzcLBzsUJKc,5848
+accusleepy/validation.py,sha256=h3iHVW71AaUC4lp1KMQXo5j81bS-IuYbTZWRSIkmHWE,6933
+accusleepy-0.9.2.dist-info/METADATA,sha256=50NVD9h3oIMy-R9WLVypyNO-KiGxHF8eeUzjvgWtUJ0,4733
+accusleepy-0.9.2.dist-info/WHEEL,sha256=3ny-bZhpXrU6vSQ1UPG34FoxZBp3lVcvK0LkgUz6VLk,88
+accusleepy-0.9.2.dist-info/RECORD,,

{accusleepy-0.8.1.dist-info → accusleepy-0.9.2.dist-info}/WHEEL

@@ -1,4 +1,4 @@
 Wheel-Version: 1.0
-Generator: poetry-core 2.2.1
+Generator: poetry-core 2.3.0
 Root-Is-Purelib: true
 Tag: py3-none-any