spacr 0.1.6__py3-none-any.whl → 0.1.7__py3-none-any.whl

spacr/measure.py

@@ -16,12 +16,6 @@ from skimage.util import img_as_bool
 
 from .logger import log_function_call
 
-#from .io import create_database, _save_settings_to_db
-#from .timelapse import _timelapse_masks_to_gif, _scmovie
-#from .plot import _plot_cropped_arrays, _save_scimg_plot
-#from .utils import _merge_overlapping_objects, _filter_object, _relabel_parent_with_child_labels, _exclude_objects
-#from .utils import _merge_and_save_to_database, _crop_center, _find_bounding_box, _generate_names, _get_percentiles, normalize_to_dtype, _map_wells_png, _list_endpoint_subdirectories, _generate_representative_images
-
 def get_components(cell_mask, nucleus_mask, pathogen_mask):
     """
     Get the components (nucleus and pathogens) for each cell in the given masks.
@@ -626,8 +620,14 @@ def _measure_crop_core(index, time_ls, file, settings):
 
         file_name = os.path.splitext(file)[0]
         data = np.load(os.path.join(settings['input_folder'], file))
-
         data_type = data.dtype
+        if data_type not in ['uint8','uint16']:
+            data_type_before = data_type
+            data = data.astype(np.uint16)
+            data_type = data.dtype
+            if settings['verbose']:
+                print(f'Converted data from {data_type_before} to {data_type}')
+
         if settings['save_measurements']:
             os.makedirs(source_folder+'/measurements', exist_ok=True)
             _create_database(source_folder+'/measurements/measurements.db')    
@@ -832,7 +832,6 @@ def _measure_crop_core(index, time_ls, file, settings):
                                     png_channels = normalize_to_dtype(png_channels, settings['normalize'][0], settings['normalize'][1], percentile_list=percentile_list)
                             else:
                                 png_channels = normalize_to_dtype(png_channels, 0, 100)
-
                             os.makedirs(png_folder, exist_ok=True)
 
                             if png_channels.shape[2] == 2:
@@ -999,12 +998,12 @@ def measure_crop(settings):
     _save_settings_to_db(settings)
 
     files = [f for f in os.listdir(settings['input_folder']) if f.endswith('.npy')]
-    max_workers = settings['max_workers'] or mp.cpu_count()-4
-    print(f'using {max_workers} cpu cores')
+    n_job = settings['n_job'] or mp.cpu_count()-4
+    print(f'using {n_job} cpu cores')
 
     with mp.Manager() as manager:
         time_ls = manager.list()
-        with mp.Pool(max_workers) as pool:
+        with mp.Pool(n_job) as pool:
             result = pool.starmap_async(_measure_crop_core, [(index, time_ls, file, settings) for index, file in enumerate(files)])
 
             # Track progress in the main process
@@ -1013,7 +1012,7 @@ def measure_crop(settings):
                 files_processed = len(time_ls)
                 files_to_process = len(files)
                 average_time = np.mean(time_ls) if len(time_ls) > 0 else 0
-                time_left = (((files_to_process-files_processed)*average_time)/max_workers)/60
+                time_left = (((files_to_process-files_processed)*average_time)/n_job)/60
                 print(f'Progress: {files_processed}/{files_to_process} Time/img {average_time:.3f}sec, Time Remaining {time_left:.3f} min.', end='\r', flush=True)
             result.get()
 

spacr/plot.py

@@ -1565,4 +1565,56 @@ def plot_feature_importance(feature_importance_df):
     ax.set_xlabel('Feature Importance', fontsize=font_size)
     ax.tick_params(axis='both', which='major', labelsize=font_size)
     plt.tight_layout()
-    return fig
+    return fig
+
+def read_and_plot__vision_results(base_dir, y_axis='accuracy', name_split='_time', y_lim=[0.8, 0.9]):
+    # List to store data from all CSV files
+    data_frames = []
+
+    dst = os.path.join(base_dir, 'result')
+    os.mkdir(dst,exists=True)
+
+    # Walk through the directory
+    for root, dirs, files in os.walk(base_dir):
+        for file in files:
+            if file.endswith("_test_result.csv"):
+                file_path = os.path.join(root, file)
+                # Extract model information from the file name
+                file_name = os.path.basename(file_path)
+                model = file_name.split(f'{name_split}')[0]
+                
+                # Extract epoch information from the file name
+                epoch_info = file_name.split('_time')[1]
+                base_folder = os.path.dirname(file_path)
+                epoch = os.path.basename(base_folder)
+                
+                # Read the CSV file
+                df = pd.read_csv(file_path)
+                df['model'] = model
+                df['epoch'] = epoch
+                
+                # Append the data frame to the list
+                data_frames.append(df)
+    
+    # Concatenate all data frames
+    if data_frames:
+        result_df = pd.concat(data_frames, ignore_index=True)
+        
+        # Calculate average y_axis per model
+        avg_metric = result_df.groupby('model')[y_axis].mean().reset_index()
+        avg_metric = avg_metric.sort_values(by=y_axis)
+        print(avg_metric)
+        
+        # Plotting the results
+        plt.figure(figsize=(10, 6))
+        plt.bar(avg_metric['model'], avg_metric[y_axis])
+        plt.xlabel('Model')
+        plt.ylabel(f'{y_axis}')
+        plt.title(f'Average {y_axis.capitalize()} per Model')
+        plt.xticks(rotation=45)
+        plt.tight_layout()
+        if y_lim is not None:
+            plt.ylim(y_lim)
+        plt.show()
+    else:
+        print("No CSV files found in the specified directory.")