deepliif 1.1.8__py3-none-any.whl → 1.1.9__py3-none-any.whl

deepliif/models/__init__.py

@@ -56,8 +56,11 @@ def get_opt(model_dir, mode='test'):
             opt = Options(path_file=os.path.join(model_dir,'test_opt.txt'), mode=mode)
         except:
             opt = Options(path_file=os.path.join(model_dir,'train_opt.txt'), mode=mode)
+        opt.use_dp = False
+        opt.gpu_ids = list(range(torch.cuda.device_count()))
     return opt
 
+
 def find_model_using_name(model_name):
     """Import the module "models/[model_name]_model.py".
 
@@ -148,7 +151,7 @@ def init_nets(model_dir, eager_mode=False, opt=None, phase='test'):
     if opt is None:
         opt = get_opt(model_dir, mode=phase)
         opt.use_dp = False
-        print_options(opt)
+        #print_options(opt)
     
     if opt.model == 'DeepLIIF':
         net_groups = [
@@ -168,7 +171,7 @@ def init_nets(model_dir, eager_mode=False, opt=None, phase='test'):
 
     number_of_gpus_all = torch.cuda.device_count()
     number_of_gpus = len(opt.gpu_ids)
-    print(number_of_gpus)
+    #print(number_of_gpus)
     if number_of_gpus > 0:
         mapping_gpu_ids = {i:idx for i,idx in enumerate(opt.gpu_ids)}
         chunks = [itertools.chain.from_iterable(c) for c in chunker(net_groups, number_of_gpus)]
@@ -351,7 +354,7 @@ def inference(img, tile_size, overlap_size, model_path, use_torchserve=False, ea
               color_dapi=False, color_marker=False, opt=None):
     if not opt:
         opt = get_opt(model_path)
-        print_options(opt)
+        #print_options(opt)
     
     if opt.model == 'DeepLIIF':
         rescaled, rows, cols = format_image_for_tiling(img, tile_size, overlap_size)
@@ -424,18 +427,18 @@ def inference(img, tile_size, overlap_size, model_path, use_torchserve=False, ea
         raise Exception(f'inference() not implemented for model {opt.model}')
 
 
-def postprocess(orig, images, tile_size, seg_thresh=150, size_thresh='default', marker_thresh='default', size_thresh_upper=None, opt=None):
-    if opt.model == 'DeepLIIF':
+def postprocess(orig, images, tile_size, model, seg_thresh=150, size_thresh='auto', marker_thresh='auto', size_thresh_upper=None):
+    if model == 'DeepLIIF':
         resolution = '40x' if tile_size > 384 else ('20x' if tile_size > 192 else '10x')
         overlay, refined, scoring = compute_results(np.array(orig), np.array(images['Seg']),
-                                                    np.array(images['Marker'].convert('L')), resolution,
-                                                    seg_thresh, size_thresh, marker_thresh, size_thresh_upper)
+                                                    np.array(images['Marker'].convert('L')) if 'Marker' in images else None,
+                                                    resolution, seg_thresh, size_thresh, marker_thresh, size_thresh_upper)
         processed_images = {}
         processed_images['SegOverlaid'] = Image.fromarray(overlay)
         processed_images['SegRefined'] = Image.fromarray(refined)
         return processed_images, scoring
 
-    elif opt.model == 'DeepLIIFExt':
+    elif model == 'DeepLIIFExt':
         resolution = '40x' if tile_size > 768 else ('20x' if tile_size > 384 else '10x')
         processed_images = {}
         scoring = {}
@@ -452,7 +455,7 @@ def postprocess(orig, images, tile_size, seg_thresh=150, size_thresh='default',
         return processed_images, scoring
 
     else:
-        raise Exception(f'postprocess() not implemented for model {opt.model}')
+        raise Exception(f'postprocess() not implemented for model {model}')
 
 
 def infer_modalities(img, tile_size, model_dir, eager_mode=False,
@@ -467,7 +470,7 @@ def infer_modalities(img, tile_size, model_dir, eager_mode=False,
     if opt is None:
         opt = get_opt(model_dir)
         opt.use_dp = False
-        print_options(opt)
+        #print_options(opt)
     
     if not tile_size:
         tile_size = check_multi_scale(Image.open('./images/target.png').convert('L'),
@@ -486,7 +489,7 @@ def infer_modalities(img, tile_size, model_dir, eager_mode=False,
     )
     
     if not hasattr(opt,'seg_gen') or (hasattr(opt,'seg_gen') and opt.seg_gen): # the first condition accounts for old settings of deepliif; the second refers to deepliifext models
-        post_images, scoring = postprocess(img, images, tile_size, opt=opt)
+        post_images, scoring = postprocess(img, images, tile_size, opt.model)
         images = {**images, **post_images}
         return images, scoring
     else:

deepliif/postprocessing.py

@@ -401,23 +401,30 @@ def calc_default_marker_thresh(marker):
         return 0
 
 
-def compute_results(orig, seg, marker, resolution=None, seg_thresh=150, size_thresh='default', marker_thresh='default', size_thresh_upper=None):
+def compute_results(orig, seg, marker, resolution=None, seg_thresh=150, size_thresh='auto', marker_thresh='auto', size_thresh_upper=None):
     mask = create_posneg_mask(seg, seg_thresh)
     mark_background(mask)
 
-    if size_thresh == 'default':
+    if size_thresh == 'auto':
         size_thresh = calc_default_size_thresh(mask, resolution)
-    if marker_thresh == 'default':
+    if marker_thresh is None:
+        marker_thresh = 0
+        marker = None
+    elif marker_thresh == 'auto':
         marker_thresh = calc_default_marker_thresh(marker)
 
-    counts = compute_cell_classification(mask, marker, size_thresh, marker_thresh, size_thresh_upper=None)
+    counts = compute_cell_classification(mask, marker, size_thresh, marker_thresh, size_thresh_upper)
     enlarge_cell_boundaries(mask)
 
     scoring = {
         'num_total': counts['num_total'],
         'num_pos': counts['num_pos'],
         'num_neg': counts['num_neg'],
-        'percent_pos': round(counts['num_pos'] / counts['num_total'] * 100, 1) if counts['num_pos'] > 0 else 0
+        'percent_pos': round(counts['num_pos'] / counts['num_total'] * 100, 1) if counts['num_pos'] > 0 else 0,
+        'prob_thresh': seg_thresh,
+        'size_thresh': size_thresh,
+        'size_thresh_upper': size_thresh_upper,
+        'marker_thresh': marker_thresh if marker is not None else None,
     }
 
     overlay = np.copy(orig)

{deepliif-1.1.8.dist-info → deepliif-1.1.9.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: deepliif
-Version: 1.1.8
+Version: 1.1.9
 Summary: DeepLIIF: Deep-Learning Inferred Multiplex Immunofluorescence for Immunohistochemical Image Quantification
 Home-page: https://github.com/nadeemlab/DeepLIIF
 Author: Parmida93
@@ -269,7 +269,7 @@ img (required)
 file: image to run the models on
 
 resolution
-string: resolution used to scan the slide (10x, 20x, 40x), defaults to 20x 
+string: resolution used to scan the slide (10x, 20x, 40x), defaults to 40x 
 
 pil
 boolean: if true, use PIL.Image.open() to load the image, instead of python-bioformats
@@ -298,9 +298,9 @@ res = requests.post(
     files={
         'img': open(f'{images_dir}/{filename}', 'rb')
     },
-    # optional param that can be 10x, 20x (default) or 40x
+    # optional param that can be 10x, 20x, or 40x (default)
     params={
-        'resolution': '20x'
+        'resolution': '40x'
     }
 )
 

{deepliif-1.1.8.dist-info → deepliif-1.1.9.dist-info}/RECORD

@@ -1,6 +1,6 @@
 cli.py,sha256=elGu-9di_LcUXPsWWknWPevbaTv2r2rpORVPH5lsnAU,39723
 deepliif/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-deepliif/postprocessing.py,sha256=QrTT7HodlLyrNFrI9hJowkQet6YVVIosf9B1iK4ajQk,17327
+deepliif/postprocessing.py,sha256=cM-cYVidY691Sjb1-B8a1jkLq5UR_hTCbuKzuF4765o,17589
 deepliif/train.py,sha256=-ZORL5vQrD0_Jq2Adgr3w8vJ7L1QcAgNTqMnBgtixgk,15757
 deepliif/data/__init__.py,sha256=euf9eUboK4RYR0jvdiyZDgPGozC1Nv7WRqRbTxSZD6A,5281
 deepliif/data/aligned_dataset.py,sha256=bAofVfgMwtb8Exe4EtQ3aP2ZYewBT4N_X8BDWSeFFj0,4405
@@ -12,7 +12,7 @@ deepliif/data/template_dataset.py,sha256=PCDBnFRzRKReaeWgKUZmW0LrzRByI9adrKDJ6SN
 deepliif/data/unaligned_dataset.py,sha256=m7j-CX-hkXbhg96NSEcaCagNVhTuXKkMsBADdMEJDBA,3393
 deepliif/models/DeepLIIFExt_model.py,sha256=Sc60rOfDJuoGrJ1CYe4beAg6as6F0o864AO6ZB7paBY,14527
 deepliif/models/DeepLIIF_model.py,sha256=ECZyM9jzoJAWSgB1ProBoarVuGcbScQMaSkRjSMgt0k,20872
-deepliif/models/__init__.py,sha256=0vseZQ6LX6zrUQuMnmpcZDjyyjuVb_vMBi_hyiAVwcw,22565
+deepliif/models/__init__.py,sha256=E2udWyU4eScFnvDO2qtwMeHRz4ihw8hhCoP666QBK4o,22674
 deepliif/models/base_model.py,sha256=HKcUOBHtL-zLs5ZcmeXT-ZV_ubqsSUo4wMCQ0W27YHU,15583
 deepliif/models/networks.py,sha256=bN4yjRdE413efUESq8pvhzPDgFCTwFKXyQOrRqHckWY,32177
 deepliif/options/__init__.py,sha256=nm231wh_hQw1mX0AfXZu7-cx6WZdxURsE-jvnjlHQzE,4581
@@ -26,9 +26,9 @@ deepliif/util/html.py,sha256=RNAONZ4opP-bViahgmpSbHwOc6jXKQRnWRAVIaeIvac,3309
 deepliif/util/image_pool.py,sha256=M89Hc7DblRWroNP71S9mAdRn7h3DrhPFPjqFxxZYSgw,2280
 deepliif/util/util.py,sha256=bTArzuhIMGgGweH0v5rkiHrqBxc24BDv12rssOE9OoI,4636
 deepliif/util/visualizer.py,sha256=5V1lWidHqssJX21jn1P5-bOVgtrEXKVaQgnMWAsMfqg,15636
-deepliif-1.1.8.dist-info/LICENSE.md,sha256=HlZw_UPS6EtJimJ_Ci7xKh-S5Iubs0Z8y8E6EZ3ZNyE,956
-deepliif-1.1.8.dist-info/METADATA,sha256=kXAkqKbOVhBUAhMuF7vTqZAXr8rJuLKU8dykdnr47jQ,25075
-deepliif-1.1.8.dist-info/WHEEL,sha256=pkctZYzUS4AYVn6dJ-7367OJZivF2e8RA9b_ZBjif18,92
-deepliif-1.1.8.dist-info/entry_points.txt,sha256=f70-10j2q68o_rDlsE3hspnv4ejlDnXwwGZ9JJ-3yF4,37
-deepliif-1.1.8.dist-info/top_level.txt,sha256=vLDK5YKmDz08E7PywuvEjAo7dM5rnIpsjR4c0ubQCnc,13
-deepliif-1.1.8.dist-info/RECORD,,
+deepliif-1.1.9.dist-info/LICENSE.md,sha256=HlZw_UPS6EtJimJ_Ci7xKh-S5Iubs0Z8y8E6EZ3ZNyE,956
+deepliif-1.1.9.dist-info/METADATA,sha256=CBgGs3wxeg1Hewt7lbAFUEH4Sbm5jV6jt4F3q59gvzM,25076
+deepliif-1.1.9.dist-info/WHEEL,sha256=pkctZYzUS4AYVn6dJ-7367OJZivF2e8RA9b_ZBjif18,92
+deepliif-1.1.9.dist-info/entry_points.txt,sha256=f70-10j2q68o_rDlsE3hspnv4ejlDnXwwGZ9JJ-3yF4,37
+deepliif-1.1.9.dist-info/top_level.txt,sha256=vLDK5YKmDz08E7PywuvEjAo7dM5rnIpsjR4c0ubQCnc,13
+deepliif-1.1.9.dist-info/RECORD,,