Preparing for public model release

classify_images.py

@@ -3,9 +3,8 @@
 # classify_images.py
 #
 # This is a test driver for running our species classifier. It classifies
-# one or more hard-coded image files. Think of this more like a demo notebook 
-# than like a script you'd actually use to run a zillion files through our
-# classifier.
+# one or more hard-coded image files, writing the top N results for each
+# to a .csv file.
 #
 # You should set the options in the 'Options' cell before running.
 #
@@ -19,11 +18,14 @@
 # By default, both point to URLs and will be downloaded to a temp directory
 # automatically.
 #
-# This code has been tested against pytorch 1.2, so dependencies can be installed as:
+# Dependencies (may work against later versions, but this was the test environment):
 #
 # conda install pytorch==1.2.0 torchvision==0.4.0 cudatoolkit=10.0 -c pytorch
-# pip install pretrainedmodels
-# pip install "pillow<7"
+# pip install pretrainedmodels==0.7.4
+# pip install pillow==6.1.0
+# pip install progressbar2==3.51.0
+# pip install cupy-cuda100==7.3.0
+# pip install torchnet==0.0.4
 #
 ####### 
 
@@ -48,7 +50,7 @@
 # little path management. This also implicitly defers PyTorch imports.
 
 # Directory to which you sync'd this repo.
-api_root = r'c:\git\speciesclassification'
+api_root = r'/home/coyote/git/speciesclassification'
 
 # If not None, pre-pended to filenames. Most useful when filenames are coming from 
 # a .csv file.
@@ -63,13 +65,13 @@
 # a single image file
 #
 # a directory, which is recursively enumerated
-images_to_classify = r'g:\temp\elephant.jpg'
-# images_to_classify = [r'g:\temp\elephant.jpg']
+# images_to_classify = r'/data/species-classification/elephant.jpg'
+# images_to_classify = [r'/data/species-classification/elephant.jpg']
 # images_to_classify = 'image_list.csv'
-# images_to_classify = r'g:\temp\animal_images'
+images_to_classify = r'/data/species-classification/images/sample_images.2019.12.28'
 
 # Classification results will be written here
-classification_output_file = 'g:\temp\classification_output.csv'
+classification_output_file = '/data/species-classification/classification_output.csv'
 
 # Path to taxa.csv, for latin --> common mapping
 #
@@ -89,7 +91,7 @@
 
 # This must be True if detection is enabled. Classification can be run
 # on the CPU or GPU.
-use_gpu = False
+use_gpu = True
 
 
 #%% Constants 
@@ -127,42 +129,6 @@
 import api as speciesapi
 
 
-#%% Build Latin --> common mapping
-
-latin_to_common = {}
-
-if taxonomy_path is not None:
-
- print('Reading taxonomy file')
-
- # Read taxonomy file; takes ~1 minute
- df = pd.read_csv(taxonomy_path)
- df = df.fillna('')
-
- # Columns are:
- #
- # taxonID,scientificName,parentNameUsageID,taxonRank,vernacularName,wikipedia_url
-
- # Create dictionary by ID
-
- nRows = df.shape[0]
-
- for index, row in df.iterrows():
-
- latin_name = row['scientificName']
- latin_name = latin_name.strip()
- if len(latin_name)==0:
- print('Warning: invalid scientific name at {}'.format(index))
- latin_name = 'unknown'
- common_name = row['vernacularName']
- common_name = common_name.strip()
- latin_name = latin_name.lower()
- common_name = common_name.lower()
- latin_to_common[latin_name] = common_name
-
- print('Finished reading taxonomy file')
-
-
 #%% Support functions
 
 class DownloadProgressBar():
@@ -255,8 +221,41 @@ def do_latin_to_common(latin_name):
  assert os.path.isfile(detection_model_path)
 
 
-#%% Create the model(s)
+#%% Build Latin --> common mapping
 
+latin_to_common = {}
+
+if taxonomy_path is not None:
+
+ print('Reading taxonomy file from {}'.format(taxonomy_path))
+
+ # Read taxonomy file; takes ~1 minute
+ df = pd.read_csv(taxonomy_path)
+ df = df.fillna('')
+
+ # Columns are:
+ #
+ # taxonID,scientificName,parentNameUsageID,taxonRank,vernacularName,wikipedia_url
+
+ nRows = df.shape[0]
+
+ for index, row in df.iterrows():
+
+ latin_name = row['scientificName']
+ latin_name = latin_name.strip()
+ if len(latin_name)==0:
+ print('Warning: invalid scientific name at {}'.format(index))
+ latin_name = 'unknown'
+ common_name = row['vernacularName']
+ common_name = common_name.strip()
+ latin_name = latin_name.lower()
+ common_name = common_name.lower()
+ latin_to_common[latin_name] = common_name
+
+ print('Finished reading taxonomy file')
+
+
+#%% Create the model(s)
 
 print('Loading model')
 model = speciesapi.DetectionClassificationAPI(classification_model_path, 
@@ -270,10 +269,6 @@ def do_latin_to_common(latin_name):
 
 queries = None
 
-# If we specified a single image
-if isinstance(images_to_classify,str) and (not images_to_classify.endswith('.csv')):
- images_to_classify = [images_to_classify]
-
 # If we specified a folder 
 if isinstance(images_to_classify,str) and os.path.isdir(images_to_classify):
 
@@ -294,13 +289,16 @@ def do_latin_to_common(latin_name):
  queries = list(df_images.query_string)
  assert(len(queries) == len(images))
 
-# If we specified a list 
-elif isinstance(images_to_classify,list):
+# If we specified a list or a single file  
+else:
 
+ if isinstance(images_to_classify,str):
+ images_to_classify = [images_to_classify]
+
  assert isinstance(images_to_classify,list)
  images = images_to_classify
  queries = None
- print('Processing list of {} images'.format(len(images)))
+ print('Processing a list of {} images'.format(len(images)))
 
 
 #%% Classify images