Merge branch 'dev' into kgr10_roberta

evaluate_tsv.py

@@ -20,13 +20,21 @@ def main(args):
     sentences = [sentence[0] for sentence in sentences_labels]
     labels = [sentence[1] for sentence in sentences_labels]
 
-    time_start = time.clock()
+    data_size = 0
+    for sentence in sentences:
+        data_size += sum([len(token) + 1 for token in sentence])
+
+    t0 = time.clock()
     predictions = ner.process(sentences)
-    time_end = time.clock()
+    time_processing = time.clock() - t0
 
     report = classification_report(labels, predictions, digits=4)
     print(report)
-    print("Processing time: %d second(s)" % (time_end-time_start))
+
+    print(f"Total time                  : {time_processing:>8.4} second(s)")
+    print(f"Data size:                  : {data_size/1000000:>8.4} M characters")
+    print(f"Speed:                      : {data_size / 1000000 / (time_processing/60):>8.4} M characters/minute")
+    print(f"Number of token labels      : {len(ner.label_list):>8} ")
 
 
 def parse_args():
@@ -39,7 +47,7 @@ def parse_args():
                         help='the maximum total input sequence length after WordPiece tokenization.')
     parser.add_argument('--device', required=False, default="cpu", metavar='cpu|cuda',
                         help='device type used for processing')
-    parser.add_argument('--squeeze', required=False, default=False,
+    parser.add_argument('--squeeze', required=False, default=False, action="store_true",
                         help='try to squeeze multiple examples into one Input Feature')
     return parser.parse_args()
 

main.py

@@ -95,7 +95,7 @@ def main():
 
     logger.info("Loading pretrained model...")
     t0 = time.time()
-    if args.pretrained_path.startswith("automodel:"):
+    if args.pretrained_path.startswith("hf:"):
         from poldeepner2.model.herbert_for_token_calssification import AutoTokenizerForTokenClassification
         pretrained_dir = args.pretrained_path.split(':')[1]
         model = AutoTokenizerForTokenClassification(

poleval_ner_test_v2.py

 import sys, json, getopt
 
+from tqdm import tqdm
 from attr import dataclass
 from dateutil import parser
 
@@ -130,9 +131,8 @@ def computeScores(goldfile, userfile, cn: CategoryNormalizer, htype="split", typ
     with open(goldfile) as json_data:
         goldjson = json.load(json_data)
 
-    for nr in range(len(goldjson['questions'])):
+    for nr in tqdm(range(len(goldjson['questions']))):
         idGold = '/'.join(goldjson['questions'][nr]['input']['fname'].split('/')[4:])
-        print(idGold)
         if idGold in idsToAnnsUser:
             found += 1
             # find the most recent answer: