Commit 6442322f authored by szr712's avatar szr712

使用新增数据训练

parent c217436c
...@@ -24,14 +24,17 @@ CUDA_VISIBLE_DEVICES=1 python train2.py -src_data data/pinyin_2.txt -trg_data da ...@@ -24,14 +24,17 @@ CUDA_VISIBLE_DEVICES=1 python train2.py -src_data data/pinyin_2.txt -trg_data da
CUDA_VISIBLE_DEVICES=1 nohup python train2.py -src_data data/pinyin_2.txt -trg_data data/hanzi_2.txt -src_lang en_core_web_sm -trg_lang fr_core_news_sm -epochs 100 -model_name token_classification CUDA_VISIBLE_DEVICES=1 nohup python train2.py -src_data data/pinyin_2.txt -trg_data data/hanzi_2.txt -src_lang en_core_web_sm -trg_lang fr_core_news_sm -epochs 100 -model_name token_classification
CUDA_VISIBLE_DEVICES=3 python translate_pkl.py -load_weights weights/token_classification/11-09_22:00:55/token_classification_35_0.055335590355098246 -pkl_dir weights/token_classification/11-09_22:00:55 -test_dir data/pkl/test-pkl -result_dir data/pkl/test-pkl-result CUDA_VISIBLE_DEVICES=3 python translate_pkl.py -load_weights weights/token_classification/11-09_22:00:55/token_classification_35_0.055335590355098246 -pkl_dir weights/token_classification/11-09_22:00:55 -test_dir data/pkl/pinyin_pkl -result_dir data/pkl/pinyin_pkl_result
CUDA_VISIBLE_DEVICES=2 nohup python train_token_classification.py -src_data data/pinyin_split.txt -trg_data data/hanzi_split.txt -src_lang en_core_web_sm -trg_lang fr_core_news_sm -epochs 100 -model_name token_classification_split -src_voc ./data/voc/pinyin.txt -trg_voc ./data/voc/hanzi.txt CUDA_VISIBLE_DEVICES=2 nohup python train_token_classification.py -src_data data/pinyin_split.txt -trg_data data/hanzi_split.txt -src_lang en_core_web_sm -trg_lang fr_core_news_sm -epochs 100 -model_name token_classification_split -src_voc ./data/voc/pinyin.txt -trg_voc ./data/voc/hanzi.txt
CUDA_VISIBLE_DEVICES=1 python translate2.py -load_weights weights/token_classification_split_2/11-19_17:16:18/token_classification_split_2_5_0.05776993067935109 -pkl_dir weights/token_classification_split_2/11-19_17:16:18 -src_voc ./data/voc/pinyin.txt -trg_voc ./data/voc/hanzi.txt CUDA_VISIBLE_DEVICES=1 python translate2.py -load_weights weights/token_classification_split_4/11-23_22:02:06/token_classification_split_4_25_0.02742394618457183 -pkl_dir weights/token_classification_split_4/11-23_22:02:06 -src_voc ./data/voc/pinyin.txt -trg_voc ./data/voc/hanzi.txt
CUDA_VISIBLE_DEVICES=4 nohup python translate_file2.py -load_weights weights/token_classification_split_3/11-22_21:56:11/token_classification_split_3_25_0.029638311734888702 -pkl_dir weights/token_classification_split_3/11-22_21:56:11 -test_dir data/test_data/pinyin_split -result_dir data/test_data/result_split -src_voc ./data/voc/pinyin.txt -trg_voc ./data/voc/hanzi.txt >log1 2>&1 & CUDA_VISIBLE_DEVICES=4 nohup python translate_file2.py -load_weights weights/token_classification_split_4/11-23_22:02:06/token_classification_split_4_25_0.02742394618457183 -pkl_dir weights/token_classification_split_4/11-23_22:02:06 -test_dir data/test_data/pinyin_split -result_dir data/test_data/result_split -src_voc ./data/voc/pinyin.txt -trg_voc ./data/voc/hanzi.txt >log1 2>&1 &
CUDA_VISIBLE_DEVICES=3 python eval_model.py -load_weights weights/token_classification_split_3/11-22_21:56:11/token_classification_split_3_1_0.09703897424042225 -pkl_dir weights/token_classification_split_3/11-22_21:56:11 -dev_dir data/dev -src_voc ./data/voc/pinyin.txt -trg_voc ./data/voc/hanzi.txt CUDA_VISIBLE_DEVICES=1 python eval_model.py -load_weights weights/token_classification_split_4/11-23_22:02:06/token_classification_split_4_1_0.09183966986835003 -pkl_dir weights/token_classification_split_4/11-23_22:02:06 -dev_dir data/dev -src_voc ./data/voc/pinyin.txt -trg_voc ./data/voc/hanzi.txt >log1 2>&1 &
CUDA_VISIBLE_DEVICES=2 nohup python train_token_classification.py -src_data data/pinyin_split.txt -trg_data data/hanzi_split.txt -src_lang en_core_web_sm -trg_lang fr_core_news_sm -epochs 100 -model_name token_classification_split_4 -src_voc ./data/voc/pinyin.txt -trg_voc ./data/voc/hanzi.txt CUDA_VISIBLE_DEVICES=6 nohup python train_token_classification.py -src_data data/pinyin_split.txt -trg_data data/hanzi_split.txt -src_lang en_core_web_sm -trg_lang fr_core_news_sm -epochs 100 -model_name token_classification_split_4 -src_voc ./data/voc/pinyin.txt -trg_voc ./data/voc/hanzi.txt
\ No newline at end of file
CUDA_VISIBLE_DEVICES=2 nohup python train_token_classification.py -src_data data/pinyin_new_split.txt -trg_data data/hanzi_new_split.txt -src_lang en_core_web_sm -trg_lang fr_core_news_sm -epochs 100 -model_name token_classification_split_new -src_voc ./data/voc/pinyin.txt -trg_voc ./data/voc/hanzi.txt >log1 2>&1 &
...@@ -214,6 +214,6 @@ if __name__ == "__main__": ...@@ -214,6 +214,6 @@ if __name__ == "__main__":
# with open("./data/voc/yunmu.txt","r",encoding="utf-8") as f: # with open("./data/voc/yunmu.txt","r",encoding="utf-8") as f:
# yunmus=f.readlines() # yunmus=f.readlines()
# yunmus=[a.strip() for a in yunmus] # yunmus=[a.strip() for a in yunmus]
build_corpus("./data/train_set_total.txt", build_corpus("./data/train_set_new.txt",
"./data/pinyin_split.txt", "./data/hanzi_split.txt") "./data/pinyin_new_split.txt", "./data/hanzi_new_split.txt")
print("Done") print("Done")
...@@ -4,8 +4,8 @@ from tqdm import tqdm ...@@ -4,8 +4,8 @@ from tqdm import tqdm
from build_corpus import split_initials_finals, wenzi2pinyin from build_corpus import split_initials_finals, wenzi2pinyin
hanzi_dir="./data/test_data/hanzi" hanzi_dir="./data/test_data/hanzi_new"
pinyin_dir="./data/test_data/pinyin_split" pinyin_dir="./data/test_data/pinyin_split_new"
with open("./data/voc/yunmu.txt","r",encoding="utf-8") as f: with open("./data/voc/yunmu.txt","r",encoding="utf-8") as f:
yunmus=f.readlines() yunmus=f.readlines()
......
...@@ -91,7 +91,7 @@ def main(): ...@@ -91,7 +91,7 @@ def main():
i=1 i=1
while i<=60: while i<=60:
for model_name in os.listdir(opt.pkl_dir): for model_name in os.listdir(opt.pkl_dir):
if "token_classification_split_3_"+str(i)+"_" in model_name: if "token_classification_split_new_"+str(i)+"_" in model_name:
print("model_name:{}".format(model_name)) print("model_name:{}".format(model_name))
opt.load_weights=os.path.join(opt.pkl_dir,model_name) opt.load_weights=os.path.join(opt.pkl_dir,model_name)
......
...@@ -165,4 +165,13 @@ model saved as token_classification_split_3_45_0.023 ...@@ -165,4 +165,13 @@ model saved as token_classification_split_3_45_0.023
1309m: epoch 46 [ ] 0% loss = ... 1309m: epoch 46 [ ] 0% loss = 0.018 1309m: epoch 46 [ ] 0% loss = 0.018 1309m: epoch 46 [ ] 1% loss = 0.020 1309m: epoch 46 [ ] 1% loss = 0.019 1309m: epoch 46 [ ] 1% loss = 0.019 1310m: epoch 46 [ ] 2% loss = 0.019 1310m: epoch 46 [ ] 2% loss = 0.020 1310m: epoch 46 [ ] 2% loss = 0.021 1310m: epoch 46 [ ] 3% loss = 0.020 1310m: epoch 46 [ ] 3% loss = 0.019 1310m: epoch 46 [ ] 3% loss = 0.021 1310m: epoch 46 [ ] 4% loss = 0.020 1310m: epoch 46 [ ] 4% loss = 0.020 1310m: epoch 46 [ ] 4% loss = 0.020 1311m: epoch 46 [# ] 5% loss = 0.020 1311m: epoch 46 [# ] 5% loss = 0.020 1311m: epoch 46 [# ] 5% loss = 0.020 1311m: epoch 46 [# ] 6% loss = 0.019 1311m: epoch 46 [# ] 6% loss = 0.020 1311m: epoch 46 [# ] 6% loss = 0.021 1311m: epoch 46 [# ] 7% loss = 0.020 1311m: epoch 46 [# ] 7% loss = 0.020 1311m: epoch 46 [# ] 7% loss = 0.019 1312m: epoch 46 [# ] 8% loss = 0.019 1312m: epoch 46 [# ] 8% loss = 0.020 1312m: epoch 46 [# ] 8% loss = 0.022 1312m: epoch 46 [# ] 9% loss = 0.019 1312m: epoch 46 [# ] 9% loss = 0.020 1312m: epoch 46 [# ] 9% loss = 0.020 1312m: epoch 46 [## ] 10% loss = 0.020 1312m: epoch 46 [## ] 10% loss = 0.020 1312m: epoch 46 [## ] 11% loss = 0.020 1313m: epoch 46 [## ] 11% loss = 0.019 1313m: epoch 46 [## ] 11% loss = 0.019 1313m: epoch 46 [## ] 12% loss = 0.020 1313m: epoch 46 [## ] 12% loss = 0.020 1313m: epoch 46 [## ] 12% loss = 0.020 1313m: epoch 46 [## ] 13% loss = 0.020 1313m: epoch 46 [## ] 13% loss = 0.019 1313m: epoch 46 [## ] 13% loss = 0.021 1313m: epoch 46 [## ] 14% loss = 0.021 1314m: epoch 46 [## ] 14% loss = 0.020 1314m: epoch 46 [## ] 14% loss = 0.019 1314m: epoch 46 [### ] 15% loss = 0.019 1314m: epoch 46 [### ] 15% loss = 0.020 1314m: epoch 46 [### ] 15% loss = 0.019 1314m: epoch 46 [### ] 16% loss = 0.022 1314m: epoch 46 [### ] 16% loss = 0.020 1314m: epoch 46 [### ] 16% loss = 0.021 1314m: epoch 46 [### ] 17% loss = 0.020 1315m: epoch 46 [### ] 17% loss = 0.020 1315m: epoch 46 [### ] 17% loss = 0.019 1315m: epoch 46 [### ] 18% loss = 0.020 1315m: epoch 46 [### ] 18% loss = 0.021 1315m: epoch 46 [### ] 18% loss = 0.021 1315m: epoch 46 [### ] 19% loss = 0.021 1315m: epoch 46 [### ] 19% loss = 0.019 1315m: epoch 46 [### ] 19% loss = 0.021 1315m: epoch 46 [#### ] 20% loss = 0.020 1316m: epoch 46 [#### ] 20% loss = 0.021 1316m: epoch 46 [#### ] 20% loss = 0.019 1316m: epoch 46 [#### ] 21% loss = 0.020 1316m: epoch 46 [#### ] 21% loss = 0.019 1316m: epoch 46 [#### ] 22% loss = 0.020 1316m: epoch 46 [#### ] 22% loss = 0.021 1316m: epoch 46 [#### ] 22% loss = 0.020 1316m: epoch 46 [#### ] 23% loss = 0.020 1316m: epoch 46 [#### ] 23% loss = 0.019 1317m: epoch 46 [#### ] 23% loss = 0.020 1317m: epoch 46 [#### ] 24% loss = 0.018 1317m: epoch 46 [#### ] 24% loss = 0.020 1317m: epoch 46 [#### ] 24% loss = 0.021 1317m: epoch 46 [##### ] 25% loss = 0.020 1317m: epoch 46 [##### ] 25% loss = 0.022 1317m: epoch 46 [##### ] 25% loss = 0.020 1317m: epoch 46 [##### ] 26% loss = 0.019 1317m: epoch 46 [##### ] 26% loss = 0.020 1318m: epoch 46 [##### ] 26% loss = 0.021 1318m: epoch 46 [##### ] 27% loss = 0.021 1318m: epoch 46 [##### ] 27% loss = 0.019 1318m: epoch 46 [##### ] 27% loss = 0.021 1318m: epoch 46 [##### ] 28% loss = 0.019 1318m: epoch 46 [##### ] 28% loss = 0.021 1318m: epoch 46 [##### ] 28% loss = 0.022 1318m: epoch 46 [##### ] 29% loss = 0.021 1318m: epoch 46 [##### ] 29% loss = 0.022 1319m: epoch 46 [##### ] 29% loss = 0.021 1319m: epoch 46 [###### ] 30% loss = 0.022 1319m: epoch 46 [###### ] 30% loss = 0.021 1319m: epoch 46 [###### ] 30% loss = 0.021 1319m: epoch 46 [###### ] 31% loss = 0.021 1319m: epoch 46 [###### ] 31% loss = 0.021 1319m: epoch 46 [###### ] 32% loss = 0.022 1319m: epoch 46 [###### ] 32% loss = 0.022 1319m: epoch 46 [###### ] 32% loss = 0.023 1320m: epoch 46 [###### ] 33% loss = 0.020 1320m: epoch 46 [###### ] 33% loss = 0.022 1320m: epoch 46 [###### ] 33% loss = 0.021 1320m: epoch 46 [###### ] 34% loss = 0.019 1320m: epoch 46 [###### ] 34% loss = 0.020 1320m: epoch 46 [###### ] 34% loss = 0.021 1320m: epoch 46 [####### ] 35% loss = 0.020 1320m: epoch 46 [####### ] 35% loss = 0.020 1320m: epoch 46 [####### ] 35% loss = 0.020 1321m: epoch 46 [####### ] 36% loss = 0.022 1321m: epoch 46 [####### ] 36% loss = 0.020 1321m: epoch 46 [####### ] 36% loss = 0.021 1321m: epoch 46 [####### ] 37% loss = 0.021 1321m: epoch 46 [####### ] 37% loss = 0.021 1321m: epoch 46 [####### ] 37% loss = 0.020 1321m: epoch 46 [####### ] 38% loss = 0.021 1321m: epoch 46 [####### ] 38% loss = 0.021 1321m: epoch 46 [####### ] 38% loss = 0.020 1322m: epoch 46 [####### ] 39% loss = 0.020 1322m: epoch 46 [####### ] 39% loss = 0.022 1322m: epoch 46 [####### ] 39% loss = 0.019 1322m: epoch 46 [######## ] 40% loss = 0.021 1322m: epoch 46 [######## ] 40% loss = 0.021 1322m: epoch 46 [######## ] 40% loss = 0.023 1322m: epoch 46 [######## ] 41% loss = 0.022 1322m: epoch 46 [######## ] 41% loss = 0.021 1322m: epoch 46 [######## ] 41% loss = 0.020 1323m: epoch 46 [######## ] 42% loss = 0.021 1323m: epoch 46 [######## ] 42% loss = 0.020 1323m: epoch 46 [######## ] 43% loss = 0.021 1323m: epoch 46 [######## ] 43% loss = 0.020 1323m: epoch 46 [######## ] 43% loss = 0.021 1323m: epoch 46 [######## ] 44% loss = 0.020 1323m: epoch 46 [######## ] 44% loss = 0.020 1323m: epoch 46 [######## ] 44% loss = 0.020 1323m: epoch 46 [######### ] 45% loss = 0.021 1324m: epoch 46 [######### ] 45% loss = 0.019 1324m: epoch 46 [######### ] 45% loss = 0.021 1324m: epoch 46 [######### ] 46% loss = 0.021 1324m: epoch 46 [######### ] 46% loss = 0.021 1324m: epoch 46 [######### ] 46% loss = 0.020 1324m: epoch 46 [######### ] 47% loss = 0.019 1324m: epoch 46 [######### ] 47% loss = 0.022 1324m: epoch 46 [######### ] 47% loss = 0.022 1324m: epoch 46 [######### ] 48% loss = 0.020 1324m: epoch 46 [######### ] 48% loss = 0.021 1325m: epoch 46 [######### ] 48% loss = 0.023 1325m: epoch 46 [######### ] 49% loss = 0.020 1325m: epoch 46 [######### ] 49% loss = 0.021 1325m: epoch 46 [######### ] 49% loss = 0.022 1325m: epoch 46 [########## ] 50% loss = 0.020 1325m: epoch 46 [########## ] 50% loss = 0.021 1325m: epoch 46 [########## ] 50% loss = 0.022 1325m: epoch 46 [########## ] 51% loss = 0.022 1325m: epoch 46 [########## ] 51% loss = 0.020 1326m: epoch 46 [########## ] 51% loss = 0.020 1326m: epoch 46 [########## ] 52% loss = 0.021 1326m: epoch 46 [########## ] 52% loss = 0.020 1326m: epoch 46 [########## ] 53% loss = 0.021 1326m: epoch 46 [########## ] 53% loss = 0.022 1326m: epoch 46 [########## ] 53% loss = 0.022 1326m: epoch 46 [########## ] 54% loss = 0.021 1326m: epoch 46 [########## ] 54% loss = 0.021 1326m: epoch 46 [########## ] 54% loss = 0.023 1327m: epoch 46 [########### ] 55% loss = 0.021 1327m: epoch 46 [########### ] 55% loss = 0.021 1327m: epoch 46 [########### ] 55% loss = 0.021 1327m: epoch 46 [########### ] 56% loss = 0.023 1327m: epoch 46 [########### ] 56% loss = 0.021 1327m: epoch 46 [########### ] 56% loss = 0.021 1327m: epoch 46 [########### ] 57% loss = 0.020 1327m: epoch 46 [########### ] 57% loss = 0.022 1327m: epoch 46 [########### ] 57% loss = 0.020 1328m: epoch 46 [########### ] 58% loss = 0.022 1328m: epoch 46 [########### ] 58% loss = 0.021 1328m: epoch 46 [########### ] 58% loss = 0.021 1328m: epoch 46 [########### ] 59% loss = 0.023 1328m: epoch 46 [########### ] 59% loss = 0.021 1328m: epoch 46 [########### ] 59% loss = 0.021 1328m: epoch 46 [############ ] 60% loss = 0.020 1328m: epoch 46 [############ ] 60% loss = 0.021 1328m: epoch 46 [############ ] 60% loss = 0.023 1328m: epoch 46 [############ ] 61% loss = 0.020 1329m: epoch 46 [############ ] 61% loss = 0.021 1329m: epoch 46 [############ ] 61% loss = 0.021 1329m: epoch 46 [############ ] 62% loss = 0.022 1329m: epoch 46 [############ ] 62% loss = 0.020 1329m: epoch 46 [############ ] 62% loss = 0.022 1329m: epoch 46 [############ ] 63% loss = 0.023 1329m: epoch 46 [############ ] 63% loss = 0.020 1329m: epoch 46 [############ ] 64% loss = 0.021 1329m: epoch 46 [############ ] 64% loss = 0.022 1330m: epoch 46 [############ ] 64% loss = 0.021 1330m: epoch 46 [############# ] 65% loss = 0.022 1330m: epoch 46 [############# ] 65% loss = 0.021 1330m: epoch 46 [############# ] 65% loss = 0.020 1330m: epoch 46 [############# ] 66% loss = 0.024 1330m: epoch 46 [############# ] 66% loss = 0.022 1330m: epoch 46 [############# ] 66% loss = 0.021 1330m: epoch 46 [############# ] 67% loss = 0.021 1330m: epoch 46 [############# ] 67% loss = 0.024 1331m: epoch 46 [############# ] 67% loss = 0.023 1331m: epoch 46 [############# ] 68% loss = 0.020 1331m: epoch 46 [############# ] 68% loss = 0.022 1331m: epoch 46 [############# ] 68% loss = 0.020 1331m: epoch 46 [############# ] 69% loss = 0.021 1331m: epoch 46 [############# ] 69% loss = 0.022 1331m: epoch 46 [############# ] 69% loss = 0.022 1331m: epoch 46 [############## ] 70% loss = 0.021 1331m: epoch 46 [############## ] 70% loss = 0.023 1331m: epoch 46 [############## ] 70% loss = 0.020 1332m: epoch 46 [############## ] 71% loss = 0.023 1332m: epoch 46 [############## ] 71% loss = 0.021 1332m: epoch 46 [############## ] 71% loss = 0.023 1332m: epoch 46 [############## ] 72% loss = 0.020 1332m: epoch 46 [############## ] 72% loss = 0.023 1332m: epoch 46 [############## ] 72% loss = 0.022 1332m: epoch 46 [############## ] 73% loss = 0.022 1332m: epoch 46 [############## ] 73% loss = 0.020 1332m: epoch 46 [############## ] 74% loss = 0.021 1333m: epoch 46 [############## ] 74% loss = 0.022 1333m: epoch 46 [############## ] 74% loss = 0.022 1333m: epoch 46 [############### ] 75% loss = 0.021 1333m: epoch 46 [############### ] 75% loss = 0.022 1333m: epoch 46 [############### ] 75% loss = 0.022 1333m: epoch 46 [############### ] 76% loss = 0.021 1333m: epoch 46 [############### ] 76% loss = 0.021 1333m: epoch 46 [############### ] 76% loss = 0.021 1333m: epoch 46 [############### ] 77% loss = 0.023 1334m: epoch 46 [############### ] 77% loss = 0.022 1334m: epoch 46 [############### ] 77% loss = 0.021 1334m: epoch 46 [############### ] 78% loss = 0.021 1334m: epoch 46 [############### ] 78% loss = 0.022 1334m: epoch 46 [############### ] 78% loss = 0.023 1334m: epoch 46 [############### ] 79% loss = 0.021 1334m: epoch 46 [############### ] 79% loss = 0.022 1334m: epoch 46 [############### ] 79% loss = 0.023 1334m: epoch 46 [################ ] 80% loss = 0.023 1335m: epoch 46 [################ ] 80% loss = 0.021 1335m: epoch 46 [################ ] 80% loss = 0.021 1335m: epoch 46 [################ ] 81% loss = 0.021 1335m: epoch 46 [################ ] 81% loss = 0.022 1335m: epoch 46 [################ ] 81% loss = 0.019 1335m: epoch 46 [################ ] 82% loss = 0.022 1335m: epoch 46 [################ ] 82% loss = 0.020 1335m: epoch 46 [################ ] 82% loss = 0.022 1335m: epoch 46 [################ ] 83% loss = 0.022 1336m: epoch 46 [################ ] 83% loss = 0.022 1336m: epoch 46 [################ ] 83% loss = 0.021 1336m: epoch 46 [################ ] 84% loss = 0.023 1336m: epoch 46 [################ ] 84% loss = 0.020 1336m: epoch 46 [################# ] 85% loss = 0.022 1336m: epoch 46 [################# ] 85% loss = 0.020 1336m: epoch 46 [################# ] 85% loss = 0.021 1336m: epoch 46 [################# ] 86% loss = 0.021 1336m: epoch 46 [################# ] 86% loss = 0.021 1337m: epoch 46 [################# ] 86% loss = 0.021 1337m: epoch 46 [################# ] 87% loss = 0.021 1337m: epoch 46 [################# ] 87% loss = 0.021 1337m: epoch 46 [################# ] 87% loss = 0.021 1337m: epoch 46 [################# ] 88% loss = 0.021 1337m: epoch 46 [################# ] 88% loss = 0.021 1337m: epoch 46 [################# ] 88% loss = 0.021 1337m: epoch 46 [################# ] 89% loss = 0.022 1337m: epoch 46 [################# ] 89% loss = 0.021 1338m: epoch 46 [################# ] 89% loss = 0.021 1338m: epoch 46 [################## ] 90% loss = 0.021 1338m: epoch 46 [################## ] 90% loss = 0.021 1338m: epoch 46 [################## ] 90% loss = 0.021 1338m: epoch 46 [################## ] 91% loss = 0.022 1338m: epoch 46 [################## ] 91% loss = 0.022 1338m: epoch 46 [################## ] 91% loss = 0.023 1338m: epoch 46 [################## ] 92% loss = 0.020 1338m: epoch 46 [################## ] 92% loss = 0.023 1339m: epoch 46 [################## ] 92% loss = 0.022 1339m: epoch 46 [################## ] 93% loss = 0.021 1339m: epoch 46 [################## ] 93% loss = 0.022 1339m: epoch 46 [################## ] 93% loss = 0.020 1339m: epoch 46 [################## ] 94% loss = 0.023 1339m: epoch 46 [################## ] 94% loss = 0.022 1339m: epoch 46 [################## ] 94% loss = 0.022 1339m: epoch 46 [################### ] 95% loss = 0.022 1339m: epoch 46 [################### ] 95% loss = 0.024 1340m: epoch 46 [################### ] 96% loss = 0.021 1340m: epoch 46 [################### ] 96% loss = 0.021 1340m: epoch 46 [################### ] 96% loss = 0.024 1340m: epoch 46 [################### ] 97% loss = 0.022 1340m: epoch 46 [################### ] 97% loss = 0.023 1340m: epoch 46 [################### ] 97% loss = 0.022 1340m: epoch 46 [################### ] 98% loss = 0.020 1340m: epoch 46 [################### ] 98% loss = 0.023 1340m: epoch 46 [################### ] 98% loss = 0.020 1341m: epoch 46 [################### ] 99% loss = 0.022 1341m: epoch 46 [################### ] 99% loss = 0.020 1341m: epoch 46 [################### ] 99% loss = 0.0211341m: epoch 46 [####################] 100% loss = 0.021 1309m: epoch 46 [ ] 0% loss = ... 1309m: epoch 46 [ ] 0% loss = 0.018 1309m: epoch 46 [ ] 0% loss = 0.018 1309m: epoch 46 [ ] 1% loss = 0.020 1309m: epoch 46 [ ] 1% loss = 0.019 1309m: epoch 46 [ ] 1% loss = 0.019 1310m: epoch 46 [ ] 2% loss = 0.019 1310m: epoch 46 [ ] 2% loss = 0.020 1310m: epoch 46 [ ] 2% loss = 0.021 1310m: epoch 46 [ ] 3% loss = 0.020 1310m: epoch 46 [ ] 3% loss = 0.019 1310m: epoch 46 [ ] 3% loss = 0.021 1310m: epoch 46 [ ] 4% loss = 0.020 1310m: epoch 46 [ ] 4% loss = 0.020 1310m: epoch 46 [ ] 4% loss = 0.020 1311m: epoch 46 [# ] 5% loss = 0.020 1311m: epoch 46 [# ] 5% loss = 0.020 1311m: epoch 46 [# ] 5% loss = 0.020 1311m: epoch 46 [# ] 6% loss = 0.019 1311m: epoch 46 [# ] 6% loss = 0.020 1311m: epoch 46 [# ] 6% loss = 0.021 1311m: epoch 46 [# ] 7% loss = 0.020 1311m: epoch 46 [# ] 7% loss = 0.020 1311m: epoch 46 [# ] 7% loss = 0.019 1312m: epoch 46 [# ] 8% loss = 0.019 1312m: epoch 46 [# ] 8% loss = 0.020 1312m: epoch 46 [# ] 8% loss = 0.022 1312m: epoch 46 [# ] 9% loss = 0.019 1312m: epoch 46 [# ] 9% loss = 0.020 1312m: epoch 46 [# ] 9% loss = 0.020 1312m: epoch 46 [## ] 10% loss = 0.020 1312m: epoch 46 [## ] 10% loss = 0.020 1312m: epoch 46 [## ] 11% loss = 0.020 1313m: epoch 46 [## ] 11% loss = 0.019 1313m: epoch 46 [## ] 11% loss = 0.019 1313m: epoch 46 [## ] 12% loss = 0.020 1313m: epoch 46 [## ] 12% loss = 0.020 1313m: epoch 46 [## ] 12% loss = 0.020 1313m: epoch 46 [## ] 13% loss = 0.020 1313m: epoch 46 [## ] 13% loss = 0.019 1313m: epoch 46 [## ] 13% loss = 0.021 1313m: epoch 46 [## ] 14% loss = 0.021 1314m: epoch 46 [## ] 14% loss = 0.020 1314m: epoch 46 [## ] 14% loss = 0.019 1314m: epoch 46 [### ] 15% loss = 0.019 1314m: epoch 46 [### ] 15% loss = 0.020 1314m: epoch 46 [### ] 15% loss = 0.019 1314m: epoch 46 [### ] 16% loss = 0.022 1314m: epoch 46 [### ] 16% loss = 0.020 1314m: epoch 46 [### ] 16% loss = 0.021 1314m: epoch 46 [### ] 17% loss = 0.020 1315m: epoch 46 [### ] 17% loss = 0.020 1315m: epoch 46 [### ] 17% loss = 0.019 1315m: epoch 46 [### ] 18% loss = 0.020 1315m: epoch 46 [### ] 18% loss = 0.021 1315m: epoch 46 [### ] 18% loss = 0.021 1315m: epoch 46 [### ] 19% loss = 0.021 1315m: epoch 46 [### ] 19% loss = 0.019 1315m: epoch 46 [### ] 19% loss = 0.021 1315m: epoch 46 [#### ] 20% loss = 0.020 1316m: epoch 46 [#### ] 20% loss = 0.021 1316m: epoch 46 [#### ] 20% loss = 0.019 1316m: epoch 46 [#### ] 21% loss = 0.020 1316m: epoch 46 [#### ] 21% loss = 0.019 1316m: epoch 46 [#### ] 22% loss = 0.020 1316m: epoch 46 [#### ] 22% loss = 0.021 1316m: epoch 46 [#### ] 22% loss = 0.020 1316m: epoch 46 [#### ] 23% loss = 0.020 1316m: epoch 46 [#### ] 23% loss = 0.019 1317m: epoch 46 [#### ] 23% loss = 0.020 1317m: epoch 46 [#### ] 24% loss = 0.018 1317m: epoch 46 [#### ] 24% loss = 0.020 1317m: epoch 46 [#### ] 24% loss = 0.021 1317m: epoch 46 [##### ] 25% loss = 0.020 1317m: epoch 46 [##### ] 25% loss = 0.022 1317m: epoch 46 [##### ] 25% loss = 0.020 1317m: epoch 46 [##### ] 26% loss = 0.019 1317m: epoch 46 [##### ] 26% loss = 0.020 1318m: epoch 46 [##### ] 26% loss = 0.021 1318m: epoch 46 [##### ] 27% loss = 0.021 1318m: epoch 46 [##### ] 27% loss = 0.019 1318m: epoch 46 [##### ] 27% loss = 0.021 1318m: epoch 46 [##### ] 28% loss = 0.019 1318m: epoch 46 [##### ] 28% loss = 0.021 1318m: epoch 46 [##### ] 28% loss = 0.022 1318m: epoch 46 [##### ] 29% loss = 0.021 1318m: epoch 46 [##### ] 29% loss = 0.022 1319m: epoch 46 [##### ] 29% loss = 0.021 1319m: epoch 46 [###### ] 30% loss = 0.022 1319m: epoch 46 [###### ] 30% loss = 0.021 1319m: epoch 46 [###### ] 30% loss = 0.021 1319m: epoch 46 [###### ] 31% loss = 0.021 1319m: epoch 46 [###### ] 31% loss = 0.021 1319m: epoch 46 [###### ] 32% loss = 0.022 1319m: epoch 46 [###### ] 32% loss = 0.022 1319m: epoch 46 [###### ] 32% loss = 0.023 1320m: epoch 46 [###### ] 33% loss = 0.020 1320m: epoch 46 [###### ] 33% loss = 0.022 1320m: epoch 46 [###### ] 33% loss = 0.021 1320m: epoch 46 [###### ] 34% loss = 0.019 1320m: epoch 46 [###### ] 34% loss = 0.020 1320m: epoch 46 [###### ] 34% loss = 0.021 1320m: epoch 46 [####### ] 35% loss = 0.020 1320m: epoch 46 [####### ] 35% loss = 0.020 1320m: epoch 46 [####### ] 35% loss = 0.020 1321m: epoch 46 [####### ] 36% loss = 0.022 1321m: epoch 46 [####### ] 36% loss = 0.020 1321m: epoch 46 [####### ] 36% loss = 0.021 1321m: epoch 46 [####### ] 37% loss = 0.021 1321m: epoch 46 [####### ] 37% loss = 0.021 1321m: epoch 46 [####### ] 37% loss = 0.020 1321m: epoch 46 [####### ] 38% loss = 0.021 1321m: epoch 46 [####### ] 38% loss = 0.021 1321m: epoch 46 [####### ] 38% loss = 0.020 1322m: epoch 46 [####### ] 39% loss = 0.020 1322m: epoch 46 [####### ] 39% loss = 0.022 1322m: epoch 46 [####### ] 39% loss = 0.019 1322m: epoch 46 [######## ] 40% loss = 0.021 1322m: epoch 46 [######## ] 40% loss = 0.021 1322m: epoch 46 [######## ] 40% loss = 0.023 1322m: epoch 46 [######## ] 41% loss = 0.022 1322m: epoch 46 [######## ] 41% loss = 0.021 1322m: epoch 46 [######## ] 41% loss = 0.020 1323m: epoch 46 [######## ] 42% loss = 0.021 1323m: epoch 46 [######## ] 42% loss = 0.020 1323m: epoch 46 [######## ] 43% loss = 0.021 1323m: epoch 46 [######## ] 43% loss = 0.020 1323m: epoch 46 [######## ] 43% loss = 0.021 1323m: epoch 46 [######## ] 44% loss = 0.020 1323m: epoch 46 [######## ] 44% loss = 0.020 1323m: epoch 46 [######## ] 44% loss = 0.020 1323m: epoch 46 [######### ] 45% loss = 0.021 1324m: epoch 46 [######### ] 45% loss = 0.019 1324m: epoch 46 [######### ] 45% loss = 0.021 1324m: epoch 46 [######### ] 46% loss = 0.021 1324m: epoch 46 [######### ] 46% loss = 0.021 1324m: epoch 46 [######### ] 46% loss = 0.020 1324m: epoch 46 [######### ] 47% loss = 0.019 1324m: epoch 46 [######### ] 47% loss = 0.022 1324m: epoch 46 [######### ] 47% loss = 0.022 1324m: epoch 46 [######### ] 48% loss = 0.020 1324m: epoch 46 [######### ] 48% loss = 0.021 1325m: epoch 46 [######### ] 48% loss = 0.023 1325m: epoch 46 [######### ] 49% loss = 0.020 1325m: epoch 46 [######### ] 49% loss = 0.021 1325m: epoch 46 [######### ] 49% loss = 0.022 1325m: epoch 46 [########## ] 50% loss = 0.020 1325m: epoch 46 [########## ] 50% loss = 0.021 1325m: epoch 46 [########## ] 50% loss = 0.022 1325m: epoch 46 [########## ] 51% loss = 0.022 1325m: epoch 46 [########## ] 51% loss = 0.020 1326m: epoch 46 [########## ] 51% loss = 0.020 1326m: epoch 46 [########## ] 52% loss = 0.021 1326m: epoch 46 [########## ] 52% loss = 0.020 1326m: epoch 46 [########## ] 53% loss = 0.021 1326m: epoch 46 [########## ] 53% loss = 0.022 1326m: epoch 46 [########## ] 53% loss = 0.022 1326m: epoch 46 [########## ] 54% loss = 0.021 1326m: epoch 46 [########## ] 54% loss = 0.021 1326m: epoch 46 [########## ] 54% loss = 0.023 1327m: epoch 46 [########### ] 55% loss = 0.021 1327m: epoch 46 [########### ] 55% loss = 0.021 1327m: epoch 46 [########### ] 55% loss = 0.021 1327m: epoch 46 [########### ] 56% loss = 0.023 1327m: epoch 46 [########### ] 56% loss = 0.021 1327m: epoch 46 [########### ] 56% loss = 0.021 1327m: epoch 46 [########### ] 57% loss = 0.020 1327m: epoch 46 [########### ] 57% loss = 0.022 1327m: epoch 46 [########### ] 57% loss = 0.020 1328m: epoch 46 [########### ] 58% loss = 0.022 1328m: epoch 46 [########### ] 58% loss = 0.021 1328m: epoch 46 [########### ] 58% loss = 0.021 1328m: epoch 46 [########### ] 59% loss = 0.023 1328m: epoch 46 [########### ] 59% loss = 0.021 1328m: epoch 46 [########### ] 59% loss = 0.021 1328m: epoch 46 [############ ] 60% loss = 0.020 1328m: epoch 46 [############ ] 60% loss = 0.021 1328m: epoch 46 [############ ] 60% loss = 0.023 1328m: epoch 46 [############ ] 61% loss = 0.020 1329m: epoch 46 [############ ] 61% loss = 0.021 1329m: epoch 46 [############ ] 61% loss = 0.021 1329m: epoch 46 [############ ] 62% loss = 0.022 1329m: epoch 46 [############ ] 62% loss = 0.020 1329m: epoch 46 [############ ] 62% loss = 0.022 1329m: epoch 46 [############ ] 63% loss = 0.023 1329m: epoch 46 [############ ] 63% loss = 0.020 1329m: epoch 46 [############ ] 64% loss = 0.021 1329m: epoch 46 [############ ] 64% loss = 0.022 1330m: epoch 46 [############ ] 64% loss = 0.021 1330m: epoch 46 [############# ] 65% loss = 0.022 1330m: epoch 46 [############# ] 65% loss = 0.021 1330m: epoch 46 [############# ] 65% loss = 0.020 1330m: epoch 46 [############# ] 66% loss = 0.024 1330m: epoch 46 [############# ] 66% loss = 0.022 1330m: epoch 46 [############# ] 66% loss = 0.021 1330m: epoch 46 [############# ] 67% loss = 0.021 1330m: epoch 46 [############# ] 67% loss = 0.024 1331m: epoch 46 [############# ] 67% loss = 0.023 1331m: epoch 46 [############# ] 68% loss = 0.020 1331m: epoch 46 [############# ] 68% loss = 0.022 1331m: epoch 46 [############# ] 68% loss = 0.020 1331m: epoch 46 [############# ] 69% loss = 0.021 1331m: epoch 46 [############# ] 69% loss = 0.022 1331m: epoch 46 [############# ] 69% loss = 0.022 1331m: epoch 46 [############## ] 70% loss = 0.021 1331m: epoch 46 [############## ] 70% loss = 0.023 1331m: epoch 46 [############## ] 70% loss = 0.020 1332m: epoch 46 [############## ] 71% loss = 0.023 1332m: epoch 46 [############## ] 71% loss = 0.021 1332m: epoch 46 [############## ] 71% loss = 0.023 1332m: epoch 46 [############## ] 72% loss = 0.020 1332m: epoch 46 [############## ] 72% loss = 0.023 1332m: epoch 46 [############## ] 72% loss = 0.022 1332m: epoch 46 [############## ] 73% loss = 0.022 1332m: epoch 46 [############## ] 73% loss = 0.020 1332m: epoch 46 [############## ] 74% loss = 0.021 1333m: epoch 46 [############## ] 74% loss = 0.022 1333m: epoch 46 [############## ] 74% loss = 0.022 1333m: epoch 46 [############### ] 75% loss = 0.021 1333m: epoch 46 [############### ] 75% loss = 0.022 1333m: epoch 46 [############### ] 75% loss = 0.022 1333m: epoch 46 [############### ] 76% loss = 0.021 1333m: epoch 46 [############### ] 76% loss = 0.021 1333m: epoch 46 [############### ] 76% loss = 0.021 1333m: epoch 46 [############### ] 77% loss = 0.023 1334m: epoch 46 [############### ] 77% loss = 0.022 1334m: epoch 46 [############### ] 77% loss = 0.021 1334m: epoch 46 [############### ] 78% loss = 0.021 1334m: epoch 46 [############### ] 78% loss = 0.022 1334m: epoch 46 [############### ] 78% loss = 0.023 1334m: epoch 46 [############### ] 79% loss = 0.021 1334m: epoch 46 [############### ] 79% loss = 0.022 1334m: epoch 46 [############### ] 79% loss = 0.023 1334m: epoch 46 [################ ] 80% loss = 0.023 1335m: epoch 46 [################ ] 80% loss = 0.021 1335m: epoch 46 [################ ] 80% loss = 0.021 1335m: epoch 46 [################ ] 81% loss = 0.021 1335m: epoch 46 [################ ] 81% loss = 0.022 1335m: epoch 46 [################ ] 81% loss = 0.019 1335m: epoch 46 [################ ] 82% loss = 0.022 1335m: epoch 46 [################ ] 82% loss = 0.020 1335m: epoch 46 [################ ] 82% loss = 0.022 1335m: epoch 46 [################ ] 83% loss = 0.022 1336m: epoch 46 [################ ] 83% loss = 0.022 1336m: epoch 46 [################ ] 83% loss = 0.021 1336m: epoch 46 [################ ] 84% loss = 0.023 1336m: epoch 46 [################ ] 84% loss = 0.020 1336m: epoch 46 [################# ] 85% loss = 0.022 1336m: epoch 46 [################# ] 85% loss = 0.020 1336m: epoch 46 [################# ] 85% loss = 0.021 1336m: epoch 46 [################# ] 86% loss = 0.021 1336m: epoch 46 [################# ] 86% loss = 0.021 1337m: epoch 46 [################# ] 86% loss = 0.021 1337m: epoch 46 [################# ] 87% loss = 0.021 1337m: epoch 46 [################# ] 87% loss = 0.021 1337m: epoch 46 [################# ] 87% loss = 0.021 1337m: epoch 46 [################# ] 88% loss = 0.021 1337m: epoch 46 [################# ] 88% loss = 0.021 1337m: epoch 46 [################# ] 88% loss = 0.021 1337m: epoch 46 [################# ] 89% loss = 0.022 1337m: epoch 46 [################# ] 89% loss = 0.021 1338m: epoch 46 [################# ] 89% loss = 0.021 1338m: epoch 46 [################## ] 90% loss = 0.021 1338m: epoch 46 [################## ] 90% loss = 0.021 1338m: epoch 46 [################## ] 90% loss = 0.021 1338m: epoch 46 [################## ] 91% loss = 0.022 1338m: epoch 46 [################## ] 91% loss = 0.022 1338m: epoch 46 [################## ] 91% loss = 0.023 1338m: epoch 46 [################## ] 92% loss = 0.020 1338m: epoch 46 [################## ] 92% loss = 0.023 1339m: epoch 46 [################## ] 92% loss = 0.022 1339m: epoch 46 [################## ] 93% loss = 0.021 1339m: epoch 46 [################## ] 93% loss = 0.022 1339m: epoch 46 [################## ] 93% loss = 0.020 1339m: epoch 46 [################## ] 94% loss = 0.023 1339m: epoch 46 [################## ] 94% loss = 0.022 1339m: epoch 46 [################## ] 94% loss = 0.022 1339m: epoch 46 [################### ] 95% loss = 0.022 1339m: epoch 46 [################### ] 95% loss = 0.024 1340m: epoch 46 [################### ] 96% loss = 0.021 1340m: epoch 46 [################### ] 96% loss = 0.021 1340m: epoch 46 [################### ] 96% loss = 0.024 1340m: epoch 46 [################### ] 97% loss = 0.022 1340m: epoch 46 [################### ] 97% loss = 0.023 1340m: epoch 46 [################### ] 97% loss = 0.022 1340m: epoch 46 [################### ] 98% loss = 0.020 1340m: epoch 46 [################### ] 98% loss = 0.023 1340m: epoch 46 [################### ] 98% loss = 0.020 1341m: epoch 46 [################### ] 99% loss = 0.022 1341m: epoch 46 [################### ] 99% loss = 0.020 1341m: epoch 46 [################### ] 99% loss = 0.0211341m: epoch 46 [####################] 100% loss = 0.021
epoch 46 complete, loss = 0.021 epoch 46 complete, loss = 0.021
model saved as token_classification_split_3_46_0.021 model saved as token_classification_split_3_46_0.021
1341m: epoch 47 [ ] 0% loss = ... 1341m: epoch 47 [ ] 0% loss = 0.019 1341m: epoch 47 [ ] 0% loss = 0.019 1341m: epoch 47 [ ] 1% loss = 0.019 1341m: epoch 47 [ ] 1% loss = 0.022 1341m: epoch 47 [ ] 1% loss = 0.017 1342m: epoch 47 [ ] 2% loss = 0.021 1342m: epoch 47 [ ] 2% loss = 0.020 1342m: epoch 47 [ ] 2% loss = 0.020 1342m: epoch 47 [ ] 3% loss = 0.018 1342m: epoch 47 [ ] 3% loss = 0.020 1342m: epoch 47 [ ] 3% loss = 0.020 1342m: epoch 47 [ ] 4% loss = 0.020 1342m: epoch 47 [ ] 4% loss = 0.020 1342m: epoch 47 [ ] 4% loss = 0.019 1343m: epoch 47 [# ] 5% loss = 0.020 1343m: epoch 47 [# ] 5% loss = 0.019 1343m: epoch 47 [# ] 5% loss = 0.020 1343m: epoch 47 [# ] 6% loss = 0.019 1343m: epoch 47 [# ] 6% loss = 0.019 1343m: epoch 47 [# ] 6% loss = 0.019 1343m: epoch 47 [# ] 7% loss = 0.019 1343m: epoch 47 [# ] 7% loss = 0.019 1343m: epoch 47 [# ] 7% loss = 0.019 1344m: epoch 47 [# ] 8% loss = 0.019 1344m: epoch 47 [# ] 8% loss = 0.019 1344m: epoch 47 [# ] 8% loss = 0.020 1344m: epoch 47 [# ] 9% loss = 0.019 1344m: epoch 47 [# ] 9% loss = 0.020 1344m: epoch 47 [# ] 9% loss = 0.020 1344m: epoch 47 [## ] 10% loss = 0.020 1344m: epoch 47 [## ] 10% loss = 0.020 1344m: epoch 47 [## ] 11% loss = 0.019 1345m: epoch 47 [## ] 11% loss = 0.020 1345m: epoch 47 [## ] 11% loss = 0.020 1345m: epoch 47 [## ] 12% loss = 0.019 1345m: epoch 47 [## ] 12% loss = 0.020 1345m: epoch 47 [## ] 12% loss = 0.019 1345m: epoch 47 [## ] 13% loss = 0.020 1345m: epoch 47 [## ] 13% loss = 0.020 1345m: epoch 47 [## ] 13% loss = 0.019 1345m: epoch 47 [## ] 14% loss = 0.020 1345m: epoch 47 [## ] 14% loss = 0.019 1346m: epoch 47 [## ] 14% loss = 0.019 1346m: epoch 47 [### ] 15% loss = 0.021 1346m: epoch 47 [### ] 15% loss = 0.019 1346m: epoch 47 [### ] 15% loss = 0.020 1346m: epoch 47 [### ] 16% loss = 0.021 1346m: epoch 47 [### ] 16% loss = 0.021 1346m: epoch 47 [### ] 16% loss = 0.020 1346m: epoch 47 [### ] 17% loss = 0.021 1346m: epoch 47 [### ] 17% loss = 0.019 1347m: epoch 47 [### ] 17% loss = 0.021 1347m: epoch 47 [### ] 18% loss = 0.020 1347m: epoch 47 [### ] 18% loss = 0.022 1347m: epoch 47 [### ] 18% loss = 0.020 1347m: epoch 47 [### ] 19% loss = 0.021 1347m: epoch 47 [### ] 19% loss = 0.020 1347m: epoch 47 [### ] 19% loss = 0.020 1347m: epoch 47 [#### ] 20% loss = 0.020 1347m: epoch 47 [#### ] 20% loss = 0.020 1348m: epoch 47 [#### ] 20% loss = 0.019 1348m: epoch 47 [#### ] 21% loss = 0.020 1348m: epoch 47 [#### ] 21% loss = 0.020 1348m: epoch 47 [#### ] 22% loss = 0.019 1348m: epoch 47 [#### ] 22% loss = 0.020 1348m: epoch 47 [#### ] 22% loss = 0.019 1348m: epoch 47 [#### ] 23% loss = 0.019 1348m: epoch 47 [#### ] 23% loss = 0.021 1348m: epoch 47 [#### ] 23% loss = 0.021 1349m: epoch 47 [#### ] 24% loss = 0.021 1349m: epoch 47 [#### ] 24% loss = 0.019 1349m: epoch 47 [#### ] 24% loss = 0.021 1349m: epoch 47 [##### ] 25% loss = 0.019 1349m: epoch 47 [##### ] 25% loss = 0.021 1349m: epoch 47 [##### ] 25% loss = 0.021 1349m: epoch 47 [##### ] 26% loss = 0.022 1349m: epoch 47 [##### ] 26% loss = 0.021 1349m: epoch 47 [##### ] 26% loss = 0.021 1350m: epoch 47 [##### ] 27% loss = 0.019 1350m: epoch 47 [##### ] 27% loss = 0.020 1350m: epoch 47 [##### ] 27% loss = 0.022 1350m: epoch 47 [##### ] 28% loss = 0.021 1350m: epoch 47 [##### ] 28% loss = 0.021 1350m: epoch 47 [##### ] 28% loss = 0.021 1350m: epoch 47 [##### ] 29% loss = 0.021 1350m: epoch 47 [##### ] 29% loss = 0.021 1350m: epoch 47 [##### ] 29% loss = 0.020 1351m: epoch 47 [###### ] 30% loss = 0.020 1351m: epoch 47 [###### ] 30% loss = 0.020 1351m: epoch 47 [###### ] 30% loss = 0.019 1351m: epoch 47 [###### ] 31% loss = 0.022 1351m: epoch 47 [###### ] 31% loss = 0.020 1351m: epoch 47 [###### ] 32% loss = 0.020 1351m: epoch 47 [###### ] 32% loss = 0.020 1351m: epoch 47 [###### ] 32% loss = 0.022 1351m: epoch 47 [###### ] 33% loss = 0.019 1352m: epoch 47 [###### ] 33% loss = 0.020 1352m: epoch 47 [###### ] 33% loss = 0.019 1352m: epoch 47 [###### ] 34% loss = 0.021 1352m: epoch 47 [###### ] 34% loss = 0.021 1352m: epoch 47 [###### ] 34% loss = 0.022 1352m: epoch 47 [####### ] 35% loss = 0.020 1352m: epoch 47 [####### ] 35% loss = 0.020 1352m: epoch 47 [####### ] 35% loss = 0.021 1352m: epoch 47 [####### ] 36% loss = 0.019 1353m: epoch 47 [####### ] 36% loss = 0.021 1353m: epoch 47 [####### ] 36% loss = 0.021 1353m: epoch 47 [####### ] 37% loss = 0.020 1353m: epoch 47 [####### ] 37% loss = 0.021 1353m: epoch 47 [####### ] 37% loss = 0.020 1353m: epoch 47 [####### ] 38% loss = 0.020 1353m: epoch 47 [####### ] 38% loss = 0.021 1353m: epoch 47 [####### ] 38% loss = 0.022 1353m: epoch 47 [####### ] 39% loss = 0.020 1354m: epoch 47 [####### ] 39% loss = 0.021 1354m: epoch 47 [####### ] 39% loss = 0.021 1354m: epoch 47 [######## ] 40% loss = 0.022 1354m: epoch 47 [######## ] 40% loss = 0.020 1354m: epoch 47 [######## ] 40% loss = 0.020 1354m: epoch 47 [######## ] 41% loss = 0.021 1354m: epoch 47 [######## ] 41% loss = 0.020 1354m: epoch 47 [######## ] 41% loss = 0.020 1354m: epoch 47 [######## ] 42% loss = 0.021 1355m: epoch 47 [######## ] 42% loss = 0.021 1355m: epoch 47 [######## ] 43% loss = 0.020 1355m: epoch 47 [######## ] 43% loss = 0.020 1355m: epoch 47 [######## ] 43% loss = 0.021 1355m: epoch 47 [######## ] 44% loss = 0.020 1355m: epoch 47 [######## ] 44% loss = 0.021 1355m: epoch 47 [######## ] 44% loss = 0.021 1355m: epoch 47 [######### ] 45% loss = 0.021 1355m: epoch 47 [######### ] 45% loss = 0.022 1356m: epoch 47 [######### ] 45% loss = 0.021 1356m: epoch 47 [######### ] 46% loss = 0.023 1356m: epoch 47 [######### ] 46% loss = 0.021 1356m: epoch 47 [######### ] 46% loss = 0.022 1356m: epoch 47 [######### ] 47% loss = 0.021 1356m: epoch 47 [######### ] 47% loss = 0.020 1356m: epoch 47 [######### ] 47% loss = 0.021 1356m: epoch 47 [######### ] 48% loss = 0.021 1356m: epoch 47 [######### ] 48% loss = 0.021 1357m: epoch 47 [######### ] 48% loss = 0.019 1357m: epoch 47 [######### ] 49% loss = 0.020 1357m: epoch 47 [######### ] 49% loss = 0.020 1357m: epoch 47 [######### ] 49% loss = 0.019 1357m: epoch 47 [########## ] 50% loss = 0.023 1357m: epoch 47 [########## ] 50% loss = 0.019 1357m: epoch 47 [########## ] 50% loss = 0.021 1357m: epoch 47 [########## ] 51% loss = 0.021 1357m: epoch 47 [########## ] 51% loss = 0.021 1358m: epoch 47 [########## ] 51% loss = 0.021 1358m: epoch 47 [########## ] 52% loss = 0.019 1358m: epoch 47 [########## ] 52% loss = 0.023 1358m: epoch 47 [########## ] 53% loss = 0.021 1358m: epoch 47 [########## ] 53% loss = 0.022 1358m: epoch 47 [########## ] 53% loss = 0.021 1358m: epoch 47 [########## ] 54% loss = 0.022 1358m: epoch 47 [########## ] 54% loss = 0.022 1358m: epoch 47 [########## ] 54% loss = 0.021 1359m: epoch 47 [########### ] 55% loss = 0.022 1359m: epoch 47 [########### ] 55% loss = 0.021 1359m: epoch 47 [########### ] 55% loss = 0.021 1359m: epoch 47 [########### ] 56% loss = 0.020 1359m: epoch 47 [########### ] 56% loss = 0.020 1359m: epoch 47 [########### ] 56% loss = 0.022 1359m: epoch 47 [########### ] 57% loss = 0.020 1359m: epoch 47 [########### ] 57% loss = 0.021 1359m: epoch 47 [########### ] 57% loss = 0.022 1360m: epoch 47 [########### ] 58% loss = 0.021 1360m: epoch 47 [########### ] 58% loss = 0.021 1360m: epoch 47 [########### ] 58% loss = 0.020 1360m: epoch 47 [########### ] 59% loss = 0.020 1360m: epoch 47 [########### ] 59% loss = 0.021 1360m: epoch 47 [########### ] 59% loss = 0.023 1360m: epoch 47 [############ ] 60% loss = 0.021 1360m: epoch 47 [############ ] 60% loss = 0.020 1360m: epoch 47 [############ ] 60% loss = 0.021 1361m: epoch 47 [############ ] 61% loss = 0.022 1361m: epoch 47 [############ ] 61% loss = 0.021 1361m: epoch 47 [############ ] 61% loss = 0.021 1361m: epoch 47 [############ ] 62% loss = 0.020 1361m: epoch 47 [############ ] 62% loss = 0.020 1361m: epoch 47 [############ ] 62% loss = 0.021 1361m: epoch 47 [############ ] 63% loss = 0.020 1361m: epoch 47 [############ ] 63% loss = 0.023 1361m: epoch 47 [############ ] 64% loss = 0.019 1362m: epoch 47 [############ ] 64% loss = 0.019 1362m: epoch 47 [############ ] 64% loss = 0.023 1362m: epoch 47 [############# ] 65% loss = 0.021 1362m: epoch 47 [############# ] 65% loss = 0.020 1362m: epoch 47 [############# ] 65% loss = 0.021 1362m: epoch 47 [############# ] 66% loss = 0.023 1362m: epoch 47 [############# ] 66% loss = 0.021 1362m: epoch 47 [############# ] 66% loss = 0.022 1362m: epoch 47 [############# ] 67% loss = 0.022 1363m: epoch 47 [############# ] 67% loss = 0.022 1363m: epoch 47 [############# ] 67% loss = 0.021 1363m: epoch 47 [############# ] 68% loss = 0.021 1363m: epoch 47 [############# ] 68% loss = 0.022 1363m: epoch 47 [############# ] 68% loss = 0.021 1363m: epoch 47 [############# ] 69% loss = 0.022 1363m: epoch 47 [############# ] 69% loss = 0.022 1363m: epoch 47 [############# ] 69% loss = 0.021 1363m: epoch 47 [############## ] 70% loss = 0.022 1364m: epoch 47 [############## ] 70% loss = 0.020 1364m: epoch 47 [############## ] 70% loss = 0.022 1364m: epoch 47 [############## ] 71% loss = 0.020 1364m: epoch 47 [############## ] 71% loss = 0.021 1364m: epoch 47 [############## ] 71% loss = 0.021 1364m: epoch 47 [############## ] 72% loss = 0.021 1364m: epoch 47 [############## ] 72% loss = 0.021 1364m: epoch 47 [############## ] 72% loss = 0.022 1364m: epoch 47 [############## ] 73% loss = 0.020 1365m: epoch 47 [############## ] 73% loss = 0.021 1365m: epoch 47 [############## ] 74% loss = 0.022 1365m: epoch 47 [############## ] 74% loss = 0.020 1365m: epoch 47 [############## ] 74% loss = 0.021 1365m: epoch 47 [############### ] 75% loss = 0.019 1365m: epoch 47 [############### ] 75% loss = 0.021 1365m: epoch 47 [############### ] 75% loss = 0.021 1365m: epoch 47 [############### ] 76% loss = 0.020 1365m: epoch 47 [############### ] 76% loss = 0.022 1366m: epoch 47 [############### ] 76% loss = 0.022 1366m: epoch 47 [############### ] 77% loss = 0.021 1366m: epoch 47 [############### ] 77% loss = 0.022 1366m: epoch 47 [############### ] 77% loss = 0.023 1366m: epoch 47 [############### ] 78% loss = 0.022 1366m: epoch 47 [############### ] 78% loss = 0.020 1366m: epoch 47 [############### ] 78% loss = 0.021 1366m: epoch 47 [############### ] 79% loss = 0.023 1366m: epoch 47 [############### ] 79% loss = 0.021 1367m: epoch 47 [############### ] 79% loss = 0.020 1367m: epoch 47 [################ ] 80% loss = 0.022 1367m: epoch 47 [################ ] 80% loss = 0.021 1367m: epoch 47 [################ ] 80% loss = 0.021 1367m: epoch 47 [################ ] 81% loss = 0.022 1341m: epoch 47 [ ] 0% loss = ... 1341m: epoch 47 [ ] 0% loss = 0.019 1341m: epoch 47 [ ] 0% loss = 0.019 1341m: epoch 47 [ ] 1% loss = 0.019 1341m: epoch 47 [ ] 1% loss = 0.022 1341m: epoch 47 [ ] 1% loss = 0.017 1342m: epoch 47 [ ] 2% loss = 0.021 1342m: epoch 47 [ ] 2% loss = 0.020 1342m: epoch 47 [ ] 2% loss = 0.020 1342m: epoch 47 [ ] 3% loss = 0.018 1342m: epoch 47 [ ] 3% loss = 0.020 1342m: epoch 47 [ ] 3% loss = 0.020 1342m: epoch 47 [ ] 4% loss = 0.020 1342m: epoch 47 [ ] 4% loss = 0.020 1342m: epoch 47 [ ] 4% loss = 0.019 1343m: epoch 47 [# ] 5% loss = 0.020 1343m: epoch 47 [# ] 5% loss = 0.019 1343m: epoch 47 [# ] 5% loss = 0.020 1343m: epoch 47 [# ] 6% loss = 0.019 1343m: epoch 47 [# ] 6% loss = 0.019 1343m: epoch 47 [# ] 6% loss = 0.019 1343m: epoch 47 [# ] 7% loss = 0.019 1343m: epoch 47 [# ] 7% loss = 0.019 1343m: epoch 47 [# ] 7% loss = 0.019 1344m: epoch 47 [# ] 8% loss = 0.019 1344m: epoch 47 [# ] 8% loss = 0.019 1344m: epoch 47 [# ] 8% loss = 0.020 1344m: epoch 47 [# ] 9% loss = 0.019 1344m: epoch 47 [# ] 9% loss = 0.020 1344m: epoch 47 [# ] 9% loss = 0.020 1344m: epoch 47 [## ] 10% loss = 0.020 1344m: epoch 47 [## ] 10% loss = 0.020 1344m: epoch 47 [## ] 11% loss = 0.019 1345m: epoch 47 [## ] 11% loss = 0.020 1345m: epoch 47 [## ] 11% loss = 0.020 1345m: epoch 47 [## ] 12% loss = 0.019 1345m: epoch 47 [## ] 12% loss = 0.020 1345m: epoch 47 [## ] 12% loss = 0.019 1345m: epoch 47 [## ] 13% loss = 0.020 1345m: epoch 47 [## ] 13% loss = 0.020 1345m: epoch 47 [## ] 13% loss = 0.019 1345m: epoch 47 [## ] 14% loss = 0.020 1345m: epoch 47 [## ] 14% loss = 0.019 1346m: epoch 47 [## ] 14% loss = 0.019 1346m: epoch 47 [### ] 15% loss = 0.021 1346m: epoch 47 [### ] 15% loss = 0.019 1346m: epoch 47 [### ] 15% loss = 0.020 1346m: epoch 47 [### ] 16% loss = 0.021 1346m: epoch 47 [### ] 16% loss = 0.021 1346m: epoch 47 [### ] 16% loss = 0.020 1346m: epoch 47 [### ] 17% loss = 0.021 1346m: epoch 47 [### ] 17% loss = 0.019 1347m: epoch 47 [### ] 17% loss = 0.021 1347m: epoch 47 [### ] 18% loss = 0.020 1347m: epoch 47 [### ] 18% loss = 0.022 1347m: epoch 47 [### ] 18% loss = 0.020 1347m: epoch 47 [### ] 19% loss = 0.021 1347m: epoch 47 [### ] 19% loss = 0.020 1347m: epoch 47 [### ] 19% loss = 0.020 1347m: epoch 47 [#### ] 20% loss = 0.020 1347m: epoch 47 [#### ] 20% loss = 0.020 1348m: epoch 47 [#### ] 20% loss = 0.019 1348m: epoch 47 [#### ] 21% loss = 0.020 1348m: epoch 47 [#### ] 21% loss = 0.020 1348m: epoch 47 [#### ] 22% loss = 0.019 1348m: epoch 47 [#### ] 22% loss = 0.020 1348m: epoch 47 [#### ] 22% loss = 0.019 1348m: epoch 47 [#### ] 23% loss = 0.019 1348m: epoch 47 [#### ] 23% loss = 0.021 1348m: epoch 47 [#### ] 23% loss = 0.021 1349m: epoch 47 [#### ] 24% loss = 0.021 1349m: epoch 47 [#### ] 24% loss = 0.019 1349m: epoch 47 [#### ] 24% loss = 0.021 1349m: epoch 47 [##### ] 25% loss = 0.019 1349m: epoch 47 [##### ] 25% loss = 0.021 1349m: epoch 47 [##### ] 25% loss = 0.021 1349m: epoch 47 [##### ] 26% loss = 0.022 1349m: epoch 47 [##### ] 26% loss = 0.021 1349m: epoch 47 [##### ] 26% loss = 0.021 1350m: epoch 47 [##### ] 27% loss = 0.019 1350m: epoch 47 [##### ] 27% loss = 0.020 1350m: epoch 47 [##### ] 27% loss = 0.022 1350m: epoch 47 [##### ] 28% loss = 0.021 1350m: epoch 47 [##### ] 28% loss = 0.021 1350m: epoch 47 [##### ] 28% loss = 0.021 1350m: epoch 47 [##### ] 29% loss = 0.021 1350m: epoch 47 [##### ] 29% loss = 0.021 1350m: epoch 47 [##### ] 29% loss = 0.020 1351m: epoch 47 [###### ] 30% loss = 0.020 1351m: epoch 47 [###### ] 30% loss = 0.020 1351m: epoch 47 [###### ] 30% loss = 0.019 1351m: epoch 47 [###### ] 31% loss = 0.022 1351m: epoch 47 [###### ] 31% loss = 0.020 1351m: epoch 47 [###### ] 32% loss = 0.020 1351m: epoch 47 [###### ] 32% loss = 0.020 1351m: epoch 47 [###### ] 32% loss = 0.022 1351m: epoch 47 [###### ] 33% loss = 0.019 1352m: epoch 47 [###### ] 33% loss = 0.020 1352m: epoch 47 [###### ] 33% loss = 0.019 1352m: epoch 47 [###### ] 34% loss = 0.021 1352m: epoch 47 [###### ] 34% loss = 0.021 1352m: epoch 47 [###### ] 34% loss = 0.022 1352m: epoch 47 [####### ] 35% loss = 0.020 1352m: epoch 47 [####### ] 35% loss = 0.020 1352m: epoch 47 [####### ] 35% loss = 0.021 1352m: epoch 47 [####### ] 36% loss = 0.019 1353m: epoch 47 [####### ] 36% loss = 0.021 1353m: epoch 47 [####### ] 36% loss = 0.021 1353m: epoch 47 [####### ] 37% loss = 0.020 1353m: epoch 47 [####### ] 37% loss = 0.021 1353m: epoch 47 [####### ] 37% loss = 0.020 1353m: epoch 47 [####### ] 38% loss = 0.020 1353m: epoch 47 [####### ] 38% loss = 0.021 1353m: epoch 47 [####### ] 38% loss = 0.022 1353m: epoch 47 [####### ] 39% loss = 0.020 1354m: epoch 47 [####### ] 39% loss = 0.021 1354m: epoch 47 [####### ] 39% loss = 0.021 1354m: epoch 47 [######## ] 40% loss = 0.022 1354m: epoch 47 [######## ] 40% loss = 0.020 1354m: epoch 47 [######## ] 40% loss = 0.020 1354m: epoch 47 [######## ] 41% loss = 0.021 1354m: epoch 47 [######## ] 41% loss = 0.020 1354m: epoch 47 [######## ] 41% loss = 0.020 1354m: epoch 47 [######## ] 42% loss = 0.021 1355m: epoch 47 [######## ] 42% loss = 0.021 1355m: epoch 47 [######## ] 43% loss = 0.020 1355m: epoch 47 [######## ] 43% loss = 0.020 1355m: epoch 47 [######## ] 43% loss = 0.021 1355m: epoch 47 [######## ] 44% loss = 0.020 1355m: epoch 47 [######## ] 44% loss = 0.021 1355m: epoch 47 [######## ] 44% loss = 0.021 1355m: epoch 47 [######### ] 45% loss = 0.021 1355m: epoch 47 [######### ] 45% loss = 0.022 1356m: epoch 47 [######### ] 45% loss = 0.021 1356m: epoch 47 [######### ] 46% loss = 0.023 1356m: epoch 47 [######### ] 46% loss = 0.021 1356m: epoch 47 [######### ] 46% loss = 0.022 1356m: epoch 47 [######### ] 47% loss = 0.021 1356m: epoch 47 [######### ] 47% loss = 0.020 1356m: epoch 47 [######### ] 47% loss = 0.021 1356m: epoch 47 [######### ] 48% loss = 0.021 1356m: epoch 47 [######### ] 48% loss = 0.021 1357m: epoch 47 [######### ] 48% loss = 0.019 1357m: epoch 47 [######### ] 49% loss = 0.020 1357m: epoch 47 [######### ] 49% loss = 0.020 1357m: epoch 47 [######### ] 49% loss = 0.019 1357m: epoch 47 [########## ] 50% loss = 0.023 1357m: epoch 47 [########## ] 50% loss = 0.019 1357m: epoch 47 [########## ] 50% loss = 0.021 1357m: epoch 47 [########## ] 51% loss = 0.021 1357m: epoch 47 [########## ] 51% loss = 0.021 1358m: epoch 47 [########## ] 51% loss = 0.021 1358m: epoch 47 [########## ] 52% loss = 0.019 1358m: epoch 47 [########## ] 52% loss = 0.023 1358m: epoch 47 [########## ] 53% loss = 0.021 1358m: epoch 47 [########## ] 53% loss = 0.022 1358m: epoch 47 [########## ] 53% loss = 0.021 1358m: epoch 47 [########## ] 54% loss = 0.022 1358m: epoch 47 [########## ] 54% loss = 0.022 1358m: epoch 47 [########## ] 54% loss = 0.021 1359m: epoch 47 [########### ] 55% loss = 0.022 1359m: epoch 47 [########### ] 55% loss = 0.021 1359m: epoch 47 [########### ] 55% loss = 0.021 1359m: epoch 47 [########### ] 56% loss = 0.020 1359m: epoch 47 [########### ] 56% loss = 0.020 1359m: epoch 47 [########### ] 56% loss = 0.022 1359m: epoch 47 [########### ] 57% loss = 0.020 1359m: epoch 47 [########### ] 57% loss = 0.021 1359m: epoch 47 [########### ] 57% loss = 0.022 1360m: epoch 47 [########### ] 58% loss = 0.021 1360m: epoch 47 [########### ] 58% loss = 0.021 1360m: epoch 47 [########### ] 58% loss = 0.020 1360m: epoch 47 [########### ] 59% loss = 0.020 1360m: epoch 47 [########### ] 59% loss = 0.021 1360m: epoch 47 [########### ] 59% loss = 0.023 1360m: epoch 47 [############ ] 60% loss = 0.021 1360m: epoch 47 [############ ] 60% loss = 0.020 1360m: epoch 47 [############ ] 60% loss = 0.021 1361m: epoch 47 [############ ] 61% loss = 0.022 1361m: epoch 47 [############ ] 61% loss = 0.021 1361m: epoch 47 [############ ] 61% loss = 0.021 1361m: epoch 47 [############ ] 62% loss = 0.020 1361m: epoch 47 [############ ] 62% loss = 0.020 1361m: epoch 47 [############ ] 62% loss = 0.021 1361m: epoch 47 [############ ] 63% loss = 0.020 1361m: epoch 47 [############ ] 63% loss = 0.023 1361m: epoch 47 [############ ] 64% loss = 0.019 1362m: epoch 47 [############ ] 64% loss = 0.019 1362m: epoch 47 [############ ] 64% loss = 0.023 1362m: epoch 47 [############# ] 65% loss = 0.021 1362m: epoch 47 [############# ] 65% loss = 0.020 1362m: epoch 47 [############# ] 65% loss = 0.021 1362m: epoch 47 [############# ] 66% loss = 0.023 1362m: epoch 47 [############# ] 66% loss = 0.021 1362m: epoch 47 [############# ] 66% loss = 0.022 1362m: epoch 47 [############# ] 67% loss = 0.022 1363m: epoch 47 [############# ] 67% loss = 0.022 1363m: epoch 47 [############# ] 67% loss = 0.021 1363m: epoch 47 [############# ] 68% loss = 0.021 1363m: epoch 47 [############# ] 68% loss = 0.022 1363m: epoch 47 [############# ] 68% loss = 0.021 1363m: epoch 47 [############# ] 69% loss = 0.022 1363m: epoch 47 [############# ] 69% loss = 0.022 1363m: epoch 47 [############# ] 69% loss = 0.021 1363m: epoch 47 [############## ] 70% loss = 0.022 1364m: epoch 47 [############## ] 70% loss = 0.020 1364m: epoch 47 [############## ] 70% loss = 0.022 1364m: epoch 47 [############## ] 71% loss = 0.020 1364m: epoch 47 [############## ] 71% loss = 0.021 1364m: epoch 47 [############## ] 71% loss = 0.021 1364m: epoch 47 [############## ] 72% loss = 0.021 1364m: epoch 47 [############## ] 72% loss = 0.021 1364m: epoch 47 [############## ] 72% loss = 0.022 1364m: epoch 47 [############## ] 73% loss = 0.020 1365m: epoch 47 [############## ] 73% loss = 0.021 1365m: epoch 47 [############## ] 74% loss = 0.022 1365m: epoch 47 [############## ] 74% loss = 0.020 1365m: epoch 47 [############## ] 74% loss = 0.021 1365m: epoch 47 [############### ] 75% loss = 0.019 1365m: epoch 47 [############### ] 75% loss = 0.021 1365m: epoch 47 [############### ] 75% loss = 0.021 1365m: epoch 47 [############### ] 76% loss = 0.020 1365m: epoch 47 [############### ] 76% loss = 0.022 1366m: epoch 47 [############### ] 76% loss = 0.022 1366m: epoch 47 [############### ] 77% loss = 0.021 1366m: epoch 47 [############### ] 77% loss = 0.022 1366m: epoch 47 [############### ] 77% loss = 0.023 1366m: epoch 47 [############### ] 78% loss = 0.022 1366m: epoch 47 [############### ] 78% loss = 0.020 1366m: epoch 47 [############### ] 78% loss = 0.021 1366m: epoch 47 [############### ] 79% loss = 0.023 1366m: epoch 47 [############### ] 79% loss = 0.021 1367m: epoch 47 [############### ] 79% loss = 0.020 1367m: epoch 47 [################ ] 80% loss = 0.022 1367m: epoch 47 [################ ] 80% loss = 0.021 1367m: epoch 47 [################ ] 80% loss = 0.021 1367m: epoch 47 [################ ] 81% loss = 0.022 1367m: epoch 47 [################ ] 81% loss = 0.023 1367m: epoch 47 [################ ] 81% loss = 0.022 1367m: epoch 47 [################ ] 82% loss = 0.022 1367m: epoch 47 [################ ] 82% loss = 0.021 1368m: epoch 47 [################ ] 82% loss = 0.021 1368m: epoch 47 [################ ] 83% loss = 0.021 1368m: epoch 47 [################ ] 83% loss = 0.022 1368m: epoch 47 [################ ] 83% loss = 0.023 1368m: epoch 47 [################ ] 84% loss = 0.022 1368m: epoch 47 [################ ] 84% loss = 0.021 1368m: epoch 47 [################# ] 85% loss = 0.021 1368m: epoch 47 [################# ] 85% loss = 0.021 1368m: epoch 47 [################# ] 85% loss = 0.022 1369m: epoch 47 [################# ] 86% loss = 0.022 1369m: epoch 47 [################# ] 86% loss = 0.022 1369m: epoch 47 [################# ] 86% loss = 0.021 1369m: epoch 47 [################# ] 87% loss = 0.022 1369m: epoch 47 [################# ] 87% loss = 0.021 1369m: epoch 47 [################# ] 87% loss = 0.023 1369m: epoch 47 [################# ] 88% loss = 0.021 1369m: epoch 47 [################# ] 88% loss = 0.022 1369m: epoch 47 [################# ] 88% loss = 0.022 1370m: epoch 47 [################# ] 89% loss = 0.022 1370m: epoch 47 [################# ] 89% loss = 0.022 1370m: epoch 47 [################# ] 89% loss = 0.023 1370m: epoch 47 [################## ] 90% loss = 0.023 1370m: epoch 47 [################## ] 90% loss = 0.021 1370m: epoch 47 [################## ] 90% loss = 0.020 1370m: epoch 47 [################## ] 91% loss = 0.021 1370m: epoch 47 [################## ] 91% loss = 0.022 1370m: epoch 47 [################## ] 91% loss = 0.022 1371m: epoch 47 [################## ] 92% loss = 0.021 1371m: epoch 47 [################## ] 92% loss = 0.021 1371m: epoch 47 [################## ] 92% loss = 0.022 1371m: epoch 47 [################## ] 93% loss = 0.020 1371m: epoch 47 [################## ] 93% loss = 0.021 1371m: epoch 47 [################## ] 93% loss = 0.022 1371m: epoch 47 [################## ] 94% loss = 0.022 1371m: epoch 47 [################## ] 94% loss = 0.021 1371m: epoch 47 [################## ] 94% loss = 0.022 1372m: epoch 47 [################### ] 95% loss = 0.022 1372m: epoch 47 [################### ] 95% loss = 0.021 1372m: epoch 47 [################### ] 96% loss = 0.021 1372m: epoch 47 [################### ] 96% loss = 0.021 1372m: epoch 47 [################### ] 96% loss = 0.022 1372m: epoch 47 [################### ] 97% loss = 0.022 1372m: epoch 47 [################### ] 97% loss = 0.021 1372m: epoch 47 [################### ] 97% loss = 0.022 1372m: epoch 47 [################### ] 98% loss = 0.020 1373m: epoch 47 [################### ] 98% loss = 0.021 1373m: epoch 47 [################### ] 98% loss = 0.022 1373m: epoch 47 [################### ] 99% loss = 0.019 1373m: epoch 47 [################### ] 99% loss = 0.021 1373m: epoch 47 [################### ] 99% loss = 0.0211373m: epoch 47 [####################] 100% loss = 0.021
\ No newline at end of file epoch 47 complete, loss = 0.021
model saved as token_classification_split_3_47_0.021
1373m: epoch 48 [ ] 0% loss = ... 1373m: epoch 48 [ ] 0% loss = 0.019 1373m: epoch 48 [ ] 0% loss = 0.019 1373m: epoch 48 [ ] 1% loss = 0.019 1374m: epoch 48 [ ] 1% loss = 0.019 1374m: epoch 48 [ ] 1% loss = 0.018 1374m: epoch 48 [ ] 2% loss = 0.020 1374m: epoch 48 [ ] 2% loss = 0.020 1374m: epoch 48 [ ] 2% loss = 0.018 1374m: epoch 48 [ ] 3% loss = 0.020 1374m: epoch 48 [ ] 3% loss = 0.020 1374m: epoch 48 [ ] 3% loss = 0.021 1374m: epoch 48 [ ] 4% loss = 0.020 1375m: epoch 48 [ ] 4% loss = 0.019 1375m: epoch 48 [ ] 4% loss = 0.019 1375m: epoch 48 [# ] 5% loss = 0.020 1375m: epoch 48 [# ] 5% loss = 0.019 1375m: epoch 48 [# ] 5% loss = 0.020 1375m: epoch 48 [# ] 6% loss = 0.019 1375m: epoch 48 [# ] 6% loss = 0.019 1375m: epoch 48 [# ] 6% loss = 0.019 1375m: epoch 48 [# ] 7% loss = 0.020 1376m: epoch 48 [# ] 7% loss = 0.019 1376m: epoch 48 [# ] 7% loss = 0.020 1376m: epoch 48 [# ] 8% loss = 0.019 1376m: epoch 48 [# ] 8% loss = 0.019 1376m: epoch 48 [# ] 8% loss = 0.019 1376m: epoch 48 [# ] 9% loss = 0.019 1376m: epoch 48 [# ] 9% loss = 0.020 1376m: epoch 48 [# ] 9% loss = 0.021 1376m: epoch 48 [## ] 10% loss = 0.021 1377m: epoch 48 [## ] 10% loss = 0.020 1377m: epoch 48 [## ] 11% loss = 0.019 1377m: epoch 48 [## ] 11% loss = 0.020 1377m: epoch 48 [## ] 11% loss = 0.018 1377m: epoch 48 [## ] 12% loss = 0.019 1377m: epoch 48 [## ] 12% loss = 0.020 1377m: epoch 48 [## ] 12% loss = 0.019 1377m: epoch 48 [## ] 13% loss = 0.020 1377m: epoch 48 [## ] 13% loss = 0.018 1378m: epoch 48 [## ] 13% loss = 0.019 1378m: epoch 48 [## ] 14% loss = 0.020 1378m: epoch 48 [## ] 14% loss = 0.020 1378m: epoch 48 [## ] 14% loss = 0.019 1378m: epoch 48 [### ] 15% loss = 0.019 1378m: epoch 48 [### ] 15% loss = 0.021 1378m: epoch 48 [### ] 15% loss = 0.020 1378m: epoch 48 [### ] 16% loss = 0.020 1378m: epoch 48 [### ] 16% loss = 0.018 1379m: epoch 48 [### ] 16% loss = 0.019 1379m: epoch 48 [### ] 17% loss = 0.020 1379m: epoch 48 [### ] 17% loss = 0.020 1379m: epoch 48 [### ] 17% loss = 0.021 1379m: epoch 48 [### ] 18% loss = 0.020 1379m: epoch 48 [### ] 18% loss = 0.020 1379m: epoch 48 [### ] 18% loss = 0.020 1379m: epoch 48 [### ] 19% loss = 0.019 1379m: epoch 48 [### ] 19% loss = 0.019 1380m: epoch 48 [### ] 19% loss = 0.019 1380m: epoch 48 [#### ] 20% loss = 0.021 1380m: epoch 48 [#### ] 20% loss = 0.020 1380m: epoch 48 [#### ] 20% loss = 0.019 1380m: epoch 48 [#### ] 21% loss = 0.019 1380m: epoch 48 [#### ] 21% loss = 0.020 1380m: epoch 48 [#### ] 22% loss = 0.020 1380m: epoch 48 [#### ] 22% loss = 0.018 1380m: epoch 48 [#### ] 22% loss = 0.019 1381m: epoch 48 [#### ] 23% loss = 0.019 1381m: epoch 48 [#### ] 23% loss = 0.020 1381m: epoch 48 [#### ] 23% loss = 0.020 1381m: epoch 48 [#### ] 24% loss = 0.019 1381m: epoch 48 [#### ] 24% loss = 0.019 1381m: epoch 48 [#### ] 24% loss = 0.019 1381m: epoch 48 [##### ] 25% loss = 0.020 1381m: epoch 48 [##### ] 25% loss = 0.018 1381m: epoch 48 [##### ] 25% loss = 0.019 1382m: epoch 48 [##### ] 26% loss = 0.019 1382m: epoch 48 [##### ] 26% loss = 0.019 1382m: epoch 48 [##### ] 26% loss = 0.019 1382m: epoch 48 [##### ] 27% loss = 0.020 1382m: epoch 48 [##### ] 27% loss = 0.021 1382m: epoch 48 [##### ] 27% loss = 0.021 1382m: epoch 48 [##### ] 28% loss = 0.018 1382m: epoch 48 [##### ] 28% loss = 0.019 1382m: epoch 48 [##### ] 28% loss = 0.019 1383m: epoch 48 [##### ] 29% loss = 0.022 1383m: epoch 48 [##### ] 29% loss = 0.019 1383m: epoch 48 [##### ] 29% loss = 0.020 1383m: epoch 48 [###### ] 30% loss = 0.020 1383m: epoch 48 [###### ] 30% loss = 0.022 1383m: epoch 48 [###### ] 30% loss = 0.020 1383m: epoch 48 [###### ] 31% loss = 0.021 1383m: epoch 48 [###### ] 31% loss = 0.019 1383m: epoch 48 [###### ] 32% loss = 0.021 1384m: epoch 48 [###### ] 32% loss = 0.020 1384m: epoch 48 [###### ] 32% loss = 0.021 1384m: epoch 48 [###### ] 33% loss = 0.022 1384m: epoch 48 [###### ] 33% loss = 0.020 1384m: epoch 48 [###### ] 33% loss = 0.020 1384m: epoch 48 [###### ] 34% loss = 0.021 1384m: epoch 48 [###### ] 34% loss = 0.020 1384m: epoch 48 [###### ] 34% loss = 0.019 1384m: epoch 48 [####### ] 35% loss = 0.021 1385m: epoch 48 [####### ] 35% loss = 0.022 1385m: epoch 48 [####### ] 35% loss = 0.021 1385m: epoch 48 [####### ] 36% loss = 0.022 1385m: epoch 48 [####### ] 36% loss = 0.019 1385m: epoch 48 [####### ] 36% loss = 0.019 1385m: epoch 48 [####### ] 37% loss = 0.020 1385m: epoch 48 [####### ] 37% loss = 0.021 1385m: epoch 48 [####### ] 37% loss = 0.020 1385m: epoch 48 [####### ] 38% loss = 0.021 1386m: epoch 48 [####### ] 38% loss = 0.020 1386m: epoch 48 [####### ] 38% loss = 0.021 1386m: epoch 48 [####### ] 39% loss = 0.020 1386m: epoch 48 [####### ] 39% loss = 0.023 1386m: epoch 48 [####### ] 39% loss = 0.021 1386m: epoch 48 [######## ] 40% loss = 0.020 1386m: epoch 48 [######## ] 40% loss = 0.020 1386m: epoch 48 [######## ] 40% loss = 0.022 1386m: epoch 48 [######## ] 41% loss = 0.020 1387m: epoch 48 [######## ] 41% loss = 0.020 1387m: epoch 48 [######## ] 41% loss = 0.020 1387m: epoch 48 [######## ] 42% loss = 0.021 1387m: epoch 48 [######## ] 42% loss = 0.021 1387m: epoch 48 [######## ] 43% loss = 0.019 1387m: epoch 48 [######## ] 43% loss = 0.022 1387m: epoch 48 [######## ] 43% loss = 0.020 1387m: epoch 48 [######## ] 44% loss = 0.019 1387m: epoch 48 [######## ] 44% loss = 0.022 1388m: epoch 48 [######## ] 44% loss = 0.020 1388m: epoch 48 [######### ] 45% loss = 0.020 1388m: epoch 48 [######### ] 45% loss = 0.020 1388m: epoch 48 [######### ] 45% loss = 0.020 1388m: epoch 48 [######### ] 46% loss = 0.020 1388m: epoch 48 [######### ] 46% loss = 0.021 1388m: epoch 48 [######### ] 46% loss = 0.020 1388m: epoch 48 [######### ] 47% loss = 0.020 1388m: epoch 48 [######### ] 47% loss = 0.022 1389m: epoch 48 [######### ] 47% loss = 0.021 1389m: epoch 48 [######### ] 48% loss = 0.020 1389m: epoch 48 [######### ] 48% loss = 0.021 1389m: epoch 48 [######### ] 48% loss = 0.021 1389m: epoch 48 [######### ] 49% loss = 0.022 1389m: epoch 48 [######### ] 49% loss = 0.022 1389m: epoch 48 [######### ] 49% loss = 0.020 1389m: epoch 48 [########## ] 50% loss = 0.021 1389m: epoch 48 [########## ] 50% loss = 0.020 1390m: epoch 48 [########## ] 50% loss = 0.021 1390m: epoch 48 [########## ] 51% loss = 0.021 1390m: epoch 48 [########## ] 51% loss = 0.020 1390m: epoch 48 [########## ] 51% loss = 0.021 1390m: epoch 48 [########## ] 52% loss = 0.021 1390m: epoch 48 [########## ] 52% loss = 0.020 1390m: epoch 48 [########## ] 53% loss = 0.020 1390m: epoch 48 [########## ] 53% loss = 0.021 1390m: epoch 48 [########## ] 53% loss = 0.021 1391m: epoch 48 [########## ] 54% loss = 0.020 1391m: epoch 48 [########## ] 54% loss = 0.019 1391m: epoch 48 [########## ] 54% loss = 0.021 1391m: epoch 48 [########### ] 55% loss = 0.020 1391m: epoch 48 [########### ] 55% loss = 0.020 1391m: epoch 48 [########### ] 55% loss = 0.022 1391m: epoch 48 [########### ] 56% loss = 0.021 1391m: epoch 48 [########### ] 56% loss = 0.021 1391m: epoch 48 [########### ] 56% loss = 0.019 1392m: epoch 48 [########### ] 57% loss = 0.021 1392m: epoch 48 [########### ] 57% loss = 0.022 1392m: epoch 48 [########### ] 57% loss = 0.021 1392m: epoch 48 [########### ] 58% loss = 0.019 1392m: epoch 48 [########### ] 58% loss = 0.022 1392m: epoch 48 [########### ] 58% loss = 0.022 1392m: epoch 48 [########### ] 59% loss = 0.020 1392m: epoch 48 [########### ] 59% loss = 0.021 1392m: epoch 48 [########### ] 59% loss = 0.019 1393m: epoch 48 [############ ] 60% loss = 0.019 1393m: epoch 48 [############ ] 60% loss = 0.023 1393m: epoch 48 [############ ] 60% loss = 0.021 1393m: epoch 48 [############ ] 61% loss = 0.023 1393m: epoch 48 [############ ] 61% loss = 0.020 1393m: epoch 48 [############ ] 61% loss = 0.021 1393m: epoch 48 [############ ] 62% loss = 0.020 1393m: epoch 48 [############ ] 62% loss = 0.021 1393m: epoch 48 [############ ] 62% loss = 0.021 1394m: epoch 48 [############ ] 63% loss = 0.021 1394m: epoch 48 [############ ] 63% loss = 0.021 1394m: epoch 48 [############ ] 64% loss = 0.021 1394m: epoch 48 [############ ] 64% loss = 0.020 1394m: epoch 48 [############ ] 64% loss = 0.020 1394m: epoch 48 [############# ] 65% loss = 0.021 1394m: epoch 48 [############# ] 65% loss = 0.020 1394m: epoch 48 [############# ] 65% loss = 0.021 1394m: epoch 48 [############# ] 66% loss = 0.022 1395m: epoch 48 [############# ] 66% loss = 0.021 1395m: epoch 48 [############# ] 66% loss = 0.022 1395m: epoch 48 [############# ] 67% loss = 0.020 1395m: epoch 48 [############# ] 67% loss = 0.021 1395m: epoch 48 [############# ] 67% loss = 0.021 1395m: epoch 48 [############# ] 68% loss = 0.021 1395m: epoch 48 [############# ] 68% loss = 0.021 1395m: epoch 48 [############# ] 68% loss = 0.022 1395m: epoch 48 [############# ] 69% loss = 0.022 1396m: epoch 48 [############# ] 69% loss = 0.022 1396m: epoch 48 [############# ] 69% loss = 0.021 1396m: epoch 48 [############## ] 70% loss = 0.021 1396m: epoch 48 [############## ] 70% loss = 0.021 1396m: epoch 48 [############## ] 70% loss = 0.020 1396m: epoch 48 [############## ] 71% loss = 0.022 1396m: epoch 48 [############## ] 71% loss = 0.019 1396m: epoch 48 [############## ] 71% loss = 0.021 1396m: epoch 48 [############## ] 72% loss = 0.021 1397m: epoch 48 [############## ] 72% loss = 0.020 1397m: epoch 48 [############## ] 72% loss = 0.020 1397m: epoch 48 [############## ] 73% loss = 0.021 1397m: epoch 48 [############## ] 73% loss = 0.020 1397m: epoch 48 [############## ] 74% loss = 0.019 1397m: epoch 48 [############## ] 74% loss = 0.020 1397m: epoch 48 [############## ] 74% loss = 0.020 1397m: epoch 48 [############### ] 75% loss = 0.022 1397m: epoch 48 [############### ] 75% loss = 0.020 1398m: epoch 48 [############### ] 75% loss = 0.021 1398m: epoch 48 [############### ] 76% loss = 0.021 1398m: epoch 48 [############### ] 76% loss = 0.022 1398m: epoch 48 [############### ] 76% loss = 0.020 1398m: epoch 48 [############### ] 77% loss = 0.021 1398m: epoch 48 [############### ] 77% loss = 0.022 1398m: epoch 48 [############### ] 77% loss = 0.023 1398m: epoch 48 [############### ] 78% loss = 0.021 1398m: epoch 48 [############### ] 78% loss = 0.023 1399m: epoch 48 [############### ] 78% loss = 0.024 1399m: epoch 48 [############### ] 79% loss = 0.023 1399m: epoch 48 [############### ] 79% loss = 0.022 1399m: epoch 48 [############### ] 79% loss = 0.020 1399m: epoch 48 [################ ] 80% loss = 0.021 1399m: epoch 48 [################ ] 80% loss = 0.021 1399m: epoch 48 [################ ] 80% loss = 0.020 1399m: epoch 48 [################ ] 81% loss = 0.021 1399m: epoch 48 [################ ] 81% loss = 0.021 1400m: epoch 48 [################ ] 81% loss = 0.022 1400m: epoch 48 [################ ] 82% loss = 0.020 1400m: epoch 48 [################ ] 82% loss = 0.022 1400m: epoch 48 [################ ] 82% loss = 0.022 1400m: epoch 48 [################ ] 83% loss = 0.019 1400m: epoch 48 [################ ] 83% loss = 0.021 1400m: epoch 48 [################ ] 83% loss = 0.020 1400m: epoch 48 [################ ] 84% loss = 0.021 1400m: epoch 48 [################ ] 84% loss = 0.023 1401m: epoch 48 [################# ] 85% loss = 0.023 1401m: epoch 48 [################# ] 85% loss = 0.024 1401m: epoch 48 [################# ] 85% loss = 0.023 1401m: epoch 48 [################# ] 86% loss = 0.022 1401m: epoch 48 [################# ] 86% loss = 0.022 1401m: epoch 48 [################# ] 86% loss = 0.021 1401m: epoch 48 [################# ] 87% loss = 0.020 1401m: epoch 48 [################# ] 87% loss = 0.021 1401m: epoch 48 [################# ] 87% loss = 0.023 1402m: epoch 48 [################# ] 88% loss = 0.021 1402m: epoch 48 [################# ] 88% loss = 0.021 1402m: epoch 48 [################# ] 88% loss = 0.022 1402m: epoch 48 [################# ] 89% loss = 0.023 1402m: epoch 48 [################# ] 89% loss = 0.020 1402m: epoch 48 [################# ] 89% loss = 0.022 1402m: epoch 48 [################## ] 90% loss = 0.021 1402m: epoch 48 [################## ] 90% loss = 0.020 1402m: epoch 48 [################## ] 90% loss = 0.023 1403m: epoch 48 [################## ] 91% loss = 0.021 1403m: epoch 48 [################## ] 91% loss = 0.021 1403m: epoch 48 [################## ] 91% loss = 0.020 1403m: epoch 48 [################## ] 92% loss = 0.022 1403m: epoch 48 [################## ] 92% loss = 0.021 1403m: epoch 48 [################## ] 92% loss = 0.021 1403m: epoch 48 [################## ] 93% loss = 0.020 1403m: epoch 48 [################## ] 93% loss = 0.021 1403m: epoch 48 [################## ] 93% loss = 0.023 1404m: epoch 48 [################## ] 94% loss = 0.023 1404m: epoch 48 [################## ] 94% loss = 0.020 1404m: epoch 48 [################## ] 94% loss = 0.023 1404m: epoch 48 [################### ] 95% loss = 0.020 1404m: epoch 48 [################### ] 95% loss = 0.022 1404m: epoch 48 [################### ] 96% loss = 0.022 1404m: epoch 48 [################### ] 96% loss = 0.023 1404m: epoch 48 [################### ] 96% loss = 0.020 1404m: epoch 48 [################### ] 97% loss = 0.020 1405m: epoch 48 [################### ] 97% loss = 0.019 1405m: epoch 48 [################### ] 97% loss = 0.022 1405m: epoch 48 [################### ] 98% loss = 0.021 1405m: epoch 48 [################### ] 98% loss = 0.020 1405m: epoch 48 [################### ] 98% loss = 0.020 1405m: epoch 48 [################### ] 99% loss = 0.022 1405m: epoch 48 [################### ] 99% loss = 0.021 1405m: epoch 48 [################### ] 99% loss = 0.0221405m: epoch 48 [####################] 100% loss = 0.022
epoch 48 complete, loss = 0.022
model saved as token_classification_split_3_48_0.022
1405m: epoch 49 [ ] 0% loss = ... 1406m: epoch 49 [ ] 0% loss = 0.020 1406m: epoch 49 [ ] 0% loss = 0.018 1406m: epoch 49 [ ] 1% loss = 0.019 1406m: epoch 49 [ ] 1% loss = 0.018 1406m: epoch 49 [ ] 1% loss = 0.018 1406m: epoch 49 [ ] 2% loss = 0.020 1406m: epoch 49 [ ] 2% loss = 0.018 1406m: epoch 49 [ ] 2% loss = 0.019 1406m: epoch 49 [ ] 3% loss = 0.021 1407m: epoch 49 [ ] 3% loss = 0.019 1407m: epoch 49 [ ] 3% loss = 0.019 1407m: epoch 49 [ ] 4% loss = 0.019 1407m: epoch 49 [ ] 4% loss = 0.020 1407m: epoch 49 [ ] 4% loss = 0.019 1407m: epoch 49 [# ] 5% loss = 0.019 1407m: epoch 49 [# ] 5% loss = 0.018 1407m: epoch 49 [# ] 5% loss = 0.019 1407m: epoch 49 [# ] 6% loss = 0.020 1408m: epoch 49 [# ] 6% loss = 0.019 1408m: epoch 49 [# ] 6% loss = 0.018 1408m: epoch 49 [# ] 7% loss = 0.019 1408m: epoch 49 [# ] 7% loss = 0.018 1408m: epoch 49 [# ] 7% loss = 0.021 1408m: epoch 49 [# ] 8% loss = 0.019 1408m: epoch 49 [# ] 8% loss = 0.019 1408m: epoch 49 [# ] 8% loss = 0.020 1408m: epoch 49 [# ] 9% loss = 0.019 1409m: epoch 49 [# ] 9% loss = 0.019 1409m: epoch 49 [# ] 9% loss = 0.019 1409m: epoch 49 [## ] 10% loss = 0.022 1409m: epoch 49 [## ] 10% loss = 0.019 1409m: epoch 49 [## ] 11% loss = 0.019 1409m: epoch 49 [## ] 11% loss = 0.019 1409m: epoch 49 [## ] 11% loss = 0.020 1409m: epoch 49 [## ] 12% loss = 0.017 1409m: epoch 49 [## ] 12% loss = 0.019 1410m: epoch 49 [## ] 12% loss = 0.020 1410m: epoch 49 [## ] 13% loss = 0.021 1410m: epoch 49 [## ] 13% loss = 0.020 1410m: epoch 49 [## ] 13% loss = 0.018 1410m: epoch 49 [## ] 14% loss = 0.020 1410m: epoch 49 [## ] 14% loss = 0.019 1410m: epoch 49 [## ] 14% loss = 0.021 1410m: epoch 49 [### ] 15% loss = 0.020 1410m: epoch 49 [### ] 15% loss = 0.019 1411m: epoch 49 [### ] 15% loss = 0.020 1411m: epoch 49 [### ] 16% loss = 0.021 1411m: epoch 49 [### ] 16% loss = 0.019 1411m: epoch 49 [### ] 16% loss = 0.020 1411m: epoch 49 [### ] 17% loss = 0.020 1411m: epoch 49 [### ] 17% loss = 0.019 1411m: epoch 49 [### ] 17% loss = 0.020 1411m: epoch 49 [### ] 18% loss = 0.021 1411m: epoch 49 [### ] 18% loss = 0.020 1412m: epoch 49 [### ] 18% loss = 0.020 1412m: epoch 49 [### ] 19% loss = 0.020 1412m: epoch 49 [### ] 19% loss = 0.019 1412m: epoch 49 [### ] 19% loss = 0.019 1412m: epoch 49 [#### ] 20% loss = 0.020 1412m: epoch 49 [#### ] 20% loss = 0.019 1412m: epoch 49 [#### ] 20% loss = 0.020 1412m: epoch 49 [#### ] 21% loss = 0.021 1412m: epoch 49 [#### ] 21% loss = 0.020 1413m: epoch 49 [#### ] 22% loss = 0.021 1413m: epoch 49 [#### ] 22% loss = 0.020 1413m: epoch 49 [#### ] 22% loss = 0.020 1413m: epoch 49 [#### ] 23% loss = 0.020 1413m: epoch 49 [#### ] 23% loss = 0.021 1413m: epoch 49 [#### ] 23% loss = 0.019 1413m: epoch 49 [#### ] 24% loss = 0.021 1413m: epoch 49 [#### ] 24% loss = 0.020 1413m: epoch 49 [#### ] 24% loss = 0.021 1414m: epoch 49 [##### ] 25% loss = 0.020 1414m: epoch 49 [##### ] 25% loss = 0.019 1414m: epoch 49 [##### ] 25% loss = 0.020 1414m: epoch 49 [##### ] 26% loss = 0.021 1414m: epoch 49 [##### ] 26% loss = 0.021 1414m: epoch 49 [##### ] 26% loss = 0.022 1414m: epoch 49 [##### ] 27% loss = 0.021 1414m: epoch 49 [##### ] 27% loss = 0.021 1414m: epoch 49 [##### ] 27% loss = 0.020 1415m: epoch 49 [##### ] 28% loss = 0.019 1415m: epoch 49 [##### ] 28% loss = 0.020 1415m: epoch 49 [##### ] 28% loss = 0.020 1415m: epoch 49 [##### ] 29% loss = 0.019 1415m: epoch 49 [##### ] 29% loss = 0.021 1415m: epoch 49 [##### ] 29% loss = 0.020 1415m: epoch 49 [###### ] 30% loss = 0.021 1415m: epoch 49 [###### ] 30% loss = 0.020 1415m: epoch 49 [###### ] 30% loss = 0.019 1416m: epoch 49 [###### ] 31% loss = 0.021 1416m: epoch 49 [###### ] 31% loss = 0.020 1416m: epoch 49 [###### ] 32% loss = 0.021 1416m: epoch 49 [###### ] 32% loss = 0.020 1416m: epoch 49 [###### ] 32% loss = 0.020 1416m: epoch 49 [###### ] 33% loss = 0.020 1416m: epoch 49 [###### ] 33% loss = 0.021 1416m: epoch 49 [###### ] 33% loss = 0.019 1416m: epoch 49 [###### ] 34% loss = 0.019 1417m: epoch 49 [###### ] 34% loss = 0.019 1417m: epoch 49 [###### ] 34% loss = 0.021 1417m: epoch 49 [####### ] 35% loss = 0.020 1417m: epoch 49 [####### ] 35% loss = 0.022 1417m: epoch 49 [####### ] 35% loss = 0.020 1417m: epoch 49 [####### ] 36% loss = 0.019 1417m: epoch 49 [####### ] 36% loss = 0.022 1417m: epoch 49 [####### ] 36% loss = 0.021 1417m: epoch 49 [####### ] 37% loss = 0.021 1418m: epoch 49 [####### ] 37% loss = 0.019 1418m: epoch 49 [####### ] 37% loss = 0.019 1418m: epoch 49 [####### ] 38% loss = 0.020 1418m: epoch 49 [####### ] 38% loss = 0.021 1418m: epoch 49 [####### ] 38% loss = 0.021 1418m: epoch 49 [####### ] 39% loss = 0.021 1418m: epoch 49 [####### ] 39% loss = 0.018 1418m: epoch 49 [####### ] 39% loss = 0.021 1418m: epoch 49 [######## ] 40% loss = 0.018 1419m: epoch 49 [######## ] 40% loss = 0.020 1419m: epoch 49 [######## ] 40% loss = 0.021 1419m: epoch 49 [######## ] 41% loss = 0.019 1419m: epoch 49 [######## ] 41% loss = 0.020 1419m: epoch 49 [######## ] 41% loss = 0.020 1419m: epoch 49 [######## ] 42% loss = 0.019 1419m: epoch 49 [######## ] 42% loss = 0.021 1419m: epoch 49 [######## ] 43% loss = 0.020 1419m: epoch 49 [######## ] 43% loss = 0.020 1420m: epoch 49 [######## ] 43% loss = 0.021 1420m: epoch 49 [######## ] 44% loss = 0.020 1420m: epoch 49 [######## ] 44% loss = 0.021 1420m: epoch 49 [######## ] 44% loss = 0.020 1420m: epoch 49 [######### ] 45% loss = 0.021 1420m: epoch 49 [######### ] 45% loss = 0.020 1420m: epoch 49 [######### ] 45% loss = 0.019 1420m: epoch 49 [######### ] 46% loss = 0.020 1420m: epoch 49 [######### ] 46% loss = 0.021 1421m: epoch 49 [######### ] 46% loss = 0.020 1421m: epoch 49 [######### ] 47% loss = 0.019 1421m: epoch 49 [######### ] 47% loss = 0.021 1421m: epoch 49 [######### ] 47% loss = 0.020 1421m: epoch 49 [######### ] 48% loss = 0.019 1421m: epoch 49 [######### ] 48% loss = 0.020 1421m: epoch 49 [######### ] 48% loss = 0.021 1421m: epoch 49 [######### ] 49% loss = 0.021 1421m: epoch 49 [######### ] 49% loss = 0.019 1422m: epoch 49 [######### ] 49% loss = 0.020 1422m: epoch 49 [########## ] 50% loss = 0.021 1422m: epoch 49 [########## ] 50% loss = 0.021 1422m: epoch 49 [########## ] 50% loss = 0.022 1422m: epoch 49 [########## ] 51% loss = 0.020 1422m: epoch 49 [########## ] 51% loss = 0.022 1422m: epoch 49 [########## ] 51% loss = 0.021 1422m: epoch 49 [########## ] 52% loss = 0.021 1422m: epoch 49 [########## ] 52% loss = 0.021 1423m: epoch 49 [########## ] 53% loss = 0.020 1423m: epoch 49 [########## ] 53% loss = 0.021 1423m: epoch 49 [########## ] 53% loss = 0.021 1423m: epoch 49 [########## ] 54% loss = 0.021 1423m: epoch 49 [########## ] 54% loss = 0.022 1423m: epoch 49 [########## ] 54% loss = 0.020 1423m: epoch 49 [########### ] 55% loss = 0.020 1423m: epoch 49 [########### ] 55% loss = 0.021 1423m: epoch 49 [########### ] 55% loss = 0.021 1424m: epoch 49 [########### ] 56% loss = 0.020 1424m: epoch 49 [########### ] 56% loss = 0.020 1424m: epoch 49 [########### ] 56% loss = 0.020 1424m: epoch 49 [########### ] 57% loss = 0.021 1424m: epoch 49 [########### ] 57% loss = 0.020 1424m: epoch 49 [########### ] 57% loss = 0.020 1424m: epoch 49 [########### ] 58% loss = 0.020 1424m: epoch 49 [########### ] 58% loss = 0.021 1424m: epoch 49 [########### ] 58% loss = 0.021 1425m: epoch 49 [########### ] 59% loss = 0.020 1425m: epoch 49 [########### ] 59% loss = 0.022 1425m: epoch 49 [########### ] 59% loss = 0.022 1425m: epoch 49 [############ ] 60% loss = 0.023 1425m: epoch 49 [############ ] 60% loss = 0.022 1425m: epoch 49 [############ ] 60% loss = 0.021 1425m: epoch 49 [############ ] 61% loss = 0.021 1425m: epoch 49 [############ ] 61% loss = 0.021 1425m: epoch 49 [############ ] 61% loss = 0.022 1426m: epoch 49 [############ ] 62% loss = 0.021 1426m: epoch 49 [############ ] 62% loss = 0.022 1426m: epoch 49 [############ ] 62% loss = 0.021 1426m: epoch 49 [############ ] 63% loss = 0.021 1426m: epoch 49 [############ ] 63% loss = 0.021 1426m: epoch 49 [############ ] 64% loss = 0.022 1426m: epoch 49 [############ ] 64% loss = 0.021 1426m: epoch 49 [############ ] 64% loss = 0.020 1426m: epoch 49 [############# ] 65% loss = 0.020 1427m: epoch 49 [############# ] 65% loss = 0.021 1427m: epoch 49 [############# ] 65% loss = 0.021 1427m: epoch 49 [############# ] 66% loss = 0.021 1427m: epoch 49 [############# ] 66% loss = 0.022 1427m: epoch 49 [############# ] 66% loss = 0.020 1427m: epoch 49 [############# ] 67% loss = 0.021 1427m: epoch 49 [############# ] 67% loss = 0.023 1427m: epoch 49 [############# ] 67% loss = 0.022 1427m: epoch 49 [############# ] 68% loss = 0.020 1428m: epoch 49 [############# ] 68% loss = 0.019 1428m: epoch 49 [############# ] 68% loss = 0.020 1428m: epoch 49 [############# ] 69% loss = 0.021 1428m: epoch 49 [############# ] 69% loss = 0.022 1428m: epoch 49 [############# ] 69% loss = 0.021 1428m: epoch 49 [############## ] 70% loss = 0.020 1428m: epoch 49 [############## ] 70% loss = 0.020 1428m: epoch 49 [############## ] 70% loss = 0.023 1428m: epoch 49 [############## ] 71% loss = 0.021 1429m: epoch 49 [############## ] 71% loss = 0.022 1429m: epoch 49 [############## ] 71% loss = 0.023 1429m: epoch 49 [############## ] 72% loss = 0.022 1429m: epoch 49 [############## ] 72% loss = 0.020 1429m: epoch 49 [############## ] 72% loss = 0.021 1429m: epoch 49 [############## ] 73% loss = 0.022 1429m: epoch 49 [############## ] 73% loss = 0.021 1429m: epoch 49 [############## ] 74% loss = 0.022 1429m: epoch 49 [############## ] 74% loss = 0.020 1430m: epoch 49 [############## ] 74% loss = 0.021 1430m: epoch 49 [############### ] 75% loss = 0.021 1430m: epoch 49 [############### ] 75% loss = 0.020 1430m: epoch 49 [############### ] 75% loss = 0.021 1430m: epoch 49 [############### ] 76% loss = 0.022 1430m: epoch 49 [############### ] 76% loss = 0.020 1430m: epoch 49 [############### ] 76% loss = 0.020 1430m: epoch 49 [############### ] 77% loss = 0.020 1430m: epoch 49 [############### ] 77% loss = 0.021 1431m: epoch 49 [############### ] 77% loss = 0.022 1431m: epoch 49 [############### ] 78% loss = 0.021 1431m: epoch 49 [############### ] 78% loss = 0.022 1431m: epoch 49 [############### ] 78% loss = 0.022 1431m: epoch 49 [############### ] 79% loss = 0.021 1431m: epoch 49 [############### ] 79% loss = 0.019 1431m: epoch 49 [############### ] 79% loss = 0.021 1431m: epoch 49 [################ ] 80% loss = 0.021 1431m: epoch 49 [################ ] 80% loss = 0.019 1432m: epoch 49 [################ ] 80% loss = 0.020 1432m: epoch 49 [################ ] 81% loss = 0.021 1432m: epoch 49 [################ ] 81% loss = 0.020 1432m: epoch 49 [################ ] 81% loss = 0.021 1432m: epoch 49 [################ ] 82% loss = 0.022 1432m: epoch 49 [################ ] 82% loss = 0.021 1432m: epoch 49 [################ ] 82% loss = 0.020 1432m: epoch 49 [################ ] 83% loss = 0.020 1433m: epoch 49 [################ ] 83% loss = 0.021 1433m: epoch 49 [################ ] 83% loss = 0.022 1433m: epoch 49 [################ ] 84% loss = 0.021 1433m: epoch 49 [################ ] 84% loss = 0.022 1433m: epoch 49 [################# ] 85% loss = 0.021 1433m: epoch 49 [################# ] 85% loss = 0.022 1433m: epoch 49 [################# ] 85% loss = 0.024 1433m: epoch 49 [################# ] 86% loss = 0.021 1433m: epoch 49 [################# ] 86% loss = 0.022 1434m: epoch 49 [################# ] 86% loss = 0.021 1434m: epoch 49 [################# ] 87% loss = 0.022 1434m: epoch 49 [################# ] 87% loss = 0.021 1434m: epoch 49 [################# ] 87% loss = 0.020 1434m: epoch 49 [################# ] 88% loss = 0.022 1434m: epoch 49 [################# ] 88% loss = 0.020 1434m: epoch 49 [################# ] 88% loss = 0.022 1434m: epoch 49 [################# ] 89% loss = 0.022 1434m: epoch 49 [################# ] 89% loss = 0.021 1435m: epoch 49 [################# ] 89% loss = 0.021 1435m: epoch 49 [################## ] 90% loss = 0.021 1435m: epoch 49 [################## ] 90% loss = 0.019 1435m: epoch 49 [################## ] 90% loss = 0.021 1435m: epoch 49 [################## ] 91% loss = 0.020 1435m: epoch 49 [################## ] 91% loss = 0.020 1435m: epoch 49 [################## ] 91% loss = 0.021 1435m: epoch 49 [################## ] 92% loss = 0.021 1435m: epoch 49 [################## ] 92% loss = 0.021 1436m: epoch 49 [################## ] 92% loss = 0.021 1436m: epoch 49 [################## ] 93% loss = 0.022 1436m: epoch 49 [################## ] 93% loss = 0.021 1436m: epoch 49 [################## ] 93% loss = 0.023 1436m: epoch 49 [################## ] 94% loss = 0.020 1436m: epoch 49 [################## ] 94% loss = 0.021 1436m: epoch 49 [################## ] 94% loss = 0.021 1436m: epoch 49 [################### ] 95% loss = 0.021 1436m: epoch 49 [################### ] 95% loss = 0.021 1437m: epoch 49 [################### ] 96% loss = 0.023 1437m: epoch 49 [################### ] 96% loss = 0.021 1437m: epoch 49 [################### ] 96% loss = 0.024 1437m: epoch 49 [################### ] 97% loss = 0.022 1437m: epoch 49 [################### ] 97% loss = 0.023 1437m: epoch 49 [################### ] 97% loss = 0.022 1437m: epoch 49 [################### ] 98% loss = 0.023 1437m: epoch 49 [################### ] 98% loss = 0.020 1437m: epoch 49 [################### ] 98% loss = 0.020 1438m: epoch 49 [################### ] 99% loss = 0.021 1438m: epoch 49 [################### ] 99% loss = 0.019 1438m: epoch 49 [################### ] 99% loss = 0.0211438m: epoch 49 [####################] 100% loss = 0.021
epoch 49 complete, loss = 0.021
model saved as token_classification_split_3_49_0.021
1438m: epoch 50 [ ] 0% loss = ... 1438m: epoch 50 [ ] 0% loss = 0.019 1438m: epoch 50 [ ] 0% loss = 0.020 1438m: epoch 50 [ ] 1% loss = 0.019 1438m: epoch 50 [ ] 1% loss = 0.018 1438m: epoch 50 [ ] 1% loss = 0.019 1439m: epoch 50 [ ] 2% loss = 0.018 1439m: epoch 50 [ ] 2% loss = 0.019 1439m: epoch 50 [ ] 2% loss = 0.019 1439m: epoch 50 [ ] 3% loss = 0.019 1439m: epoch 50 [ ] 3% loss = 0.020 1439m: epoch 50 [ ] 3% loss = 0.017 1439m: epoch 50 [ ] 4% loss = 0.019 1439m: epoch 50 [ ] 4% loss = 0.020 1439m: epoch 50 [ ] 4% loss = 0.020 1440m: epoch 50 [# ] 5% loss = 0.018 1440m: epoch 50 [# ] 5% loss = 0.019 1440m: epoch 50 [# ] 5% loss = 0.020 1440m: epoch 50 [# ] 6% loss = 0.020 1440m: epoch 50 [# ] 6% loss = 0.019 1440m: epoch 50 [# ] 6% loss = 0.018 1440m: epoch 50 [# ] 7% loss = 0.018 1440m: epoch 50 [# ] 7% loss = 0.020 1440m: epoch 50 [# ] 7% loss = 0.019 1441m: epoch 50 [# ] 8% loss = 0.020 1441m: epoch 50 [# ] 8% loss = 0.019 1441m: epoch 50 [# ] 8% loss = 0.019 1441m: epoch 50 [# ] 9% loss = 0.018 1441m: epoch 50 [# ] 9% loss = 0.020 1441m: epoch 50 [# ] 9% loss = 0.020 1441m: epoch 50 [## ] 10% loss = 0.019 1441m: epoch 50 [## ] 10% loss = 0.021 1441m: epoch 50 [## ] 11% loss = 0.020
\ No newline at end of file
......
This source diff could not be displayed because it is too large. You can view the blob instead.
This source diff could not be displayed because it is too large. You can view the blob instead.
...@@ -8,7 +8,7 @@ with open("./data/voc/yunmu.txt","r",encoding="utf-8") as f: ...@@ -8,7 +8,7 @@ with open("./data/voc/yunmu.txt","r",encoding="utf-8") as f:
yunmus=f.readlines() yunmus=f.readlines()
yunmus=[a.strip() for a in yunmus] yunmus=[a.strip() for a in yunmus]
a = 'ying' a = 'zhi'
a=convert_finals(a) a=convert_finals(a)
print(a) print(a)
strict = True strict = True
......
from tqdm import tqdm
import os
import pickle
import itertools
pkl_dir="./data/pkl/pinyin_pkl(1)"
for file in tqdm(os.listdir(pkl_dir)):
# print("filename:{}".format(file))
contents = pickle.load(open(os.path.join(pkl_dir,file),"rb"))
# contents=random.sample(contents,10)
# print(contents)
# contents=[list(itertools.chain.from_iterable(lines)) for lines in contents]
# contents=list(itertools.chain.from_iterable(contents))
with open(os.path.join("data/pkl/pinyin_pkl_txt",file[:-4]+".txt"),"w",encoding="utf-8") as f:
f.write(" ".join(contents))
...@@ -20,7 +20,8 @@ def get_result(src, model, SRC, TRG, opt): ...@@ -20,7 +20,8 @@ def get_result(src, model, SRC, TRG, opt):
output = model(src, src_mask) output = model(src, src_mask)
output = F.softmax(output,dim=-1) output = F.softmax(output,dim=-1)
preds = torch.argmax(output, dim=-1) preds = torch.argmax(output, dim=-1)
return ''.join([TRG.vocab.itos[tok] for tok in preds[0][:]]).replace("_", "") # return ''.join([TRG.vocab.itos[tok] for tok in preds[0][:]]).replace("_", "")
return ' '.join([TRG.vocab.itos[tok] for tok in preds[0][:]])
def translate_sentence(sentence, model, opt, SRC, TRG): def translate_sentence(sentence, model, opt, SRC, TRG):
......
...@@ -38,6 +38,8 @@ def main(): ...@@ -38,6 +38,8 @@ def main():
parser.add_argument('-floyd', action='store_true') parser.add_argument('-floyd', action='store_true')
parser.add_argument('-test_dir', type=str, required=True) parser.add_argument('-test_dir', type=str, required=True)
parser.add_argument('-result_dir', type=str, required=True) parser.add_argument('-result_dir', type=str, required=True)
parser.add_argument('-src_voc')
parser.add_argument('-trg_voc')
opt = parser.parse_args() opt = parser.parse_args()
...@@ -54,8 +56,10 @@ def main(): ...@@ -54,8 +56,10 @@ def main():
contents = pickle.load(open(os.path.join(opt.test_dir,file),"rb")) contents = pickle.load(open(os.path.join(opt.test_dir,file),"rb"))
# contents=random.sample(contents,10) # contents=random.sample(contents,10)
contents=[list(itertools.chain.from_iterable(lines)) for lines in contents] # contents=[list(itertools.chain.from_iterable(lines)) for lines in contents]
contents=list(itertools.chain.from_iterable(contents)) # # print(contents)
# contents=list(itertools.chain.from_iterable(contents))
# print(contents)
# print(contents) # print(contents)
start=time.time() start=time.time()
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment