diff --git a/code/config_global_1.2.json b/code/config_global_1.2.json
index a6c9f8ccf5aba587ac066181ccaee169ccc38717..7a6b920522984459dccd26c878b43bf7821c9797 100644
--- a/code/config_global_1.2.json
+++ b/code/config_global_1.2.json
@@ -1,19 +1,19 @@
{
- "usecase_description": "Config file for usecase_1 : from a raw text, get the same text but with EDU bracket.",
+ "usecase_description": "Config file for usecase_1 : from a tokenized text, get the same text but with EDU bracket.",
"data_raw": {
- "name": "edgar_poe_en",
- "exte": ".txt",
+ "name": "edgar_poe_short",
+ "exte": ".conll",
"language": "en",
"existing_metadata": true
},
"steps":{
"main": "annotation",
"pre-processing": {
- "to_do": true,
+ "to_do": false,
"syntactic_tool": "stanza",
"sentence_split": true,
"tokenization": true,
- "syntactic_parsing": false,
+ "syntactic_parsing": true,
"create_metadata": {
"to_do": true,
"line": "paragraph",
@@ -30,16 +30,17 @@
"validation_data_path": null
}
},
+ "evaluation": true,
"gold_test_data_path": null
},
"output":{
"conll_file":{
"to_do": true,
"metadata": true,
- "with_gold_labels": false
+ "with_gold_labels": true
},
"txt_file":{
- "to_do": false,
+ "to_do": true,
"metadata": true
}
}
diff --git a/code/config_global_1.21.json b/code/config_global_1.21.json
deleted file mode 100644
index 21351a824b3bbaf6558f3f1b307911bd3af3c570..0000000000000000000000000000000000000000
--- a/code/config_global_1.21.json
+++ /dev/null
@@ -1,49 +0,0 @@
-{
- "usecase_description": "Config file for usecase_1 : from a tokenized text, get the same text but with EDU bracket.",
- "data_raw": {
- "name": "edgar_poe_short",
- "exte": ".conll",
- "language": "en",
- "existing_metadata": true
- },
- "steps":{
- "main": "annotation",
- "pre-processing": {
- "to_do": false,
- "syntactic_tool": "stanza",
- "sentence_split": true,
- "tokenization": true,
- "syntactic_parsing": true,
- "create_metadata": {
- "to_do": true,
- "line": "paragraph",
- "sent": "sent"
- }
- },
- "discourse_segmenter": {
- "model": "/home/lriviere/andiamo/discut/Results_conllu/results_eng.rst.gum-eng_bert/model.tar.gz",
- "training": {
- "toolkit": null,
- "pre_trained_lm": null,
- "config_file": null,
- "train_data_path": null,
- "validation_data_path": null
- }
- },
- "gold_test_data_path": null
- },
- "output":{
- "conll_file":{
- "to_do": true,
- "metadata": true,
- "with_gold_labels": true
- },
- "txt_file":{
- "to_do": true,
- "metadata": true
- }
- }
-}
-
-
-
diff --git a/code/config_global_2.2.json b/code/config_global_2.2.json
deleted file mode 100644
index 395e75891f2c6a204c2e194eab4c9ffb08ec3572..0000000000000000000000000000000000000000
--- a/code/config_global_2.2.json
+++ /dev/null
@@ -1,49 +0,0 @@
-{
- "usecase_description": "Config file for usecase_2 : from a gold text, make predictions with an existing model, compare gold and predictions with metrics.",
- "data_raw": {
- "name": "eng.pdtb.pdtb_dev",
- "exte": ".conllu",
- "language": "en",
- "existing_metadata": true
- },
- "steps":{
- "main": "test",
- "pre-processing": {
- "to_do": false,
- "syntactic_tool": "stanza",
- "sentence_split": true,
- "tokenization": true,
- "syntactic_parsing": true,
- "create_metadata": {
- "to_do": true,
- "line": "paragraph",
- "sent": "sent"
- }
- },
- "discourse_segmenter": {
- "model": "/home/lriviere/andiamo/morteza/discut/Results_conllu/results_eng.pdtb.pdtb_bert/model.tar.gz",
- "training": {
- "toolkit": null,
- "pre_trained_lm": null,
- "config_file": null,
- "train_data_path": null,
- "validation_data_path": null
- }
- },
- "gold_test_data_path": null
- },
- "output":{
- "conll_file":{
- "to_do": true,
- "metadata": true,
- "with_gold_labels": true
- },
- "txt_file":{
- "to_do": true,
- "metadata": true
- }
- }
-}
-
-
-
diff --git a/code/config_global_3.json b/code/config_global_3.json
index ad453a45e30169b6ce0d95ca9e63fcd0986c8a4c..c24b4ff27dd98227b58b2e886e40017582902503 100644
--- a/code/config_global_3.json
+++ b/code/config_global_3.json
@@ -30,6 +30,7 @@
"validation_data_path": "eng.rst.rstdt_dev"
}
},
+ "evaluation": true,
"gold_test_data_path": "eng.rst.rstdt_dev"
},
"output":{
diff --git a/code/config_global_4.json b/code/config_global_4.json
index 9e1a95cd1e65a47fc99e3de9fa454d2a70393386..52e75d35240168369a151ec37f75b5796626d0b7 100644
--- a/code/config_global_4.json
+++ b/code/config_global_4.json
@@ -30,6 +30,7 @@
"validation_data_path": "/home/lriviere/andiamo/discut22/data/eng.sdrt.stac/eng.sdrt.stac_dev.conllu"
}
},
+ "evaluation": true,
"gold_test_data_path": "eng.rst.rstdt_dev"
},
"output":{
diff --git a/code/discut22_2.py b/code/discut22_2.py
index abdc9e58cbb7b5d5d08265553f0bf14eac3539cd..4c5caaacfef951694f827e867fa9c1e3cd866301 100644
--- a/code/discut22_2.py
+++ b/code/discut22_2.py
@@ -248,6 +248,7 @@ class Process:
self.crea_meta = infos['pre-processing']['create_metadata']['to_do']
self.meta_line = infos['pre-processing']['create_metadata']['line']
self.meta_sent = infos['pre-processing']['create_metadata']['sent']
+ self.eval = infos['evaluation']
if self.main == "train" or "fine_tune":
self.set_train = infos['discourse_segmenter']['training']['train_data_path']
@@ -261,11 +262,6 @@ class Process:
self.model = infos['discourse_segmenter']['model'] # ezpz for Tony
self.test_data = infos['gold_test_data_path']
- def get_evaluation_status(self):
- if self.main == "test" or self.main == "train" or self.main == "fine_tune":
- self.eval = True
- else:
- self.eval = False # "annotation"
def get_model(self):
self.model_path = ""
@@ -397,7 +393,6 @@ if __name__ == '__main__':
data.make_ner_format()
steps.get_model()
data.make_predictions(steps) # output allennlp JSON
- steps.get_evaluation_status()
if steps.eval == True:
data.evaluation(steps, prod)
else:
@@ -410,7 +405,6 @@ if __name__ == '__main__':
steps.update_training_config()
steps.training(data)
data.make_predictions(steps, js_name=steps.set_test, fi_ner=steps.test_ner)
- steps.get_evaluation_status()
if steps.eval == True:
data.evaluation(steps, prod, name=steps.test_data)
@@ -422,7 +416,6 @@ if __name__ == '__main__':
steps.update_training_config()
steps.fine_tuning(data)
data.make_predictions(steps, js_name=steps.set_test, fi_ner=steps.test_ner, model=steps.model_ft_path)
- steps.get_evaluation_status()
if steps.eval == True:
data.evaluation(steps, prod, name=steps.test_data, model=steps.model_ft_path)