Spaces:
Sleeping
Sleeping
Update src/core.py
Browse files- src/core.py +17 -27
src/core.py
CHANGED
|
@@ -4,16 +4,18 @@ from src.services.processor import *
|
|
| 4 |
global_tech, global_tech_embeddings = load_technologies()
|
| 5 |
|
| 6 |
|
| 7 |
-
def process_input(data, global_tech, global_tech_embeddings):
|
| 8 |
-
|
| 9 |
-
|
| 10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
constraints_stemmed = stem(constraints, "constraints")
|
| 12 |
|
| 13 |
save_dataframe(constraints_stemmed, "constraints_stemmed.xlsx")
|
| 14 |
|
| 15 |
-
#global_tech, keys, original_tech = preprocess_tech_data(df)
|
| 16 |
-
|
| 17 |
save_dataframe(global_tech, "global_tech.xlsx")
|
| 18 |
|
| 19 |
result_similarities, matrix = get_contrastive_similarities(constraints_stemmed, global_tech, global_tech_embeddings, )
|
|
@@ -28,25 +30,13 @@ def process_input(data, global_tech, global_tech_embeddings):
|
|
| 28 |
|
| 29 |
return best_technologies
|
| 30 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 31 |
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
constraints_stemmed = stem(constraints, "constraints")
|
| 35 |
-
|
| 36 |
-
save_dataframe(constraints_stemmed, "constraints_stemmed.xlsx")
|
| 37 |
-
|
| 38 |
-
#global_tech, keys, original_tech = preprocess_tech_data(df)
|
| 39 |
-
|
| 40 |
-
save_dataframe(global_tech, "global_tech.xlsx")
|
| 41 |
-
|
| 42 |
-
result_similarities, matrix = get_contrastive_similarities(constraints_stemmed, global_tech, global_tech_embeddings, )
|
| 43 |
-
|
| 44 |
-
save_to_pickle(result_similarities)
|
| 45 |
-
|
| 46 |
-
print(f"Matrix : {matrix} \n Constraints : {constraints_stemmed} \n Gloabl tech : {global_tech}")
|
| 47 |
-
|
| 48 |
-
best_combinations = find_best_list_combinations(constraints_stemmed, global_tech, matrix)
|
| 49 |
-
best_technologies_id = select_technologies(best_combinations)
|
| 50 |
-
best_technologies = get_technologies_by_id(best_technologies_id,global_tech)
|
| 51 |
-
|
| 52 |
-
return best_technologies
|
|
|
|
| 4 |
global_tech, global_tech_embeddings = load_technologies()
|
| 5 |
|
| 6 |
|
| 7 |
+
def process_input(data, global_tech, global_tech_embeddings, data_type):
|
| 8 |
+
if data_type == "problem":
|
| 9 |
+
prompt = set_prompt(data.problem)
|
| 10 |
+
constraints = retrieve_constraints(prompt)
|
| 11 |
+
|
| 12 |
+
elif data_type == "constraints":
|
| 13 |
+
constraints = data
|
| 14 |
+
|
| 15 |
constraints_stemmed = stem(constraints, "constraints")
|
| 16 |
|
| 17 |
save_dataframe(constraints_stemmed, "constraints_stemmed.xlsx")
|
| 18 |
|
|
|
|
|
|
|
| 19 |
save_dataframe(global_tech, "global_tech.xlsx")
|
| 20 |
|
| 21 |
result_similarities, matrix = get_contrastive_similarities(constraints_stemmed, global_tech, global_tech_embeddings, )
|
|
|
|
| 30 |
|
| 31 |
return best_technologies
|
| 32 |
|
| 33 |
+
def process_prior_art(technologies, data, data_type):
|
| 34 |
+
try:
|
| 35 |
+
prior_art_reponse = search_prior_art(technologies, data, data_type)
|
| 36 |
+
prior_art_search = add_citations_and_collect_uris(prior_art_reponse)
|
| 37 |
+
except Exception as e:
|
| 38 |
+
print(f"An error occured during the process, trying again : {e}")
|
| 39 |
+
prior_art_reponse = search_prior_art(technologies, data, data_type)
|
| 40 |
+
prior_art_search = add_citations_and_collect_uris(prior_art_reponse)
|
| 41 |
|
| 42 |
+
return prior_art_search
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|