Merge pull request #80 from mongulu-cm/kmer-support

Kmer support
mongulu-cm · Jun 21, 2023 · 446b311 · 446b311
2 parents 8bfd444 + 7a9c8f1
commit 446b311
Show file tree

Hide file tree

Showing 8 changed files with 19 additions and 5 deletions.
diff --git a/README.md b/README.md
@@ -84,4 +84,4 @@ Thanks goes to these wonderful people ([emoji key](https://allcontributors.org/d
 
 <!-- ALL-CONTRIBUTORS-LIST:END -->
 
-This project follows the [all-contributors](https://github.com/all-contributors/all-contributors) specification. Contributions of any kind welcome!!
+This project follows the [all-contributors](https://github.com/all-contributors/all-contributors) specification. Contributions of any kind welcome!!
diff --git a/etl/filter-cameroon.py b/etl/filter-cameroon.py
@@ -49,7 +49,7 @@
 
 
 def filter_cameroon(df):
-    return df[df['titre'].str.contains("CAMEROUN", case=False, na=False) | df['objet'].str.contains("CAMEROUN", case=False, na=False)]
+    return df[df['titre'].str.contains("CAMEROUN", case=False, na=False) | df['objet'].str.contains("CAMEROUN", case=False, na=False) | df['titre'].str.contains("KMER", case=False, na=False) | df['objet'].str.contains("KMER", case=False, na=False)]
 
 
 def remove_closed(df):
@@ -222,10 +222,9 @@ def select_relevant_columns(df):
 waldec_csv[6035] = "CULTURE, PRATIQUES D'ACTIVITÉS ARTISTIQUES, PRATIQUES CULTURELLES"
 waldec_csv[40510] = "ACTIVITÉS RELIGIEUSES, SPIRITUELLES OU PHILOSOPHIQUES"
 waldec_csv[17035] = "SANTÉ"
+waldec_csv[40580] = "ACTIVTÉS RELIGIEUSES, SPIRITUELLES OU PHILOSOPHIQUES"
 
 # %%
-
-
 def add_dept_and_region(df):
 
     def get_dept_region(code_postal):

diff --git a/etl/geocode_cache.sqlite b/etl/geocode_cache.sqlite
diff --git a/etl/prompt_address_cache/cache.db b/etl/prompt_address_cache/cache.db
diff --git a/etl/prompt_address_cache/cache.db-shm b/etl/prompt_address_cache/cache.db-shm
diff --git a/etl/prompt_address_cache/cache.db-wal b/etl/prompt_address_cache/cache.db-wal
diff --git a/etl/requirements.txt b/etl/requirements.txt
@@ -7,7 +7,6 @@ jupyter==1.0.0
 jupyter-utils==1.2.6
 pandarallel==1.6.1
 pipdeptree==2.2.1
-pyarrow==8.0.0
 requests-cache==0.9.4
 tqdm==4.64.0
 wheel==0.38.1
@@ -22,3 +21,4 @@ lambdaprompt==0.4.2
 openai==0.27.0
 boto3==1.26.82
 diskcache==5.4.0
+numpy==1.24.3
diff --git a/etl/rna-real-mars-2022-new.csv b/etl/rna-real-mars-2022-new.csv