junte/ML
1
0
Fork 0
mirror of https://github.com/Hestia-Homes/ML.git synced 2026-06-08 11:17:25 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge pull request #90 from Hestia-Homes/sap-dev-model

Browse source 
Sap dev model
This commit is contained in:
KhalimCK 2024-01-16 17:37:12 +00:00 • committed by GitHub
commit 03364036db
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
13 changed files with 79 additions and 72 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
.github/workflows/Deploy.yml vendored
View file

@ -2,7 +2,7 @@ name: Sap Change Model Deploy
on:
push:
branches: [ sap-dev, sap-prod ]
branches: [ sap-dev, sap-prod, heat-dev, heat-prod, carbon-dev, carbon-prod]
jobs:
deploy:

3
modules/ml-pipeline/.dvc/.gitignore vendored
View file

@ -1,3 +0,0 @@
/config.local
/tmp
/cache

2
modules/ml-pipeline/.dvc/config
View file

@ -1,2 +0,0 @@
['remote "myremote"']
url = /tmp/dvcstore

3
modules/ml-pipeline/.dvcignore
View file

@ -1,3 +0,0 @@
# Add patterns of files dvc should ignore, which could improve
# the performance. Learn more at
# https://dvc.org/doc/user-guide/dvcignore

2
modules/ml-pipeline/.gto
View file

@ -1,2 +0,0 @@
# .gto config file
stages: [dev, stage, prod] # list of allowed Stages

1
modules/ml-pipeline/src/pipeline/2_build_model.py
View file

@ -67,7 +67,6 @@ def build_model(
test_data: Union[pd.DataFrame, None] = None,
pipeline_mode: bool = False,
):
logger.info("--- Loading Data for build process ---")
if train_data is None:

2
modules/ml-pipeline/src/pipeline/configs/build_model.yaml
View file

@ -13,7 +13,7 @@ default:
output_filepath: ./data/model/allmodels/
problem_type: regression
eval_metric: mean_squared_error #mean_absolute_error
time_limit: 4000
time_limit: 400
presets: medium_quality
excluded_model_types: ['KNN', 'RF']
infer_limit: 0.05

10
modules/ml-pipeline/src/pipeline/configs/feature_processor_logic.py
View file

@ -9,11 +9,11 @@ Business Logic dict + functions
def remove_starting_columns(df):
keep_column_index = [
False if col_name.endswith("_STARTING") else True
False if col_name.endswith("_starting") else True
for col_name in list(df.columns)
]
keep_columns = df.columns[keep_column_index].to_list()
keep_columns.append("SAP_STARTING")
keep_columns.append("sap_starting")
df = df[keep_columns]
return df
@ -22,7 +22,7 @@ def remove_floor_height_ending(df):
# df.describe(percentiles=[0.005,0.99])['FLOOR_HEIGHT_ENDING']
# shows bottom 0.5 percentile is 1.665
# So keep anything above this
df = df[df["FLOOR_HEIGHT_ENDING"] > 1.665].reset_index(drop=True)
df = df[df["floor_height_ending"] > 1.665].reset_index(drop=True)
print("we in here")
return df
@ -30,13 +30,13 @@ def remove_floor_height_ending(df):
def remove_minimum_habitable_room_size(df):
# Need minimum of 6.5m per habitable room
df = df[
df["TOTAL_FLOOR_AREA_ENDING"] / df["NUMBER_HABITABLE_ROOMS"] > 6.5
df["total_floor_area_ending"] / df["number_habitable_rooms"] > 6.5
].reset_index(drop=True)
return df
def keep_flats(df):
df = df[df["PROPERTY_TYPE"] == "Flat"]
df = df[df["property_type"] == "Flat"]
return df

4
modules/ml-pipeline/src/pipeline/configs/post_prediction_logic.py
View file

@ -12,9 +12,9 @@ def clip_predictions_to_minimum_value(
predictions.name = "predictions"
predictions_df = pd.concat([data, predictions], axis=1)
# We expect all prediction to be atleast one point improvement
replace_index = predictions_df["SAP_STARTING"] + 1 > predictions_df["predictions"]
replace_index = predictions_df["sap_starting"] + 1 > predictions_df["predictions"]
predictions_df.loc[replace_index, "predictions"] = (
predictions_df.loc[replace_index, "SAP_STARTING"] + minimum_value
predictions_df.loc[replace_index, "sap_starting"] + minimum_value
)
predictions_new = predictions_df["predictions"]

9
modules/ml-pipeline/src/pipeline/configs/settings.yaml
View file

@ -21,7 +21,8 @@ default:
# data_filepath: s3://retrofit-data-dev/sap_change_model/dataset_with_differencing.parquet
# data_filepath: s3://retrofit-data-dev/sap_change_model/floor_area_clean_test.parquet
# data_filepath: s3://retrofit-data-dev/sap_change_model/dataset_without_differencing.parquet
data_filepath: s3://retrofit-data-dev/sap_change_model/dataset_test.parquet
# data_filepath: s3://retrofit-data-dev/sap_change_model/dataset_test.parquet
data_filepath: s3://retrofit-data-dev/sap_change_model/dataset_refactor.parquet
train_proportion: 0.9
output_train_filepath: ./data/prepared_data/train.parquet
output_test_filepath: ./data/prepared_data/test.parquet
@ -31,9 +32,9 @@ default:
feature_processor_config:
subsample_amount: null
subsample_seed: 0
target: SAP_ENDING
identifier_columns: ["UPRN"]
drop_columns: ["HEAT_DEMAND_CHANGE", "CARBON_CHANGE", "RDSAP_CHANGE", "HEAT_DEMAND_ENDING", "CARBON_ENDING"]
target: sap_ending
identifier_columns: ["uprn"]
drop_columns: ["heat_demand_change", "carbon_change", "rdsap_change", "heat_demand_ending", "carbon_ending"]
# retain_features: ["SAP_STARTING", "TOTAL_FLOOR_AREA_DIFF"]
retain_features: null

62
modules/ml-pipeline/src/pipeline/dvc.lock
View file

@ -10,17 +10,17 @@ stages:
params:
configs/settings.yaml:
default.feature_processor.feature_processor_config.drop_columns:
- HEAT_DEMAND_CHANGE
- CARBON_CHANGE
- RDSAP_CHANGE
- HEAT_DEMAND_ENDING
- CARBON_ENDING
- heat_demand_change
- carbon_change
- rdsap_change
- heat_demand_ending
- carbon_ending
default.feature_processor.feature_processor_config.retain_features:
default.feature_processor.feature_processor_config.subsample_amount:
default.feature_processor.feature_processor_config.subsample_seed: 0
default.feature_processor.feature_processor_config.target: SAP_ENDING
default.feature_processor.feature_processor_config.target: sap_ending
default.feature_processor.feature_processor_type: dataframe
default.prepare_data.data_filepath: s3://retrofit-data-dev/sap_change_model/dataset_test.parquet
default.prepare_data.data_filepath: s3://retrofit-data-dev/sap_change_model/dataset_refactor.parquet
default.prepare_data.input_dataclient_type: aws-s3
default.prepare_data.output_dataclient_type: local
default.prepare_data.output_test_filepath: ./data/prepared_data/test.parquet
@ -29,20 +29,20 @@ stages:
outs:
- path: data/prepared_data/
hash: md5
md5: 6bfdb621b608648c017bf2323f7b5052.dir
size: 37048968
md5: 5d29397fcafe6b3dc4d51ffaf1e55239.dir
size: 39303409
nfiles: 2
build_model:
cmd: python 2_build_model.py
deps:
- path: 2_build_model.py
hash: md5
md5: 7b79f280b8b0d5bc6f07669e7cc37c6a
size: 4150
md5: b824822475c222521516493e68eef9c5
size: 4149
- path: data/prepared_data
hash: md5
md5: 6bfdb621b608648c017bf2323f7b5052.dir
size: 37048968
md5: 5d29397fcafe6b3dc4d51ffaf1e55239.dir
size: 39303409
nfiles: 2
params:
configs/build_model.yaml:
@ -58,7 +58,7 @@ stages:
output_filepath: ./data/model/allmodels/
problem_type: regression
eval_metric: mean_squared_error
time_limit: 4000
time_limit: 400
presets: medium_quality
excluded_model_types:
- KNN
@ -68,13 +68,13 @@ stages:
outs:
- path: data/model/
hash: md5
md5: f2999107de7572ea5ff0f2d774fa83b8.dir
size: 424943352
nfiles: 27
md5: 6265dafedf579905c31c676e81c2a9c7.dir
size: 344212462
nfiles: 24
- path: metrics/fit_metrics.json
hash: md5
md5: 9537e7ebc2eb32b421a7cabd2005f00b
size: 223
md5: 5cd6b92af1b1df753e20e9ea33629c4d
size: 224
generate_predictions:
cmd: python 3_generate_predictions.py
deps:
@ -84,13 +84,13 @@ stages:
size: 2464
- path: data/model
hash: md5
md5: f2999107de7572ea5ff0f2d774fa83b8.dir
size: 424943352
nfiles: 27
md5: 6265dafedf579905c31c676e81c2a9c7.dir
size: 344212462
nfiles: 24
- path: data/prepared_data
hash: md5
md5: 6bfdb621b608648c017bf2323f7b5052.dir
size: 37048968
md5: 5d29397fcafe6b3dc4d51ffaf1e55239.dir
size: 39303409
nfiles: 2
params:
configs/settings.yaml:
@ -102,8 +102,8 @@ stages:
outs:
- path: data/predictions/
hash: md5
md5: f4439a56669f84bc51a9fcb4cd08353f.dir
size: 346539
md5: b130faf5117b06897b2deed97f5868ee.dir
size: 367038
nfiles: 1
generate_metrics:
cmd: python 4_generate_metrics.py
@ -114,13 +114,13 @@ stages:
size: 3484
- path: data/predictions
hash: md5
md5: f4439a56669f84bc51a9fcb4cd08353f.dir
size: 346539
md5: b130faf5117b06897b2deed97f5868ee.dir
size: 367038
nfiles: 1
- path: data/prepared_data
hash: md5
md5: 6bfdb621b608648c017bf2323f7b5052.dir
size: 37048968
md5: 5d29397fcafe6b3dc4d51ffaf1e55239.dir
size: 39303409
nfiles: 2
params:
configs/settings.yaml:
@ -130,7 +130,7 @@ stages:
outs:
- path: metrics/metrics.json
hash: md5
md5: 357904cf106279be5a578e8faefa5d80
md5: 3900cc1697d6d7308728b3d5b3025f85
size: 224
startup_cleanup:
cmd: python 0_startup_cleanup.py

43
modules/ml-pipeline/src/pipeline/eda.py
View file

@ -190,28 +190,35 @@ prediction_analysis_params = settings.prediction_analysis
model = model_factory(build_model_params["model_type"])
model.load_model(build_model_params["model_save_filepath"])
dataclient_type = prediction_analysis_params["dataclient_type"]
dataclient = dataclient_factory(
dataclient_type=dataclient_type,
dataclient_config=client_params[dataclient_type],
)
# dataclient_type = 'aws-s3'
# dataclient = dataclient_factory(
# dataclient_type=dataclient_type,
# dataclient_config=client_params[dataclient_type],
# )
# data = dataclient.load_data("s3://retrofit-data-dev/sap_change_model/dataset.parquet")
target = feature_process_params["feature_processor_config"]["target"]
predictions_column_name = generate_predictions_params["predictions_column_name"]
output_test_filepath = prepare_data_params["output_test_filepath"]
predictions_output_filepath = generate_predictions_params["predictions_output_filepath"]
test_df = dataclient.load_data(output_test_filepath)
predictions = dataclient.load_data(predictions_output_filepath)
# score_data = dataclient.load_data("s3://retrofit-data-dev/carbon_change_predictions/51/2023-11-28T21:01:21.869339.parquet")
local_dataclient = dataclient_factory(
dataclient_type="local",
dataclient_config=client_params["local"],
)
test_df = local_dataclient.load_data(output_test_filepath)
predictions = local_dataclient.load_data(predictions_output_filepath)
mix_df = pd.concat([test_df.copy(), predictions], axis=1)
mix_df["residual"] = abs(mix_df[predictions_column_name] - mix_df[target])
mix_df = mix_df.sort_values("residual", ascending=False)
cosine_similarity_df = mix_df[
mix_df.columns.difference(["UPRN", "predictions", "residual", "SAP_ENDING"])
]
cosine_similarity_df = mix_df[mix_df.columns.difference(["predictions", "residual"])]
from sklearn.metrics.pairwise import cosine_similarity
row_index = 20695
row_index = 0
from sklearn.preprocessing import LabelEncoder
@ -224,8 +231,18 @@ cosine_similarity_df[object_columns.columns] = cosine_similarity_df[
feature_vector = cosine_similarity_df.loc[[row_index]]
cosine_similarity_df["cosine"] = cosine_similarity(cosine_similarity_df, feature_vector)
similar_df = cosine_similarity_df.sort_values("cosine", ascending=False).head(5)
similar_index = similar_df.index
similar_index = (
cosine_similarity_df.sort_values("cosine", ascending=False).head(15).index
)
check_df = mix_df.loc[similar_index]
columns_to_check = [
"LOW_ENERGY_LIGHTING_ENDING",
"walls_thermal_transmittance_ENDING",
"floor_thermal_transmittance_ENDING",
"roof_thermal_transmittance_ENDING",
"roof_insulation_thickness_ENDING",
]
cosine_similarity_df = mix_df[columns_to_check]

8
modules/ml-pipeline/src/pipeline/requirements/version_control/requirements.txt
View file

@ -1,4 +1,4 @@
dvc==3.18.0
dvc-s3==2.23.0
gto==1.0.4
pyOpenSSL==23.2.0
dvc==3.36.0
dvc-s3==3.0.1
gto==1.6.1
pyOpenSSL==23.3.0