diff --git a/modules/ml-pipeline/src/pipeline/src/configs/prepare_data.yaml b/modules/ml-pipeline/src/pipeline/src/configs/prepare_data.yaml index de82af2..c17b906 100644 --- a/modules/ml-pipeline/src/pipeline/src/configs/prepare_data.yaml +++ b/modules/ml-pipeline/src/pipeline/src/configs/prepare_data.yaml @@ -1,6 +1,6 @@ dataclient_type: minio data_location: s3://dev_bucket -train_proportion: 0.75 +train_proportion: 0.2 output_train_filepath: ./data/prepared_data/train.parquet output_test_filepath: ./data/prepared_data/test.parquet diff --git a/modules/ml-pipeline/src/pipeline/src/dvc.lock b/modules/ml-pipeline/src/pipeline/src/dvc.lock index 4f0e5d4..b4026e5 100644 --- a/modules/ml-pipeline/src/pipeline/src/dvc.lock +++ b/modules/ml-pipeline/src/pipeline/src/dvc.lock @@ -11,12 +11,12 @@ stages: configs/prepare_data.yaml: output_test_filepath: ./data/prepared_data/test.parquet output_train_filepath: ./data/prepared_data/train.parquet - train_proportion: 0.75 + train_proportion: 0.2 outs: - path: data/prepared_data/ hash: md5 - md5: 5cededc181e2b86c8ff63498a3e79462.dir - size: 36481 + md5: b9dd6ec988430bd4d4ec08824bf43b70.dir + size: 36061 nfiles: 2 build_model: cmd: python build_model.py @@ -27,8 +27,8 @@ stages: size: 3576 - path: data/prepared_data hash: md5 - md5: 5cededc181e2b86c8ff63498a3e79462.dir - size: 36481 + md5: b9dd6ec988430bd4d4ec08824bf43b70.dir + size: 36061 nfiles: 2 params: configs/build_model.yaml: @@ -41,7 +41,7 @@ stages: outs: - path: data/model/ hash: md5 - md5: e4904765c79b0139ccb80d56ef044383.dir + md5: 2d4977388d8acf7a99bd6351ddb03d71.dir size: 1096 nfiles: 1 generate_predictions: @@ -49,13 +49,13 @@ stages: deps: - path: data/model hash: md5 - md5: e4904765c79b0139ccb80d56ef044383.dir + md5: 2d4977388d8acf7a99bd6351ddb03d71.dir size: 1096 nfiles: 1 - path: data/prepared_data hash: md5 - md5: 5cededc181e2b86c8ff63498a3e79462.dir - size: 36481 + md5: b9dd6ec988430bd4d4ec08824bf43b70.dir + size: 36061 nfiles: 2 - path: generate_predictions.py hash: md5 @@ -68,21 +68,21 @@ stages: outs: - path: data/predictions/ hash: md5 - md5: d78af354c4ecb0b0296ac093e515a01b.dir - size: 2734 + md5: 25c8d2bd3ce9d1ab09c68aef722babad.dir + size: 4987 nfiles: 1 generate_metrics: cmd: python generate_metrics.py deps: - path: data/predictions hash: md5 - md5: d78af354c4ecb0b0296ac093e515a01b.dir - size: 2734 + md5: 25c8d2bd3ce9d1ab09c68aef722babad.dir + size: 4987 nfiles: 1 - path: data/prepared_data hash: md5 - md5: 5cededc181e2b86c8ff63498a3e79462.dir - size: 36481 + md5: b9dd6ec988430bd4d4ec08824bf43b70.dir + size: 36061 nfiles: 2 - path: generate_metrics.py hash: md5 @@ -95,5 +95,5 @@ stages: outs: - path: metrics/metrics.json hash: md5 - md5: a32cd5e65ca89db0920297a7c6ca39bf + md5: e7c575cadca7fe73d9b7659b71457815 size: 182 diff --git a/modules/ml-pipeline/src/pipeline/src/metrics/.gitignore b/modules/ml-pipeline/src/pipeline/src/metrics/.gitignore index fbadd1c..e69de29 100644 --- a/modules/ml-pipeline/src/pipeline/src/metrics/.gitignore +++ b/modules/ml-pipeline/src/pipeline/src/metrics/.gitignore @@ -1 +0,0 @@ -/metrics.json diff --git a/modules/ml-pipeline/src/pipeline/src/metrics/metrics.json b/modules/ml-pipeline/src/pipeline/src/metrics/metrics.json new file mode 100644 index 0000000..d5da667 --- /dev/null +++ b/modules/ml-pipeline/src/pipeline/src/metrics/metrics.json @@ -0,0 +1 @@ +{"mean_absolute_error": 46.93719007928077, "median_absolute_error": 40.45077035567739, "mean_squared_error": 3288.600558310513, "mean_absolute_percentage_error": 0.45448480340645764}