]> code.communitydata.science - ml_measurement_error_public.git/blobdiff - simulations/Makefile
Add exploratory data analysis to come up with a real-data example.
[ml_measurement_error_public.git] / simulations / Makefile
index af54727127fbef6f3a961e329d7a8f57fa495624..b3ab77a3c62d773b5b577940c8e1e373bac955d5 100644 (file)
@@ -1,9 +1,9 @@
 
 SHELL=bash
 
-Ns=[1000, 2000, 4000]
-ms=[100, 200, 400, 800]
-seeds=[$(shell seq -s, 1 250)]
+Ns=[1000, 5000, 10000]
+ms=[100, 200, 400]
+seeds=[$(shell seq -s, 1 500)]
 explained_variances=[0.1]
 
 all:remembr.RDS remember_irr.RDS
@@ -23,21 +23,28 @@ joblists:example_1_jobs example_2_jobs example_3_jobs
 #      sbatch --wait --verbose --array=3001-6001 run_simulation.sbatch 0 test_true_z_jobs
 
 
-example_1_jobs: 01_two_covariates.R simulation_base.R grid_sweep.py
-       sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 01_two_covariates.R" --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["example_1.feather"], "y_explained_variance":${explained_variances}, "Bzx":[0.3]}' --outfile example_1_jobs
+example_1_jobs: 01_two_covariates.R simulation_base.R grid_sweep.py pl_methods.R
+       sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 01_two_covariates.R" --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["example_1.feather"], "y_explained_variance":${explained_variances}, "Bzx":[1]}' --outfile example_1_jobs
 
 example_1.feather: example_1_jobs 
        rm -f example_1.feather
-       sbatch --wait --verbose --array=1-$(shell cat example_1_jobs | wc -l) run_simulation.sbatch 0 example_1_jobs
-#      sbatch --wait --verbose --array=3001-6001 run_simulation.sbatch 0 example_1_jobs
+       sbatch --wait --verbose --array=1-1000  run_simulation.sbatch 0 example_1_jobs
+       sbatch --wait --verbose --array=1001-2000  run_simulation.sbatch 0 example_1_jobs
+       sbatch --wait --verbose --array=2001-3000  run_simulation.sbatch 0 example_1_jobs
+       sbatch --wait --verbose --array=3001-4000  run_simulation.sbatch 0 example_1_jobs
+       sbatch --wait --verbose --array=4001-$(shell cat example_1_jobs | wc -l)  run_simulation.sbatch 0 example_1_jobs
 
-example_2_jobs: 02_indep_differential.R simulation_base.R grid_sweep.py
-       sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 02_indep_differential.R" --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["example_2.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.3], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y*z*x"]}' --outfile example_2_jobs
+example_2_jobs: 02_indep_differential.R simulation_base.R grid_sweep.py pl_methods.R
+       sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 02_indep_differential.R" --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["example_2.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y*z*x"]}' --outfile example_2_jobs
 
 example_2.feather: example_2_jobs 
        rm -f example_2.feather
-       sbatch --wait --verbose --array=1-$(shell cat example_1_jobs | wc -l) run_simulation.sbatch 0 example_2_jobs
-#      sbatch --wait --verbose --array=3001-6001 run_simulation.sbatch 0 example_2_jobs
+       sbatch --wait --verbose --array=1-1000  run_simulation.sbatch 0 example_2_jobs
+       sbatch --wait --verbose --array=1001-2000  run_simulation.sbatch 0 example_2_jobs
+       sbatch --wait --verbose --array=2001-3000  run_simulation.sbatch 0 example_2_jobs
+       sbatch --wait --verbose --array=3001-4000  run_simulation.sbatch 0 example_2_jobs
+       sbatch --wait --verbose --array=4001-$(shell cat example_2_jobs | wc -l)  run_simulation.sbatch 0 example_2_jobs
+
 
 # example_2_B_jobs: example_2_B.R
 #      sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript example_2_B.R" --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["example_2_B.feather"]}' --outfile example_2_B_jobs
@@ -46,19 +53,28 @@ example_2.feather: example_2_jobs
 #      rm -f example_2_B.feather
 #      sbatch --wait --verbose --array=1-3000 run_simulation.sbatch 0 example_2_B_jobs
 
-example_3_jobs: 03_depvar.R simulation_base.R grid_sweep.py
-       sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 03_depvar.R" --arg_dict '{"N":${Ns},"m":${ms}, "Bxy":[0.01],"Bzy":[-0.01],"seed":${seeds}, "outfile":["example_3.feather"], "y_explained_variance":${explained_variances}}' --outfile example_3_jobs
+example_3_jobs: 03_depvar.R simulation_base.R grid_sweep.py pl_methods.R
+       sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 03_depvar.R" --arg_dict '{"N":${Ns},"m":${ms}, "Bxy":[0.7],"Bzy":[-0.7],"seed":${seeds}, "outfile":["example_3.feather"], "y_explained_variance":${explained_variances}}' --outfile example_3_jobs
 
 example_3.feather: example_3_jobs
        rm -f example_3.feather 
-       sbatch --wait --verbose --array=1-$(shell cat example_3_jobs | wc -l)  run_simulation.sbatch 0 example_3_jobs
+       sbatch --wait --verbose --array=1-1000  run_simulation.sbatch 0 example_3_jobs
+       sbatch --wait --verbose --array=1001-2000  run_simulation.sbatch 0 example_3_jobs
+       sbatch --wait --verbose --array=2001-3000  run_simulation.sbatch 0 example_3_jobs
+       sbatch --wait --verbose --array=3001-4000  run_simulation.sbatch 0 example_3_jobs
+       sbatch --wait --verbose --array=4001-$(shell cat example_3_jobs | wc -l)  run_simulation.sbatch 0 example_3_jobs
 
-example_4_jobs: 04_depvar_differential.R simulation_base.R grid_sweep.py
-       sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 04_depvar_differential.R" --arg_dict '{"N":${Ns},"Bxy":[0.01],"Bzy":[-0.01],"m":${ms}, "seed":${seeds}, "outfile":["example_4.feather"], "y_explained_variance":${explained_variances}}' --outfile example_4_jobs
+example_4_jobs: 04_depvar_differential.R simulation_base.R grid_sweep.py pl_methods.R
+       sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 04_depvar_differential.R" --arg_dict '{"N":${Ns},"Bxy":[0.7],"Bzy":[-0.7],"m":${ms}, "seed":${seeds}, "outfile":["example_4.feather"], "z_bias":[0.5]}' --outfile example_4_jobs
 
 example_4.feather: example_4_jobs
        rm -f example_4.feather 
-       sbatch --wait --verbose --array=1-$(shell cat example_4_jobs | wc -l)  run_simulation.sbatch 0 example_4_jobs
+       sbatch --wait --verbose --array=1-1000  run_simulation.sbatch 0 example_4_jobs
+       sbatch --wait --verbose --array=1001-2000  run_simulation.sbatch 0 example_4_jobs
+       sbatch --wait --verbose --array=2001-3000  run_simulation.sbatch 0 example_4_jobs
+       sbatch --wait --verbose --array=3001-4000  run_simulation.sbatch 0 example_4_jobs
+       sbatch --wait --verbose --array=4001-$(shell cat example_4_jobs | wc -l)  run_simulation.sbatch 0 example_4_jobs
+
 
 
 remembr.RDS:example_1.feather example_2.feather example_3.feather example_4.feather plot_example.R plot_dv_example.R summarize_estimator.R
@@ -69,30 +85,39 @@ remembr.RDS:example_1.feather example_2.feather example_3.feather example_4.feat
        ${srun} Rscript plot_dv_example.R --infile example_4.feather --name "plot.df.example.4"
 
 
-irr_Ns = ${Ns}
-irr_ms = ${ms}
+irr_Ns = [1000]
+irr_ms = [150,300,600]
 irr_seeds=${seeds}
 irr_explained_variances=${explained_variances}
+irr_coder_accuracy=[0.80]
 
-example_5_jobs: 05_irr_indep.R irr_simulation_base.R grid_sweep.py
-       sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 05_irr_indep.R" --arg_dict '{"N":${irr_Ns},"m":${irr_ms}, "seed":${irr_seeds}, "outfile":["example_5.feather"], "y_explained_variance":${irr_explained_variances}}' --outfile example_5_jobs
+example_5_jobs: 05_irr_indep.R irr_simulation_base.R grid_sweep.py pl_methods.R measerr_methods.R
+       sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 05_irr_indep.R" --arg_dict '{"N":${irr_Ns},"m":${irr_ms}, "seed":${irr_seeds}, "outfile":["example_5.feather"], "y_explained_variance":${irr_explained_variances}, "coder_accuracy":${irr_coder_accuracy}}' --outfile example_5_jobs
 
 example_5.feather:example_5_jobs
        rm -f example_5.feather
-       sbatch --wait --verbose --array=1-$(shell cat example_5_jobs | wc -l)  run_simulation.sbatch 0 example_5_jobs
+       sbatch --wait --verbose --array=1-1000  run_simulation.sbatch 0 example_5_jobs
+       sbatch --wait --verbose --array=1001-$(shell cat example_5_jobs | wc -l)  run_simulation.sbatch 1000 example_5_jobs
+       # sbatch --wait --verbose --array=2001-3000  run_simulation.sbatch 2000 example_5_jobs
+       # sbatch --wait --verbose --array=3001-4000  run_simulation.sbatch 3000 example_5_jobs
+       # sbatch --wait --verbose --array=2001-$(shell cat example_5_jobs | wc -l)  run_simulation.sbatch 4000 example_5_jobs
+
+
 
+# example_6_jobs: 06_irr_dv.R irr_dv_simulation_base.R grid_sweep.py pl_methods.R
+#      sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 06_irr_dv.R" --arg_dict '{"N":${irr_Ns},"m":${irr_ms}, "seed":${irr_seeds}, "outfile":["example_6.feather"], "y_explained_variance":${irr_explained_variances},"coder_accuracy":${irr_coder_accuracy}}' --outfile example_6_jobs
 
-example_6_jobs: 06_irr_dv.R irr_dv_simulation_base.R grid_sweep.py
-       sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 06_irr_dv.R" --arg_dict '{"N":${irr_Ns},"m":${irr_ms}, "seed":${irr_seeds}, "outfile":["example_6.feather"], "y_explained_variance":${irr_explained_variances}}' --outfile example_6_jobs
+# example_6.feather:example_6_jobs
+#      rm -f example_6.feather
+#      sbatch --wait --verbose --array=1-1000  run_simulation.sbatch 0 example_6_jobs
+#      sbatch --wait --verbose --array=1001-2000  run_simulation.sbatch 1000 example_6_jobs
+#      sbatch --wait --verbose --array=2001-$(shell cat example_6_jobs | wc -l)  run_simulation.sbatch 2000 example_6_jobs
 
-example_6.feather:example_6_jobs
-       rm -f example_6.feather
-       sbatch --wait --verbose --array=1-$(shell cat example_6_jobs | wc -l)  run_simulation.sbatch 0 example_6_jobs
 
-remember_irr.RDS: example_5.feather example_6.feather plot_irr_example.R plot_irr_dv_example.R summarize_estimator.R
+remember_irr.RDS: example_5.feather plot_irr_example.R plot_irr_dv_example.R summarize_estimator.R
        rm -f remember_irr.RDS
        sbatch --wait --verbose run_job.sbatch Rscript plot_irr_example.R --infile example_5.feather --name "plot.df.example.5"
-       sbatch --wait --verbose run_job.sbatch Rscript plot_irr_dv_example.R --infile example_6.feather --name "plot.df.example.6"
+#      sbatch --wait --verbose run_job.sbatch Rscript plot_irr_dv_example.R --infile example_6.feather --name "plot.df.example.6"
 
 
 

Community Data Science Collective || Want to submit a patch?