diff --git a/measure.py b/config/measure.py similarity index 97% rename from measure.py rename to config/measure.py index 1eb83e7..7ac8072 100644 --- a/measure.py +++ b/config/measure.py @@ -65,12 +65,12 @@ if __name__ == "__main__": ] models = [ - ("auto-genome-full-256.pt", 256, "autoencoder", files_genome + files_genome_cnn), - ("auto-genome-full-128.pt", 128, "autoencoder", files_genome + files_genome_cnn), + ("auto-genome-full-256.pt", 256, "autoencoder", files_genome), + ("auto-genome-full-128.pt", 128, "autoencoder", files_genome), ("cnn-genome-full-256.pt", 256, "cnn", files_genome_cnn), ("cnn-genome-full-128.pt", 128, "cnn", files_genome_cnn), - ("auto-enwik9-full-256.pt", 256, "autoencoder", files_enwik9 + files_enwik9_cnn), - ("auto-enwik9-full-128.pt", 128, "autoencoder", files_enwik9 + files_enwik9_cnn), + ("auto-enwik9-full-256.pt", 256, "autoencoder", files_enwik9), + ("auto-enwik9-full-128.pt", 128, "autoencoder", files_enwik9), ("cnn-enwik9-full-256.pt", 256, "cnn", files_enwik9_cnn), ("cnn-enwik9-full-128.pt", 128, "cnn", files_enwik9_cnn), ] diff --git a/measure_gzip_lz4.sh b/config/measure_gzip_lz4.sh similarity index 100% rename from measure_gzip_lz4.sh rename to config/measure_gzip_lz4.sh diff --git a/results/auto/AutoEncoder-losses-128-enwik9.csv b/models/autoencoder/AutoEncoder-losses-128-enwik9.csv similarity index 100% rename from results/auto/AutoEncoder-losses-128-enwik9.csv rename to models/autoencoder/AutoEncoder-losses-128-enwik9.csv diff --git a/results/auto/AutoEncoder-losses-128-enwik9.png b/models/autoencoder/AutoEncoder-losses-128-enwik9.png similarity index 100% rename from results/auto/AutoEncoder-losses-128-enwik9.png rename to models/autoencoder/AutoEncoder-losses-128-enwik9.png diff --git a/results/auto/AutoEncoder-losses-128-genome.csv b/models/autoencoder/AutoEncoder-losses-128-genome.csv similarity index 100% rename from results/auto/AutoEncoder-losses-128-genome.csv rename to models/autoencoder/AutoEncoder-losses-128-genome.csv diff --git a/results/auto/AutoEncoder-losses-128-genome.png b/models/autoencoder/AutoEncoder-losses-128-genome.png similarity index 100% rename from results/auto/AutoEncoder-losses-128-genome.png rename to models/autoencoder/AutoEncoder-losses-128-genome.png diff --git a/results/auto/AutoEncoder-losses-256-enwik9.csv b/models/autoencoder/AutoEncoder-losses-256-enwik9.csv similarity index 100% rename from results/auto/AutoEncoder-losses-256-enwik9.csv rename to models/autoencoder/AutoEncoder-losses-256-enwik9.csv diff --git a/results/auto/AutoEncoder-losses-256-enwik9.png b/models/autoencoder/AutoEncoder-losses-256-enwik9.png similarity index 100% rename from results/auto/AutoEncoder-losses-256-enwik9.png rename to models/autoencoder/AutoEncoder-losses-256-enwik9.png diff --git a/results/auto/AutoEncoder-losses-256-genome.csv b/models/autoencoder/AutoEncoder-losses-256-genome.csv similarity index 100% rename from results/auto/AutoEncoder-losses-256-genome.csv rename to models/autoencoder/AutoEncoder-losses-256-genome.csv diff --git a/results/auto/AutoEncoder-losses-256-genome.png b/models/autoencoder/AutoEncoder-losses-256-genome.png similarity index 100% rename from results/auto/AutoEncoder-losses-256-genome.png rename to models/autoencoder/AutoEncoder-losses-256-genome.png diff --git a/results/cnn/CNNPredictor-losses-128-enwik9.csv b/models/cnn/CNNPredictor-losses-128-enwik9.csv similarity index 100% rename from results/cnn/CNNPredictor-losses-128-enwik9.csv rename to models/cnn/CNNPredictor-losses-128-enwik9.csv diff --git a/results/cnn/CNNPredictor-losses-128-enwik9.png b/models/cnn/CNNPredictor-losses-128-enwik9.png similarity index 100% rename from results/cnn/CNNPredictor-losses-128-enwik9.png rename to models/cnn/CNNPredictor-losses-128-enwik9.png diff --git a/results/cnn/CNNPredictor-losses-128-genome.csv b/models/cnn/CNNPredictor-losses-128-genome.csv similarity index 100% rename from results/cnn/CNNPredictor-losses-128-genome.csv rename to models/cnn/CNNPredictor-losses-128-genome.csv diff --git a/results/cnn/CNNPredictor-losses-128-genome.png b/models/cnn/CNNPredictor-losses-128-genome.png similarity index 100% rename from results/cnn/CNNPredictor-losses-128-genome.png rename to models/cnn/CNNPredictor-losses-128-genome.png diff --git a/results/cnn/CNNPredictor-losses-256-enwik9.csv b/models/cnn/CNNPredictor-losses-256-enwik9.csv similarity index 100% rename from results/cnn/CNNPredictor-losses-256-enwik9.csv rename to models/cnn/CNNPredictor-losses-256-enwik9.csv diff --git a/results/cnn/CNNPredictor-losses-256-enwik9.png b/models/cnn/CNNPredictor-losses-256-enwik9.png similarity index 100% rename from results/cnn/CNNPredictor-losses-256-enwik9.png rename to models/cnn/CNNPredictor-losses-256-enwik9.png diff --git a/results/cnn/CNNPredictor-losses-256-genome.csv b/models/cnn/CNNPredictor-losses-256-genome.csv similarity index 100% rename from results/cnn/CNNPredictor-losses-256-genome.csv rename to models/cnn/CNNPredictor-losses-256-genome.csv diff --git a/results/cnn/CNNPredictor-losses-256-genome.png b/models/cnn/CNNPredictor-losses-256-genome.png similarity index 100% rename from results/cnn/CNNPredictor-losses-256-genome.png rename to models/cnn/CNNPredictor-losses-256-genome.png diff --git a/pyproject.toml b/pyproject.toml index 97b31c3..96c0308 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -9,7 +9,7 @@ dependencies = [ "fsspec==2024.9.0", "lorem>=0.1.1", "arithmeticencodingpython", - "pandas-stubs~=2.3.3", + "pandas-stubs==2.3.3.251201", "seaborn>=0.13.2", ] diff --git a/results/compress/compression_results_auto_small.csv b/results/compress/compression_results_auto_small.csv deleted file mode 100644 index 0cd438e..0000000 --- a/results/compress/compression_results_auto_small.csv +++ /dev/null @@ -1,13 +0,0 @@ -model_type,model_name,context_length,dataset_type,input_file_name,original_file_size,compressed_file_size,mse_loss,compression_time,decompression_time -autoencoder,auto-genome-full-256.pt,256,genome,genome_small.fna,1367,1392,220.29188537597656,237708436,29378622 -autoencoder,auto-genome-full-256.pt,256,genome,genome_xsmall.fna,1043,1160,263.97027587890625,5293080,4488761 -autoencoder,auto-genome-full-256.pt,256,genome,genome_xxsmall.fna,800,928,319.5350036621094,4873340,4381310 -autoencoder,auto-genome-full-128.pt,128,genome,genome_small.fna,1367,2816,120.52304077148438,251659317,5284661 -autoencoder,auto-genome-full-128.pt,128,genome,genome_xsmall.fna,1043,2304,133.8379669189453,152590538,4432469 -autoencoder,auto-genome-full-128.pt,128,genome,genome_xxsmall.fna,800,1792,159.2862548828125,153622653,4532139 -autoencoder,auto-enwik9-full-256.pt,256,enwik9,text_small.txt,1022,976,746.8033447265625,6699780,22065927 -autoencoder,auto-enwik9-full-256.pt,256,enwik9,text_xsmall.txt,825,976,787.5345458984375,3715750,3665629 -autoencoder,auto-enwik9-full-256.pt,256,enwik9,text_xxsmall.txt,492,488,828.8658447265625,4892560,4176979 -autoencoder,auto-enwik9-full-128.pt,128,enwik9,text_small.txt,1022,2016,231.42955017089844,5005880,4400159 -autoencoder,auto-enwik9-full-128.pt,128,enwik9,text_xsmall.txt,825,1764,236.55636596679688,5024449,4180100 -autoencoder,auto-enwik9-full-128.pt,128,enwik9,text_xxsmall.txt,492,1008,237.20529174804688,4856779,4145850 diff --git a/results/compress/compression_metrics_20251216_182603.csv b/results/compression_metrics_20251216_182603.csv similarity index 100% rename from results/compress/compression_metrics_20251216_182603.csv rename to results/compression_metrics_20251216_182603.csv diff --git a/results/compress/compression_results.csv b/results/compression_results.csv similarity index 100% rename from results/compress/compression_results.csv rename to results/compression_results.csv diff --git a/cpu_compression_graphs.py b/results/cpu_compression_graphs.py similarity index 97% rename from cpu_compression_graphs.py rename to results/cpu_compression_graphs.py index 6bee7b8..3e293f1 100644 --- a/cpu_compression_graphs.py +++ b/results/cpu_compression_graphs.py @@ -110,5 +110,5 @@ def plot_compression_metrics(csv_path, output_dir="results/compress/plots"): if __name__ == "__main__": # You can modify this path to point to your CSV file - csv_path = "results/compress/compression_metrics_20251216_182603.csv" + csv_path = "compression_metrics_20251216_182603.csv" plot_compression_metrics(csv_path) \ No newline at end of file diff --git a/graphs/autoencoder_enwik9_compression_ratio.png b/results/graphs/autoencoder_enwik9_compression_ratio.png similarity index 100% rename from graphs/autoencoder_enwik9_compression_ratio.png rename to results/graphs/autoencoder_enwik9_compression_ratio.png diff --git a/graphs/autoencoder_enwik9_execution_time.png b/results/graphs/autoencoder_enwik9_execution_time.png similarity index 100% rename from graphs/autoencoder_enwik9_execution_time.png rename to results/graphs/autoencoder_enwik9_execution_time.png diff --git a/graphs/autoencoder_genome_compression_ratio.png b/results/graphs/autoencoder_genome_compression_ratio.png similarity index 100% rename from graphs/autoencoder_genome_compression_ratio.png rename to results/graphs/autoencoder_genome_compression_ratio.png diff --git a/graphs/autoencoder_genome_execution_time.png b/results/graphs/autoencoder_genome_execution_time.png similarity index 100% rename from graphs/autoencoder_genome_execution_time.png rename to results/graphs/autoencoder_genome_execution_time.png diff --git a/graphs/autoencoder_loss.png b/results/graphs/autoencoder_loss.png similarity index 100% rename from graphs/autoencoder_loss.png rename to results/graphs/autoencoder_loss.png diff --git a/graphs/cnn_enwik9_compression_ratio.png b/results/graphs/cnn_enwik9_compression_ratio.png similarity index 100% rename from graphs/cnn_enwik9_compression_ratio.png rename to results/graphs/cnn_enwik9_compression_ratio.png diff --git a/graphs/cnn_enwik9_execution_time.png b/results/graphs/cnn_enwik9_execution_time.png similarity index 100% rename from graphs/cnn_enwik9_execution_time.png rename to results/graphs/cnn_enwik9_execution_time.png diff --git a/graphs/cnn_enwik9_extrapolated_execution_time.png b/results/graphs/cnn_enwik9_extrapolated_execution_time.png similarity index 100% rename from graphs/cnn_enwik9_extrapolated_execution_time.png rename to results/graphs/cnn_enwik9_extrapolated_execution_time.png diff --git a/graphs/cnn_genome_compression_ratio.png b/results/graphs/cnn_genome_compression_ratio.png similarity index 100% rename from graphs/cnn_genome_compression_ratio.png rename to results/graphs/cnn_genome_compression_ratio.png diff --git a/graphs/cnn_genome_execution_time.png b/results/graphs/cnn_genome_execution_time.png similarity index 100% rename from graphs/cnn_genome_execution_time.png rename to results/graphs/cnn_genome_execution_time.png diff --git a/graphs/cnn_genome_extrapolated_execution_time.png b/results/graphs/cnn_genome_extrapolated_execution_time.png similarity index 100% rename from graphs/cnn_genome_extrapolated_execution_time.png rename to results/graphs/cnn_genome_extrapolated_execution_time.png diff --git a/graphs/cnn_loss.png b/results/graphs/cnn_loss.png similarity index 100% rename from graphs/cnn_loss.png rename to results/graphs/cnn_loss.png diff --git a/graphs/plots/compression_ratio_comparison.png b/results/graphs/plots/compression_ratio_comparison.png similarity index 100% rename from graphs/plots/compression_ratio_comparison.png rename to results/graphs/plots/compression_ratio_comparison.png diff --git a/graphs/plots/compression_time_comparison.png b/results/graphs/plots/compression_time_comparison.png similarity index 100% rename from graphs/plots/compression_time_comparison.png rename to results/graphs/plots/compression_time_comparison.png diff --git a/graphs/plots/decompression_time_comparison.png b/results/graphs/plots/decompression_time_comparison.png similarity index 100% rename from graphs/plots/decompression_time_comparison.png rename to results/graphs/plots/decompression_time_comparison.png diff --git a/graphs/plots/ratio_vs_time.png b/results/graphs/plots/ratio_vs_time.png similarity index 100% rename from graphs/plots/ratio_vs_time.png rename to results/graphs/plots/ratio_vs_time.png diff --git a/make_graphs.py b/results/make_graphs.py similarity index 98% rename from make_graphs.py rename to results/make_graphs.py index 8380030..f02fe24 100644 --- a/make_graphs.py +++ b/results/make_graphs.py @@ -5,7 +5,7 @@ import numpy as np if __name__ == "__main__": # read in the csv - df = pd.read_csv("./results/compress/compression_results.csv") + df = pd.read_csv("compression_results.csv") for dataset_type in df["dataset_type"].unique(): for model_type in df["model_type"].unique(): diff --git a/uv.lock b/uv.lock index 95a3d23..b6f68cf 100644 --- a/uv.lock +++ b/uv.lock @@ -1761,7 +1761,7 @@ requires-dist = [ { name = "matplotlib", marker = "extra == 'dev'", specifier = ">=3.10.7" }, { name = "memray", marker = "extra == 'dev'", specifier = ">=1.19.1" }, { name = "optuna", marker = "extra == 'dev'", specifier = "==4.5.0" }, - { name = "pandas-stubs", specifier = "~=2.3.3" }, + { name = "pandas-stubs", specifier = "==2.3.3.251201" }, { name = "regex", marker = "extra == 'dataset'", specifier = ">=2025.11.3" }, { name = "seaborn", specifier = ">=0.13.2" }, { name = "torch", marker = "extra == 'dev'", specifier = "==2.9.0" },