diff --git a/README.md b/README.md index b0c6f8ff3da20948959b88595cae61f82ad3c2e1..0741f1403d20baff469f3895407ed12c32f042a9 100644 --- a/README.md +++ b/README.md @@ -21,51 +21,51 @@ For more details please read our project report. # Project structure - |_ cli.py - | The command line interface for our project. Imports and runs the files in scripts/. See Usage for more info. - | - |_ data/ - | All necessary inputs and the generated outputs. - | | - | |_ cora - | | |_embeddings - the output of EP on Cora - | | |_graph - the input for EP - a pickled networkX graph - | | |_models - the saved TF models; can be used to save the embeddings after 40 epochs, for instance - | | |_raw - the raw Cora data, input for the Cora preprocessing and the node classification - | | |_summaries - the TF loss summaries for the two Cora label types; used to produce the figures in the report - | | - | |_ other - files that don't belong to a particular dataset - | | - | |_ senseval2/senseval3 - | | |_processed - the processed raw S2 or S3 data; input for the WSD - | | |_raw - the raw S2 or S3 data - | | |_wsd-answers - outputs of the WSD on the S2 or S3 data - | | - | |_ wordnet - | |_embeddings - the output of EP on WordNet - | |_graph - the input for EP - a pickled networkX graph - | |_models - the saved TF models; can be used to save the embeddings after 50 epochs, for instance - | |_raw - the raw WN data, input for the preprocessing scripts - | |_summaries - the TF loss summaries for the five WN label types; used to produce the figures in the report - | |_mappings - various mappings for the synset IDs, lemmata, WN3.0->WN1.7 etc.; used in the WSD - | - |_ __init__.py - | - |_ README.md - | This file. - | - |_ requirements.txt - | An image of the virtualenv. -> pip install -r requirements.txt - | - |_ scripts/ - | Python scripts for EP, preprocessing, node classification and WSD. - | | - | |_ embedding_propagation - the EP algorithm - | |_ node_classification - the NC experiment on the Cora dataset - | |_ preprocessing - preprocessing scripts for Cora, SensEval, and WordNet. Not part of the CLI, so partly with aux. files. - | |_ scoring - the official S2 and S3 All Words Task scorer for the WSD - | |_ wsd - the two WSD methods - | + |_ cli.py + | The command line interface for our project. Imports and runs the files in scripts/. See Usage for more info. + | + |_ data/ + | All necessary inputs and the generated outputs. + | | + | |_ cora/ + | | |_embeddings/ - the output of EP on Cora + | | |_graph/ - the input for EP - a pickled networkX graph + | | |_models/ - the saved TF models; can be used to save the embeddings after 40 epochs, for instance + | | |_raw/ - the raw Cora data, input for the Cora preprocessing and the node classification + | | |_summaries/ - the TF loss summaries for the two Cora label types; used to produce the figures in the report + | | + | |_ other/ - files that don't belong to a particular dataset + | | + | |_ senseval2/senseval3 + | | |_processed/ - the processed raw S2 or S3 data; input for the WSD + | | |_raw/ - the raw S2 or S3 data + | | |_wsd-answers/ - outputs of the WSD on the S2 or S3 data + | | + | |_ wordnet/ + | |_embeddings/ - the output of EP on WordNet + | |_graph/ - the input for EP - a pickled networkX graph + | |_models/ - the saved TF models; can be used to save the embeddings after 50 epochs, for instance + | |_raw/ - the raw WN data, input for the preprocessing scripts + | |_summaries/ - the TF loss summaries for the five WN label types; used to produce the figures in the report + | |_mappings/ - various mappings for the synset IDs, lemmata, WN3.0->WN1.7 etc.; used in the WSD + | + |_ __init__.py + | + |_ README.md + | This file. + | + |_ requirements.txt + | An image of the virtualenv. -> pip install -r requirements.txt + | + |_ scripts/ + | Python scripts for EP, preprocessing, node classification and WSD. + | | + | |_ embedding_propagation/ - the EP algorithm + | |_ node_classification/ - the NC experiment on the Cora dataset + | |_ preprocessing/ - preprocessing scripts for Cora, SensEval, and WordNet. Not part of the CLI, so partly with aux. files. + | |_ scoring/ - the official S2 and S3 All Words Task scorer for the WSD + | |_ wsd/ - the two WSD methods + | # Usage