Contributor: Qiuchi Li
- Set up the configurations in config/config_of_the_run.ini
- python run.py -config config/config_of_the_run.ini
- python run.py -config config/download.ini (Data download)
- python run.py -config config/align.ini (Data alignment)
- python run.py -config config/extract_cmusdk.ini (Extract data to pickle file)
- Download data from https://github.com/declare-lab/conv-emotion
- python run.py -config config/pretrain_visual.ini (Optional, pretrain visual features, only for MELD)
- python run.py -config config/add_act.ini (Optional, adding dialogue act annotations)
- python run.py -config config/extract_conv_emotion.ini (Extract data to pickle file)
- python run.py -config config/add_act.ini
- mode = add_act.
- dataset_name in
{'meld','iemocap'}
. Name of the dataset. - in_path. Pickle file storing the original data (with no act information).
- out_path. Path storing the output pickle data (with act information).
- act_annotation_path. CSV file storing the dialogue act annotations.
- python run.py -config config/align.ini
- mode = align.
- dataset_name in
{'cmumosei','cmumosi','iemocap'}
. Name of the dataset. - dataset_type = multimodal.
- datasets_dir. Path storing all the datasets.
- align_function = avg. Alignment function. Only average is supported for the moment.
- modality in
{'acoustic', 'visual', 'textual'}
. The pivot modality for alignment. Other modalities will be aligned to this modality. - align_output_dir. Output directory storing the aligned features. Must be an existing path.
- python run.py -config config/download.ini
- mode = download.
- dataset_name in
{'cmumosei','cmumosi','iemocap'}
. Name of the dataset. - datasets_dir. Path where the dataset will be stored.
- python run.py -config config/extract_cmusdk.ini
- mode = extract_cmusdk.
- datasets_dir. Path storing all the datasets.
- data_aligned in
{'True', 'False'}
. Whether the data is in an aligned format. - data_dir. Name of the folder storing the dataset in CMU-MultimodalSDK format.
- dataset_type = multimodal.
- dataset_name in
{'cmumosei','cmumosi','iemocap'}
. Name of the dataset. - label = sentiment.
- max_seq_len. Length of a multimodal sequence. Zero-padding shorter sequence and truncating longer sequence.
- The output pickle file will be self.datasets_dir/cmusdk_data/
{'self.dataset_name'}
_{'self.label'}
.pkl
Extracting dialogue datasets in https://github.com/declare-lab/conv-emotion to pickle file
- python run.py -config config/extract_conv_emotion.ini
- mode = extract_conv_emotion.
- in_path. Pickle file storing the original data in Conv-Emotion format.
- out_path. Path storing the output pickle data.
- dataset_name in
{'meld','iemocap'}
. Name of the dataset. - max_seq_len. Length of a multimodal sequence. Zero-padding shorter sequence and truncating longer sequence.
- python run.py -config config/pretrain_visual.ini
- mode = extract_raw_feature.
- file_path. Pickle file storing the original data in Conv-Emotion format.
- stored_path. Folder of pickle files storing video clip data.
- videos_dir. Folder storing raw MELD video clips.
- output_path. Path storing the output data with pre-trained visual features, in Conv-Emotion format.
- Other network configurations.