oasislmf.pytools.fm.stream_sparse¶
Attributes¶
Classes¶
when reading the stream we store relenvant value into a slithly modified version of the CSR sparse matrix where |
|
Functions¶
|
|
|
|
|
|
|
|
|
|
|
Module Contents¶
- oasislmf.pytools.fm.stream_sparse.reset_empty_items(compute_idx, sidx_indptr, sidx_val, loss_val, computes)[source]¶
- oasislmf.pytools.fm.stream_sparse.add_new_loss(sidx, loss, compute_i, sidx_indptr, sidx_val, loss_val)[source]¶
- oasislmf.pytools.fm.stream_sparse.event_log_msg(event_id, sidx_indptr, len_array, node_count)[source]¶
- oasislmf.pytools.fm.stream_sparse.read_buffer(byte_mv, cursor, valid_buff, event_id, item_id, nodes_array, sidx_indexes, sidx_indptr, sidx_val, loss_indptr, loss_val, pass_through, computes, compute_idx)[source]¶
- class oasislmf.pytools.fm.stream_sparse.FMReader(nodes_array, sidx_indexes, sidx_indptr, sidx_val, loss_indptr, loss_val, pass_through, len_array, computes, compute_idx)[source]¶
Bases:
oasislmf.pytools.common.event_stream.EventReader
when reading the stream we store relenvant value into a slithly modified version of the CSR sparse matrix where the column indices for row i are stored in indices[indptr[i]:indptr[i+1]] and their corresponding values are stored in data[indptr[i]:indptr[i+1]].
nodes_array: array containing all the static information on the nodes loss_indptr: array containing the indexes of the beginning and end of samples of an item loss_sidx: array containing the sidx of the samples loss_val: array containing the loss of the samples
- oasislmf.pytools.fm.stream_sparse.load_event(byte_mv, event_id, nodes_array, sidx_indexes, sidx_indptr, sidx_val, loss_indptr, loss_val, pass_through, computes, compute_idx, output_array, i_layer, i_index)[source]¶
- class oasislmf.pytools.fm.stream_sparse.EventWriterSparse(files_out, nodes_array, output_array, sidx_indexes, sidx_indptr, sidx_val, loss_indptr, loss_val, pass_through, len_sample, computes)[source]¶
- class oasislmf.pytools.fm.stream_sparse.EventWriterOrderedOutputSparse(files_out, nodes_array, output_array, sidx_indexes, sidx_indptr, sidx_val, loss_indptr, loss_val, pass_through, len_sample, computes)[source]¶
Bases:
EventWriterSparse