Benign python Download

metatensor-torch-cxx-0.10.5/src/array.cpp

Source identifier named flags
SHA-256e4478569605175808126fbaca643de9e5a27da88687347f66883e830bf2504c4
MaleculeH(Db)

Evidence

English function-word token "and" lines 166–174
166:9… rn dt;
167}
168
169
170// We need to register a data origin with metatensor, that will be used for all
171// mts_array_t containing a C++ torch tensor. This is initialized to 0 (meaning
172// "no origin"), and will be set by `MetatensorOriginRegistration` in
173// `TorchDataArray::origin`.
174mts_data_origin_t …
English function-word token "the" lines 214–218
214:78… ersion);
215
216 // Validate fill_value shape from the DLPack tensor directly
217 if (fill_value_dlpack->dl_tensor.ndim != 0) {
218 if (fill_value …
Source identifier named flags lines 408–414
408:17… nsor->deleter = dlpack_versioned_deleter;
409 versioned_tensor->flags = 0;
410 // Copy metadata
411 versioned_tensor->dl_tensor = legacy_tensor->dl_tensor;
412 return versioned_tensor;
413#else
414 th …
English function-word token "have" lines 422–424
422:4… This is used for PyTorch <=2.8, which does not have at::toDLPackVersioned()
423struct ManagedTensorCtx {
424 DLManagedTensorVersioned* versioned_tenso …

Keyboard shortcuts on this page: j for the next sample, k for the previous one, x to go back to the feed, d to download the original bytes, r to re-queue the sample for analysis.