Module record
Expand description
Module for the recorder.
Structs§
- BinBytes
Recorder - In memory recorder using the bincode format.
- BinFile
Recorder - File recorder using the bincode format.
- BinGz
File Recorder - File recorder using the bincode format compressed with gzip.
- Burn
Metadata - Metadata of a record.
- Burn
Record - Record that can be saved by a Recorder.
- Burn
Record NoItem - Record that can be saved by a Recorder without the item.
- Double
Precision Settings - Precision settings optimized for precision.
- Full
Precision Settings - Default precision settings.
- Half
Precision Settings - Precision settings optimized for compactness.
- Json
GzFile Recorder - File recorder using the json format compressed with gzip.
- Named
MpkBytes Recorder - In memory recorder using the Named MessagePack.
- Named
MpkFile Recorder - File recorder using the named msgpack format.
- Named
MpkGz File Recorder - File recorder using the named msgpack format compressed with gzip.
- Param
Serde - (De)serialize parameters into a clean format.
- Pretty
Json File Recorder - File recorder using pretty json format for easy readability.
Enums§
- Recorder
Error - Error that can occur when using a Recorder.
Traits§
- Bytes
Recorder - Recorder trait specialized to save and load data to and from bytes.
- File
Recorder - Recorder trait specialized to save and load data to and from files.
- Precision
Settings - Settings allowing to control the precision when (de)serializing items.
- Record
- Trait to define a family of types which can be recorded using any settings.
- Recorder
- Record any item implementing Serialize and DeserializeOwned.
Type Aliases§
- Compact
Recorder - Recorder optimized for compactness.
- Debug
Record Settings - Debug recorder.
- Default
File Recorder - Default file recorder.
- Default
Recorder - Default recorder.
- NoStd
Inference Recorder - Inference recorder compatible with no-std.
- NoStd
Training Recorder - Training recorder compatible with no-std inference.
- Sensitive
Compact Recorder - Recorder optimized for compactness making it a good choice for model deployment.
Derive Macros§
- Record
- Derive macro for the record.