o 0 i8*@sddlmZddlmZddlmZddlZddlZd$d d Zd%d&ddZ d'ddZ d(d)ddZ eGd d!d!Z Gd"d#d#Z dS)*) annotations) dataclass)LiteralNfilenamestr output_type str | NonereturncCs<|dur|nd}|dur|nd}|j|||||dS)N)ZouttypeZftypeZOUTTYPEZFTYPE)lowerupperformat)rrZftype_lowercaseZftype_uppercaserX/home/app/PaddleOCR-VL-test/.venv_paddleocr/lib/python3.10/site-packages/gguf/utility.pyfill_templated_filename s rmodel_params_countint min_digitscCs|dkr |d}d}n|dkr|d}d}n|dkr!|d}d }n|d }d }t|ttt|d d }|d|d|S)NgmBg-q=TgeAg& .>Bg.Agư>MgMbP?K0r.f)maxlenrroundlstrip)rrZscaled_model_paramsZ scale_suffixZfixrrr#model_weight_count_rounded_notations r total_params shared_params expert_params expert_countcCsF|dkrtt|t|dd}|d|}|Stt|dd}|S)Nrr)rx)r abs)r!r"r#r$Z pretty_sizeZ size_classrrr size_label*s r' model_name base_namefinetune_stringversion_string model_typeLiteral['vocab', 'LoRA'] | Nonec Cs|dur|dddd}n|dur"|dddd}nd}|dur-d|nd}|dur>d|ddnd} |durOd|ddnd} |durbd|ddnd} |dursd|ddnd} ||| | | | S)N -/z ggml-modelr )stripreplacer ) r(r)r*r+r'rr,name parametersZfinetuneversionencodingkindrrrnaming_convention5s""&"r8c@s@eZdZUded<ded<ded<ded<ded<dd d Zd S) RemoteTensorrdtypeztuple[int, ...]shaper offset_startsizeurlr bytearraycCsttj|j|j|jd}|S)N)r>startr=)r?SafetensorRemoteget_data_by_ranger>r<r=)selfdatarrrrDTszRemoteTensor.dataN)r r?)__name__ __module__ __qualname____annotations__rDrrrrr9Ls r9c@sneZdZdZdZdZeddd Zedd d Zed ddZ ed!d"ddZ ed#ddZ ed$ddZ dS)%rAa Uility class to handle remote safetensor files. This class is designed to work with Hugging Face model repositories. Example (one model has single safetensor file, the other has multiple): for model_id in ["ngxson/TEST-Tiny-Llama4", "Qwen/Qwen2.5-7B-Instruct"]: tensors = SafetensorRemote.get_list_tensors_hf_model(model_id) print(tensors) Example reading tensor data: tensors = SafetensorRemote.get_list_tensors_hf_model(model_id) for name, meta in tensors.items(): dtype, shape, offset_start, size, remote_safetensor_url = meta # read the tensor data data = SafetensorRemote.get_data_by_range(remote_safetensor_url, offset_start, size) print(data) zhttps://huggingface.comodel_idrr dict[str, RemoteTensor]cCs||jd|d}|r|jd|d}||S|jd|d}||}|r{||d}|d}t|}|ddusGJd|d} tt | } | i} | D]} |jd|d | }|| D]\} }|| | <qoq[| St d |d ) z Get list of tensors from a Hugging Face model repository. Returns a dictionary of tensor names and their metadata. Each tensor is represented as a tuple of (dtype, shape, offset_start, size, remote_safetensor_url) r0z/resolve/main/model.safetensorsz*/resolve/main/model.safetensors.index.jsonrutf-8 weight_mapNz"weight_map not found in index filez/resolve/main/zModel z# does not have any safetensor files)check_file_exist BASE_DOMAINget_list_tensorsrBdecodejsonloadsgetlistsetvaluessortitems ValueError)clsrJZis_single_filer> index_urlZis_multiple_filesZ index_dataZ index_strZ index_jsonrM all_filesZtensorsfilekeyvalrrrget_list_tensors_hf_modelqs,       z*SafetensorRemote.get_list_tensors_hf_modelr>c Cs||\}}i}|D]V\}}|dkrq t|ts%td|d|z%|d}|d}|d\} } | | } || } t|t|| | |d||<Wq tyc} z td|d| d |d } ~ ww|S) z Get list of tensors from a remote safetensor file. Returns a dictionary of tensor names and their metadata. Each tensor is represented as a tuple of (dtype, shape, offset_start, size) __metadata__zInvalid metadata for tensor 'z': r:r;Z data_offsets)r:r;r<r=r>z$Missing key in metadata for tensor 'z , meta = N) get_metadatarY isinstancedictrZr9tupleKeyError)r[r>metadatadata_start_offsetresr3metar:r;Zoffset_start_relativeZoffset_end_relativer=r<errrrPs&  z!SafetensorRemote.get_list_tensorstuple[dict, int]c Csd}||d|}t|dkrtdtj|dddd}d|}tj}||dkr3||||7}t|d|krItdd|d t||dd|}|d }z t |} | |fWStj yt} ztd | d} ~ ww) zz Get JSON metadata from a remote safetensor file. Returns tuple of (metadata, data_start_offset) iPrrIz%Not enough data to read metadata sizeNlittle) byteorderz'Could not read complete metadata. Need z bytes, got rLz-Failed to parse safetensor metadata as JSON: ) rBrrZr from_bytesrA ALIGNMENTrQrRrSJSONDecodeError) r[r> read_sizeraw_dataZmetadata_lengthriZ alignmentZmetadata_bytesZ metadata_strrhrlrrrrcs(     zSafetensorRemote.get_metadatar@rr=bytesc Csddl}ddlm}||}|jr|jstd||}|dkr/d|d|||d<|j|d |d }||j t |dkrF|SdS) z Get raw byte data from a remote file by range. If size is not specified, it will read the entire file. rNurlparse Invalid URL: ruzbytes=r/RangeTallow_redirectsheaders) requests urllib.parserxschemenetlocrZ_get_request_headersrTraise_for_statuscontentslice) r[r>r@r=r~rx parsed_urlr}responserrrrBs  z"SafetensorRemote.get_data_by_rangeboolcCsddl}ddlm}||}|jr|jstd|z|}d|d<|j|d|d}d |jko7d kWSWS|j yEYd Sw) zt Check if a file exists at the given URL. Returns True if the file exists, False otherwise. rNrwryz bytes=0-0rzTr{iF) r~rrxrrrZrhead status_codeRequestException)r[r>r~rxrr}rrrrrNs  z!SafetensorRemote.check_file_existdict[str, str]cCs,ddi}tjdrdtjd|d<|S)z$Prepare common headers for requests.z User-AgentZconvert_hf_to_ggufZHF_TOKENzBearer Authorization)osenvironrT)r[r}rrrrs z%SafetensorRemote._get_request_headersN)rJrr rK)r>rr rK)r>rr rm)ru)r>rr@rr=rr rv)r>rr r)r r) rErFrG__doc__rOrq classmethodrarPrcrBrNrrrrrrA[s  $  # rA)rrrrr r)r)rrrrr r) r!rr"rr#rr$rr r)N)r(rr)rr*rr+rr'rrrr,r-r r) __future__r dataclassesrtypingrrrRrr r'r8r9rArrrrs