Hacktoberfest 2026: le issue che i maintainer hanno segnato per ottobre, aperte e adatte ai principianti. Sfoglia le issue Hacktoberfest

cannot pickle 'Stream' object

Aperta
#903 0 commenti 0 reazioni 0 assegnatari Vedi su GitHub

Nessuno ha ancora preso questa issue.

Valutazione

Difficoltà
3/5
Tempo stimato
1-2 giorni
Idoneità per principianti
25/100
Tipo di issue
Bug
Chiarezza
Da chiarire
Stato di attività
Ferma
Stack tecnologico
python, pytorch

Direzione di ricerca

Riproduci il fallimento di DeepSpeedExamples/inference/huggingface/zero_inference/deep.py usando il comando mostrato deepspeed deep.py, quindi traccia il percorso di AutoModelForCausalLM.from_pretrained attraverso la chiamata deepcopy segnalata di Transformers e bitsandbytes. Non è indicato alcun test; il lavoro è completato quando è stato stabilito un percorso di esecuzione supportato per questo esempio che non solleva più l’errore di pickling Stream segnalato.

Scritto dal modello di indicizzazione a partire dal testo della issue.

Descrizione

Trying to run llama3 70B

Hardware:
3x 4090

deepspeed deep.py

deep.py:

# ---------------------------------------
# New automatic tensor parallelism method
# ---------------------------------------
import os
import torch
import transformers
from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig
import deepspeed
local_rank = int(os.getenv("LOCAL_RANK", "0"))
world_size = int(os.getenv("WORLD_SIZE", "1"))

# specify the model id
model_id = "meta-llama/Meta-Llama-3-70B-Instruct"
# create the model pipeline
bnb_config = BitsAndBytesConfig(load_in_4bit=True, bnb_4bit_quant_type="nf4", bnb_4bit_compute_dtype=torch.float16, bnb_4bit_use_double_quant=True, llm_int8_enable_fp32_cpu_offload=False)
model = AutoModelForCausalLM.from_pretrained(model_id, quantization_config=bnb_config, torch_dtype="auto",device_map="auto", low_cpu_mem_usage = True)


pipe = transformers.pipeline(task="text-generation", model=model, device=local_rank)
# Initialize the DeepSpeed-Inference engine
pipe.model = deepspeed.init_inference(
    pipe.model,
    mp_size=world_size,
    dtype=torch.float
)
output = pipe('What is deepspeed')
print(output)

Traceback:

Traceback (most recent call last):
  File "/home/sw/bulldozer/code/deepspeed/DeepSpeedExamples/inference/huggingface/zero_inference/deep.py", line 17, in <module>
    model = AutoModelForCausalLM.from_pretrained(model_id, quantization_config=bnb_config, torch_dtype="auto",device_map="auto", low_cpu_mem_usage = True)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/site-packages/transformers/models/auto/auto_factory.py", line 555, in from_pretrained
    return model_class.from_pretrained(
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/site-packages/transformers/modeling_utils.py", line 2974, in from_pretrained
    modules_to_not_convert = get_keys_to_not_convert(model)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/site-packages/transformers/utils/bitsandbytes.py", line 257, in get_keys_to_not_convert
    tied_model = deepcopy(model)  # this has 0 cost since it is done inside `init_empty_weights` context manager`
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 172, in deepcopy
    y = _reconstruct(x, memo, *rv)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 271, in _reconstruct
    state = deepcopy(state, memo)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 146, in deepcopy
    y = copier(x, memo)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 231, in _deepcopy_dict
    y[deepcopy(key, memo)] = deepcopy(value, memo)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 172, in deepcopy
    y = _reconstruct(x, memo, *rv)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 297, in _reconstruct
    value = deepcopy(value, memo)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 172, in deepcopy
    y = _reconstruct(x, memo, *rv)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 271, in _reconstruct
    state = deepcopy(state, memo)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 146, in deepcopy
    y = copier(x, memo)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 231, in _deepcopy_dict
    y[deepcopy(key, memo)] = deepcopy(value, memo)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 172, in deepcopy
    y = _reconstruct(x, memo, *rv)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 297, in _reconstruct
    value = deepcopy(value, memo)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 172, in deepcopy
    y = _reconstruct(x, memo, *rv)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 271, in _reconstruct
    state = deepcopy(state, memo)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 146, in deepcopy
    y = copier(x, memo)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 231, in _deepcopy_dict
    y[deepcopy(key, memo)] = deepcopy(value, memo)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 172, in deepcopy
    y = _reconstruct(x, memo, *rv)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 297, in _reconstruct
    value = deepcopy(value, memo)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 172, in deepcopy
    y = _reconstruct(x, memo, *rv)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 271, in _reconstruct
    state = deepcopy(state, memo)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 146, in deepcopy
    y = copier(x, memo)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 231, in _deepcopy_dict
    y[deepcopy(key, memo)] = deepcopy(value, memo)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 172, in deepcopy
    y = _reconstruct(x, memo, *rv)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 297, in _reconstruct
    value = deepcopy(value, memo)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 172, in deepcopy
    y = _reconstruct(x, memo, *rv)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 271, in _reconstruct
    state = deepcopy(state, memo)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 146, in deepcopy
    y = copier(x, memo)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 231, in _deepcopy_dict
    y[deepcopy(key, memo)] = deepcopy(value, memo)
  File "/home/sw/anaconda3/envs/deepseed/lib/python3.10/copy.py", line 161, in deepcopy
    rv = reductor(4)
TypeError: cannot pickle 'Stream' object
Lingua principale
Python
Stelle
6.8k
Fork
1.1k
Merge medio
2g 16h
PR unite (30g)
1

Preparare l'ambiente

Non abbiamo ancora controllato i file di configurazione di questo progetto. Parti dal suo README e consulta la nostra guida al primo contributo per i passaggi generali.

Come iniziare

  1. Leggi tutta la issue e poi la guida ai contributi del progetto.
  2. Commenta sulla issue per dire che te ne occupi tu — evita che due persone facciano lo stesso lavoro.
  3. Fai un fork del repository e lavora su un branch.
  4. Apri una pull request che faccia riferimento al numero della issue.

Altre issue di deepspeedai/DeepSpeedExamples

Tutte le issue di deepspeedai/DeepSpeedExamples

Issue simili

Altre issue su Python

Ricevi le nuove issue nella tua casella

Un breve riepilogo di issue GitHub adatte ai principianti.