erfanzar / EasyDeL

Accelerate, Optimize performance with streamlined training and serving options with JAX.

https://easydel.readthedocs.io/en/latest/

erfanzar/EasyDeL Issues

Issue saving and converting the Gemma 2 model after training
Closed 23 days ago2
TPU v4-32 set-up not working
Updated 24 days ago13
Import error EasyDeL libraries examples/flash_attention_training_example.py
Closed 2 months ago6
oom when llama2-7b sft
Closed 2 months ago5
EasyDeL
Closed 2 months ago1
TPU-v3 Kaggle not working after update
Closed 3 months ago5
NaN loss in ORPOTrainer with legacy_sharded_vanilla
Closed 3 months ago9
value error using flash attention
Closed 3 months ago1
Falcon-11B: Dict key mismatch; expected keys: ['input_layernorm', 'mlp', 'self_attention']; dict: {'self_attention': {'query_key_value': {'kernel': Array
Closed 3 months ago9
Logging into wandb.ai
Closed 3 months ago2
Out of Memory issue in new easydel version.
Closed 3 months ago6
[Feature Request] Add support for tiiuae/falcon-11B
Closed 3 months ago4
checkpoint's size is increasing everytime.
Closed 3 months ago3
Import Error
Closed 3 months ago1
Mosaic kernels cannot be automatically partitioned. Please wrap the call in a shard_map or xmap
Closed 4 months ago3
AssertionError: Precision DEFAULT requested together with quantization.
Closed 4 months ago5
Kaggle training examples don't work
Closed 4 months ago14
Out of memory for serving example
Closed 4 months ago3
Can't load checkpoints continue training
Closed 4 months ago7
training does not start using latest easydel
Closed 4 months ago6
load_in_8bit doesn't work on Kaggle TPU
Closed 4 months ago2
'LoraWeight' object has no attribute 'tolist'
Closed 4 months ago4
Please provide support for LLama3 or provide example on how to serve it using Easydel
Closed 4 months ago4
How to reduce TPU RAM when finetuning?
Closed 5 months ago8
Attention Mask for Packed Sequences (via Attention Bias)
Closed 5 months ago3
Transformers-like API for inference
Closed 5 months ago19
Unable to Load EasyDeL State
Closed 5 months ago6
Error converting easydel checkpoint to huggingface model.
Closed 6 months ago2
a question about how to increase batch size.
Closed 6 months ago6
How to continue training from a previous saved easydel checkpoint?
Closed 6 months ago9
Training with Ring Attention Failed
Closed 6 months ago10
Docs site is broken https://erfanzar.github.io
Closed 6 months ago1
Install from git not working
Closed 6 months ago8
Training in kaggle's TPU is failing
Closed 6 months ago5
Error while finetuning Tinyllama on Kaggle TPU
Closed 6 months ago5
Output Differs from Hugging Face Transformer Result and EasyDel Results
Closed 6 months ago4
[Urgent] Exception while load AdaptLLM/medicine-chat, variant of llama
Closed 6 months ago6
Support HuggingFaceH4/zephyr-7b-beta serving using EasyDel
Closed 6 months ago1
Easydel support on TPU v4.8 - getting exception
Closed 6 months ago1
Example shown on https://pypi.org/project/EasyDeL/ to finetune tinyllama raise exception on kaggle
Closed 6 months ago3
GPT2 (150M model) support on Tv2.8. Example scripts goes out of memory
Closed 6 months ago1
Exception while running any model - einops.EinopsError: Error while processing rearrange-reduction pattern "b (c n) d -> b c n d".
Closed 6 months ago1
None of the examples scripts works, that used to work earlier. Please test your examples again and update docs
Closed 6 months ago1
QLoRA Finetune Example
Closed 7 months ago11
Error while training GPT2 on the kaggle
Closed 7 months ago3
Potential regression causing resource exhausted after recent commit
Closed 7 months ago3
Error running remote model that has custom code
Closed 7 months ago2
ValueError: `params` cannot be accessed from model when the model is created with `_do_init=False`.
Closed 7 months ago2
Error while training a Phi2 model
Closed 7 months ago1
What is the hardware spec are you using to tran a LLAMA model with 7B params
Closed 7 months ago1