search

YeonwooSung / MLOps

Miscellaneous codes and writings for MLOps
GNU General Public License v3.0
11 stars 1 forks source link

build(deps): bump the pip group across 11 directories with 1 update #90

Closed dependabot[bot] closed 7 months ago

dependabot[bot] commented 7 months ago

Bumps the pip group with 1 update in the /AWS/SageMaker/sagemaker-huggingface-toolkit/notebooks/accelerate_sagemaker/src/seq2seq directory: transformers. Bumps the pip group with 1 update in the /AWS/SageMaker/sagemaker-huggingface-toolkit/notebooks/inference/stable_diffusion_inference/code directory: transformers. Bumps the pip group with 1 update in the /LLM/RAG/transcription-and-semantic-search directory: transformers. Bumps the pip group with 1 update in the /LLM/large_laguage_models/mixtral_8x7b directory: transformers. Bumps the pip group with 1 update in the /LLM/llama_index/samples/mixtral_ollama directory: transformers. Bumps the pip group with 1 update in the /ml-serving/custom-serving/fastapi/local_streaming_llm directory: transformers. Bumps the pip group with 1 update in the /ml-serving/custom-serving/fastapi/ray/ray_distilbert directory: transformers. Bumps the pip group with 1 update in the /ml-serving/custom-serving/fastapi/ray/ray_stablediffusion directory: transformers. Bumps the pip group with 1 update in the /ml-serving/custom-serving/fastapi/ray/ray_yolov5s directory: transformers. Bumps the pip group with 1 update in the /ray/ray-air-with-gpt-j-6b directory: transformers. Bumps the pip group with 1 update in the /ray/zerocopy_loading directory: transformers.

Updates transformers from 4.36.0 to 4.38.0

Release notes

Sourced from transformers's releases.

v4.38: Gemma, Depth Anything, Stable LM; Static Cache, HF Quantizer, AQLM

New model additions

💎 Gemma 💎

Gemma is a new opensource Language Model series from Google AI that comes with a 2B and 7B variant. The release comes with the pre-trained and instruction fine-tuned versions and you can use them via AutoModelForCausalLM, GemmaForCausalLM or pipeline interface!

Read more about it in the Gemma release blogpost: https://hf.co/blog/gemma

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")
model = AutoModelForCausalLM.from_pretrained("google/gemma-2b", device_map="auto", torch_dtype=torch.float16)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

You can use the model with Flash Attention, SDPA, Static cache and quantization API for further optimizations !

  • Flash Attention 2
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", torch_dtype=torch.float16, attn_implementation="flash_attention_2"
)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

  • bitsandbytes-4bit
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", load_in_4bit=True
)
</tr></table>

... (truncated)

Commits
  • 08ab54a [ gemma] Adds support for Gemma 💎 (#29167)
  • 2de9314 [Maskformer] safely get backbone config (#29166)
  • 476957b 🚨 Llama: update rope scaling to match static cache changes (#29143)
  • 7a4bec6 Release: 4.38.0
  • ee3af60 Add support for fine-tuning CLIP-like models using contrastive-image-text exa...
  • 0996a10 Revert low cpu mem tie weights (#29135)
  • 15cfe38 [Core tokenization] add_dummy_prefix_space option to help with latest is...
  • efdd436 FIX [PEFT / Trainer ] Handle better peft + quantized compiled models (#29...
  • 5e95dca [cuda kernels] only compile them when initializing (#29133)
  • a7755d2 Generate: unset GenerationConfig parameters do not raise warning (#29119)
  • Additional commits viewable in compare view


Updates transformers from 4.36.0 to 4.38.0

Release notes

Sourced from transformers's releases.

v4.38: Gemma, Depth Anything, Stable LM; Static Cache, HF Quantizer, AQLM

New model additions

💎 Gemma 💎

Gemma is a new opensource Language Model series from Google AI that comes with a 2B and 7B variant. The release comes with the pre-trained and instruction fine-tuned versions and you can use them via AutoModelForCausalLM, GemmaForCausalLM or pipeline interface!

Read more about it in the Gemma release blogpost: https://hf.co/blog/gemma

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")
model = AutoModelForCausalLM.from_pretrained("google/gemma-2b", device_map="auto", torch_dtype=torch.float16)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

You can use the model with Flash Attention, SDPA, Static cache and quantization API for further optimizations !

  • Flash Attention 2
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", torch_dtype=torch.float16, attn_implementation="flash_attention_2"
)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

  • bitsandbytes-4bit
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", load_in_4bit=True
)
</tr></table>

... (truncated)

Commits
  • 08ab54a [ gemma] Adds support for Gemma 💎 (#29167)
  • 2de9314 [Maskformer] safely get backbone config (#29166)
  • 476957b 🚨 Llama: update rope scaling to match static cache changes (#29143)
  • 7a4bec6 Release: 4.38.0
  • ee3af60 Add support for fine-tuning CLIP-like models using contrastive-image-text exa...
  • 0996a10 Revert low cpu mem tie weights (#29135)
  • 15cfe38 [Core tokenization] add_dummy_prefix_space option to help with latest is...
  • efdd436 FIX [PEFT / Trainer ] Handle better peft + quantized compiled models (#29...
  • 5e95dca [cuda kernels] only compile them when initializing (#29133)
  • a7755d2 Generate: unset GenerationConfig parameters do not raise warning (#29119)
  • Additional commits viewable in compare view


Updates transformers from 4.30.0 to 4.38.0

Release notes

Sourced from transformers's releases.

v4.38: Gemma, Depth Anything, Stable LM; Static Cache, HF Quantizer, AQLM

New model additions

💎 Gemma 💎

Gemma is a new opensource Language Model series from Google AI that comes with a 2B and 7B variant. The release comes with the pre-trained and instruction fine-tuned versions and you can use them via AutoModelForCausalLM, GemmaForCausalLM or pipeline interface!

Read more about it in the Gemma release blogpost: https://hf.co/blog/gemma

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")
model = AutoModelForCausalLM.from_pretrained("google/gemma-2b", device_map="auto", torch_dtype=torch.float16)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

You can use the model with Flash Attention, SDPA, Static cache and quantization API for further optimizations !

  • Flash Attention 2
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", torch_dtype=torch.float16, attn_implementation="flash_attention_2"
)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

  • bitsandbytes-4bit
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", load_in_4bit=True
)
</tr></table>

... (truncated)

Commits
  • 08ab54a [ gemma] Adds support for Gemma 💎 (#29167)
  • 2de9314 [Maskformer] safely get backbone config (#29166)
  • 476957b 🚨 Llama: update rope scaling to match static cache changes (#29143)
  • 7a4bec6 Release: 4.38.0
  • ee3af60 Add support for fine-tuning CLIP-like models using contrastive-image-text exa...
  • 0996a10 Revert low cpu mem tie weights (#29135)
  • 15cfe38 [Core tokenization] add_dummy_prefix_space option to help with latest is...
  • efdd436 FIX [PEFT / Trainer ] Handle better peft + quantized compiled models (#29...
  • 5e95dca [cuda kernels] only compile them when initializing (#29133)
  • a7755d2 Generate: unset GenerationConfig parameters do not raise warning (#29119)
  • Additional commits viewable in compare view


Updates transformers from 4.30.0 to 4.38.0

Release notes

Sourced from transformers's releases.

v4.38: Gemma, Depth Anything, Stable LM; Static Cache, HF Quantizer, AQLM

New model additions

💎 Gemma 💎

Gemma is a new opensource Language Model series from Google AI that comes with a 2B and 7B variant. The release comes with the pre-trained and instruction fine-tuned versions and you can use them via AutoModelForCausalLM, GemmaForCausalLM or pipeline interface!

Read more about it in the Gemma release blogpost: https://hf.co/blog/gemma

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")
model = AutoModelForCausalLM.from_pretrained("google/gemma-2b", device_map="auto", torch_dtype=torch.float16)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

You can use the model with Flash Attention, SDPA, Static cache and quantization API for further optimizations !

  • Flash Attention 2
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", torch_dtype=torch.float16, attn_implementation="flash_attention_2"
)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

  • bitsandbytes-4bit
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", load_in_4bit=True
)
</tr></table>

... (truncated)

Commits
  • 08ab54a [ gemma] Adds support for Gemma 💎 (#29167)
  • 2de9314 [Maskformer] safely get backbone config (#29166)
  • 476957b 🚨 Llama: update rope scaling to match static cache changes (#29143)
  • 7a4bec6 Release: 4.38.0
  • ee3af60 Add support for fine-tuning CLIP-like models using contrastive-image-text exa...
  • 0996a10 Revert low cpu mem tie weights (#29135)
  • 15cfe38 [Core tokenization] add_dummy_prefix_space option to help with latest is...
  • efdd436 FIX [PEFT / Trainer ] Handle better peft + quantized compiled models (#29...
  • 5e95dca [cuda kernels] only compile them when initializing (#29133)
  • a7755d2 Generate: unset GenerationConfig parameters do not raise warning (#29119)
  • Additional commits viewable in compare view


Updates transformers from 4.37.2 to 4.38.0

Release notes

Sourced from transformers's releases.

v4.38: Gemma, Depth Anything, Stable LM; Static Cache, HF Quantizer, AQLM

New model additions

💎 Gemma 💎

Gemma is a new opensource Language Model series from Google AI that comes with a 2B and 7B variant. The release comes with the pre-trained and instruction fine-tuned versions and you can use them via AutoModelForCausalLM, GemmaForCausalLM or pipeline interface!

Read more about it in the Gemma release blogpost: https://hf.co/blog/gemma

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")
model = AutoModelForCausalLM.from_pretrained("google/gemma-2b", device_map="auto", torch_dtype=torch.float16)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

You can use the model with Flash Attention, SDPA, Static cache and quantization API for further optimizations !

  • Flash Attention 2
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", torch_dtype=torch.float16, attn_implementation="flash_attention_2"
)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

  • bitsandbytes-4bit
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", load_in_4bit=True
)
</tr></table>

... (truncated)

Commits
  • 08ab54a [ gemma] Adds support for Gemma 💎 (#29167)
  • 2de9314 [Maskformer] safely get backbone config (#29166)
  • 476957b 🚨 Llama: update rope scaling to match static cache changes (#29143)
  • 7a4bec6 Release: 4.38.0
  • ee3af60 Add support for fine-tuning CLIP-like models using contrastive-image-text exa...
  • 0996a10 Revert low cpu mem tie weights (#29135)
  • 15cfe38 [Core tokenization] add_dummy_prefix_space option to help with latest is...
  • efdd436 FIX [PEFT / Trainer ] Handle better peft + quantized compiled models (#29...
  • 5e95dca [cuda kernels] only compile them when initializing (#29133)
  • a7755d2 Generate: unset GenerationConfig parameters do not raise warning (#29119)
  • Additional commits viewable in compare view


Updates transformers from 4.36.2 to 4.38.0

Release notes

Sourced from transformers's releases.

v4.38: Gemma, Depth Anything, Stable LM; Static Cache, HF Quantizer, AQLM

New model additions

💎 Gemma 💎

Gemma is a new opensource Language Model series from Google AI that comes with a 2B and 7B variant. The release comes with the pre-trained and instruction fine-tuned versions and you can use them via AutoModelForCausalLM, GemmaForCausalLM or pipeline interface!

Read more about it in the Gemma release blogpost: https://hf.co/blog/gemma

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")
model = AutoModelForCausalLM.from_pretrained("google/gemma-2b", device_map="auto", torch_dtype=torch.float16)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

You can use the model with Flash Attention, SDPA, Static cache and quantization API for further optimizations !

  • Flash Attention 2
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", torch_dtype=torch.float16, attn_implementation="flash_attention_2"
)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

  • bitsandbytes-4bit
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", load_in_4bit=True
)
</tr></table>

... (truncated)

Commits
  • 08ab54a [ gemma] Adds support for Gemma 💎 (#29167)
  • 2de9314 [Maskformer] safely get backbone config (#29166)
  • 476957b 🚨 Llama: update rope scaling to match static cache changes (#29143)
  • 7a4bec6 Release: 4.38.0
  • ee3af60 Add support for fine-tuning CLIP-like models using contrastive-image-text exa...
  • 0996a10 Revert low cpu mem tie weights (#29135)
  • 15cfe38 [Core tokenization] add_dummy_prefix_space option to help with latest is...
  • efdd436 FIX [PEFT / Trainer ] Handle better peft + quantized compiled models (#29...
  • 5e95dca [cuda kernels] only compile them when initializing (#29133)
  • a7755d2 Generate: unset GenerationConfig parameters do not raise warning (#29119)
  • Additional commits viewable in compare view


Updates transformers from 4.37.2 to 4.38.0

Release notes

Sourced from transformers's releases.

v4.38: Gemma, Depth Anything, Stable LM; Static Cache, HF Quantizer, AQLM

New model additions

💎 Gemma 💎

Gemma is a new opensource Language Model series from Google AI that comes with a 2B and 7B variant. The release comes with the pre-trained and instruction fine-tuned versions and you can use them via AutoModelForCausalLM, GemmaForCausalLM or pipeline interface!

Read more about it in the Gemma release blogpost: https://hf.co/blog/gemma

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")
model = AutoModelForCausalLM.from_pretrained("google/gemma-2b", device_map="auto", torch_dtype=torch.float16)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

You can use the model with Flash Attention, SDPA, Static cache and quantization API for further optimizations !

  • Flash Attention 2
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", torch_dtype=torch.float16, attn_implementation="flash_attention_2"
)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

  • bitsandbytes-4bit
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", load_in_4bit=True
)
</tr></table>

... (truncated)

Commits
  • 08ab54a [ gemma] Adds support for Gemma 💎 (#29167)
  • 2de9314 [Maskformer] safely get backbone config (#29166)
  • 476957b 🚨 Llama: update rope scaling to match static cache changes (#29143)
  • 7a4bec6 Release: 4.38.0
  • ee3af60 Add support for fine-tuning CLIP-like models using contrastive-image-text exa...
  • 0996a10 Revert low cpu mem tie weights (#29135)
  • 15cfe38 [Core tokenization] add_dummy_prefix_space option to help with latest is...
  • efdd436 FIX [PEFT / Trainer ] Handle better peft + quantized compiled models (#29...
  • 5e95dca [cuda kernels] only compile them when initializing (#29133)
  • a7755d2 Generate: unset GenerationConfig parameters do not raise warning (#29119)
  • Additional commits viewable in compare view


Updates transformers from 4.37.2 to 4.38.0

Release notes

Sourced from transformers's releases.

v4.38: Gemma, Depth Anything, Stable LM; Static Cache, HF Quantizer, AQLM

New model additions

💎 Gemma 💎

Gemma is a new opensource Language Model series from Google AI that comes with a 2B and 7B variant. The release comes with the pre-trained and instruction fine-tuned versions and you can use them via AutoModelForCausalLM, GemmaForCausalLM or pipeline interface!

Read more about it in the Gemma release blogpost: https://hf.co/blog/gemma

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")
model = AutoModelForCausalLM.from_pretrained("google/gemma-2b", device_map="auto", torch_dtype=torch.float16)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

You can use the model with Flash Attention, SDPA, Static cache and quantization API for further optimizations !

  • Flash Attention 2
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", torch_dtype=torch.float16, attn_implementation="flash_attention_2"
)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

  • bitsandbytes-4bit
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", load_in_4bit=True
)
</tr></table>

... (truncated)

Commits
  • 08ab54a [ gemma] Adds support for Gemma 💎 (#29167)
  • 2de9314 [Maskformer] safely get backbone config (#29166)
  • 476957b 🚨 Llama: update rope scaling to match static cache changes (#29143)
  • 7a4bec6 Release: 4.38.0
  • ee3af60 Add support for fine-tuning CLIP-like models using contrastive-image-text exa...
  • 0996a10 Revert low cpu mem tie weights (#29135)
  • 15cfe38 [Core tokenization] add_dummy_prefix_space option to help with latest is...
  • efdd436 FIX [PEFT / Trainer ] Handle better peft + quantized compiled models (#29...
  • 5e95dca [cuda kernels] only compile them when initializing (#29133)
  • a7755d2 Generate: unset GenerationConfig parameters do not raise warning (#29119)
  • Additional commits viewable in compare view


Updates transformers from 4.37.2 to 4.38.0

Release notes

Sourced from transformers's releases.

v4.38: Gemma, Depth Anything, Stable LM; Static Cache, HF Quantizer, AQLM

New model additions

💎 Gemma 💎

Gemma is a new opensource Language Model series from Google AI that comes with a 2B and 7B variant. The release comes with the pre-trained and instruction fine-tuned versions and you can use them via AutoModelForCausalLM, GemmaForCausalLM or pipeline interface!

Read more about it in the Gemma release blogpost: https://hf.co/blog/gemma

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")
model = AutoModelForCausalLM.from_pretrained("google/gemma-2b", device_map="auto", torch_dtype=torch.float16)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

You can use the model with Flash Attention, SDPA, Static cache and quantization API for further optimizations !

  • Flash Attention 2
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", torch_dtype=torch.float16, attn_implementation="flash_attention_2"
)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

  • bitsandbytes-4bit
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", load_in_4bit=True
)
</tr></table>

... (truncated)

Commits
  • 08ab54a [ gemma] Adds support for Gemma 💎 (#29167)
  • 2de9314 [Maskformer] safely get backbone config (#29166)
  • 476957b 🚨 Llama: update rope scaling to match static cache changes (#29143)
  • 7a4bec6 Release: 4.38.0
  • ee3af60 Add support for fine-tuning CLIP-like models using contrastive-image-text exa...
  • 0996a10 Revert low cpu mem tie weights (#29135)
  • 15cfe38 [Core tokenization] add_dummy_prefix_space option to help with latest is...
  • efdd436 FIX [PEFT / Trainer ] Handle better peft + quantized compiled models (#29...
  • 5e95dca [cuda kernels] only compile them when initializing (#29133)
  • a7755d2 Generate: unset GenerationConfig parameters do not raise warning (#29119)
  • Additional commits viewable in compare view


Updates transformers from 4.36.0 to 4.38.0

Release notes

Sourced from transformers's releases.

v4.38: Gemma, Depth Anything, Stable LM; Static Cache, HF Quantizer, AQLM

New model additions

💎 Gemma 💎

Gemma is a new opensource Language Model series from Google AI that comes with a 2B and 7B variant. The release comes with the pre-trained and instruction fine-tuned versions and you can use them via AutoModelForCausalLM, GemmaForCausalLM or pipeline interface!

Read more about it in the Gemma release blogpost: https://hf.co/blog/gemma

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")
model = AutoModelForCausalLM.from_pretrained("google/gemma-2b", device_map="auto", torch_dtype=torch.float16)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

You can use the model with Flash Attention, SDPA, Static cache and quantization API for further optimizations !

  • Flash Attention 2
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", torch_dtype=torch.float16, attn_implementation="flash_attention_2"
)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

  • bitsandbytes-4bit
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", load_in_4bit=True
)
</tr></table>

... (truncated)

Commits
  • 08ab54a [ gemma] Adds support for Gemma 💎 (#29167)
  • 2de9314 [Maskformer] safely get backbone config (#29166)
  • 476957b 🚨 Llama: update rope scaling to match static cache changes (#29143)
  • 7a4bec6 Release: 4.38.0
  • ee3af60 Add support for fine-tuning CLIP-like models using contrastive-image-text exa...
  • 0996a10 Revert low cpu mem tie weights (#29135)
  • 15cfe38 [Core tokenization] add_dummy_prefix_space option to help with latest is...
  • efdd436 FIX [PEFT / Trainer ] Handle better peft + quantized compiled models (#29...
  • 5e95dca [cuda kernels] only compile them when initializing (#29133)
  • a7755d2 Generate: unset GenerationConfig parameters do not raise warning (#29119)
  • Additional commits viewable in compare view


Updates transformers from 4.36.0 to 4.38.0

Release notes

Sourced from transformers's releases.

v4.38: Gemma, Depth Anything, Stable LM; Static Cache, HF Quantizer, AQLM

New model additions

💎 Gemma 💎

Gemma is a new opensource Language Model series from Google AI that comes with a 2B and 7B variant. The release comes with the pre-trained and instruction fine-tuned versions and you can use them via AutoModelForCausalLM, GemmaForCausalLM or pipeline interface!

Read more about it in the Gemma release blogpost: https://hf.co/blog/gemma

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")
model = AutoModelForCausalLM.from_pretrained("google/gemma-2b", device_map="auto", torch_dtype=torch.float16)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

You can use the model with Flash Attention, SDPA, Static cache and quantization API for further optimizations !

  • Flash Attention 2
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", torch_dtype=torch.float16, attn_implementation="flash_attention_2"
)


input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")


outputs = model.generate(**input_ids)

  • bitsandbytes-4bit
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")


model = AutoModelForCausalLM.from_pretrained(
"google/gemma-2b", device_map="auto", load_in_4bit=True
)
</tr></table>

... (truncated)

Commits
  • 08ab54a [ gemma] Adds support for Gemma 💎 (#29167)
  • 2de9314 [Maskformer] safely get backbone config (#29166)
  • 476957b 🚨 Llama: update rope scaling to match static cache changes (#29143)
  • 7a4bec6 Release: 4.38.0
  • ee3af60 Add support for fine-tuning CLIP-like models using contrastive-image-text exa...
  • 0996a10 Revert low cpu mem tie weights (#29135)
  • 15cfe38 [Core tokenization] add_dummy_prefix_space option to help with latest is...
  • efdd436 FIX [PEFT / Trainer ] Handle better peft + quantized compiled models (#29...
  • 5e95dca [cuda kernels] only compile them when initializing (#29133)
  • a7755d2 Generate: unset GenerationConfig parameters do not raise warning (#29119)
  • Additional commits viewable in compare view


Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

Dependabot commands and options
You can trigger Dependabot actions by commenting on this PR: - `@dependabot rebase` will rebase this PR - `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it - `@dependabot merge` will merge this PR after your CI passes on it - `@dependabot squash and merge` will squash and merge this PR after your CI passes on it - `@dependabot cancel merge` will cancel a previously requested merge and block automerging - `@dependabot reopen` will reopen this PR if it is closed - `@dependabot close` will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually - `@dependabot show ignore conditions` will show all of the ignore conditions of the specified dependency - `@dependabot ignore major version` will close this group update PR and stop Dependabot creating any more for the specific dependency's major version (unless you unignore this specific dependency's major version or upgrade to it yourself) - `@dependabot ignore minor version` will close this group update PR and stop Dependabot creating any more for the specific dependency's minor version (unless you unignore this specific dependency's minor version or upgrade to it yourself) - `@dependabot ignore ` will close this group update PR and stop Dependabot creating any more for the specific dependency (unless you unignore this specific dependency or upgrade to it yourself) - `@dependabot unignore ` will remove all of the ignore conditions of the specified dependency - `@dependabot unignore ` will remove the ignore condition of the specified dependency and ignore conditions You can disable automated security fix PRs for this repo from the [Security Alerts page](https://github.com/YeonwooSung/MLOps/network/alerts).