Downloads · 30 days
6
1% of all-time downloads
ed001_-_datagemma-2b-gguf
RichardErkhov/ed001_-_datagemma-2b-gguf
ed001_-_datagemma-2b-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
All-time downloads
434
Public
Repo size
26.9 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.gguf31.7 GB · 100%
At a glance
- Access
- Public
- Created
- Oct 29, 2024
- Updated
- Oct 29, 2024
- SHA
- b5a37eae
Model card
From the Hugging Face model README
Quantization made by Richard Erkhov.
datagemma-2b - GGUF
- Model creator: https://huggingface.co/ed001/
- Original model: https://huggingface.co/ed001/datagemma-2b/
| Name | Quant method | Size |
|---|---|---|
| datagemma-2b.Q2_K.gguf | Q2_K | 1.08GB |
| datagemma-2b.Q3_K_S.gguf | Q3_K_S | 1.2GB |
| datagemma-2b.Q3_K.gguf | Q3_K | 1.29GB |
| datagemma-2b.Q3_K_M.gguf | Q3_K_M | 1.29GB |
| datagemma-2b.Q3_K_L.gguf | Q3_K_L | 1.36GB |
| datagemma-2b.IQ4_XS.gguf | IQ4_XS | 1.4GB |
| datagemma-2b.Q4_0.gguf | Q4_0 | 1.44GB |
| datagemma-2b.IQ4_NL.gguf | IQ4_NL | 1.45GB |
| datagemma-2b.Q4_K_S.gguf | Q4_K_S | 1.45GB |
| datagemma-2b.Q4_K.gguf | Q4_K | 1.52GB |
| datagemma-2b.Q4_K_M.gguf | Q4_K_M | 1.52GB |
| datagemma-2b.Q4_1.gguf | Q4_1 | 1.56GB |
| datagemma-2b.Q5_0.gguf | Q5_0 | 1.68GB |
| datagemma-2b.Q5_K_S.gguf | Q5_K_S | 1.68GB |
| datagemma-2b.Q5_K.gguf | Q5_K | 1.71GB |
| datagemma-2b.Q5_K_M.gguf | Q5_K_M | 1.71GB |
| datagemma-2b.Q5_1.gguf | Q5_1 | 1.79GB |
| datagemma-2b.Q6_K.gguf | Q6_K | 1.92GB |
| datagemma-2b.Q8_0.gguf | Q8_0 | 2.49GB |
Original model description:
language:
- en license: mit tags:
- code
- data science datasets:
- ed001/ds-coder-instruct-v2 pipeline_tag: text-generation
datagemma-2b
The datagemma-2b is a model designated for data science code generation from natural language instruction. It is fine-tuned from codegemma-2b model. Fine tuning was performed on the ed001/ds-coder-instruct-v2 dataset which is constructed by filtering publicly available datasets on HuggingFace.
Usage
from transformers import AutoTokenizer, AutoModelForCausalLM, GenerationConfig
model = AutoModelForCausalLM.from_pretrained(
"ed001/datagemma-2b",
low_cpu_mem_usage=True
).cuda()
# Reload tokenizer to save it
tokenizer = AutoTokenizer.from_pretrained("ed001/datagemma-2b", trust_remote_code=True)
tokenizer.padding_side = "right"
prompt_template = "### Question: {}\n ### Answer: "
generation_config = GenerationConfig(max_new_tokens=512, top_p=0.5, do_sample=True, repetition_penalty=1)
prompt = "How can I profile speed of my neural network using PyTorch?"
input = tokenizer(prompt_template.format(prompt), return_tensors="pt").to(model.device)["input_ids"]
print(tokenizer.decode(model.generate(input, generation_config=generation_config)[0]))
Training Details
lora_r: 32
lora_alpha: 16
lora_dropout: 0.05
target_modules: q, k, v, o, gate_proj, down_proj, up_proj
weight_decay: 0
optmizer: paged_adamw_8bit
lr: 1e-4
lr_scheduler: cosine
max_seq_len: 1536
batch_size: 1
grad_acc: 4
max_grad_norm: 0.5
warmup_ratio: 0.05
num_epochs: 1
Contact
GitHub: Ea0011
- Created
- Oct 29, 2024
- Updated
- Oct 29, 2024