Hugging Face's logo

AmareshHebbar
/
leetcode-python-qwen25-coder-7b

Text Generation
PEFT
Safetensors
GGUF
English
code
leetcode
python
code-generation
competitive-programming
qwen2.5-coder
dora
qdora
weight-decomposed-lora
instruction-tuned
sft
algorithm-generation
function-generation
coding-assistant
on-device
ollama
vllm
text-generation-inference
doocs-leetcode
synthetic-verification
quantized
algorithms
conversational
Model card Files
xet
 Community
AmareshHebbar commited on
Commit
2ce7cae
·
verified ·
1 Parent(s): 4dacce0

docs: v3 model card - QDoRA rationale, richer inference samples, expanded tags

Browse files
Files changed (1) hide show
  1. README.md +5 -5
README.md CHANGED
@@ -30,7 +30,7 @@ language:
30
  library_name: peft
31
  pipeline_tag: text-generation
32
  datasets:
33
- - AmareshHebbar/leetcode-code-gen-datasets
34
  co2_eq_emissions:
35
  emissions: 0
36
  source: "estimate, not measured with a carbon-tracking tool"
@@ -48,7 +48,7 @@ model-index:
48
  ### Qwen2.5-Coder-7B, QDoRA fine-tuned to solve LeetCode problems in Python
49
 
50
  [![Hugging Face](https://img.shields.io/badge/%F0%9F%A4%97%20Model-leetcode--python--qwen25--coder--7b-FFD21E)](https://huggingface.co/AmareshHebbar/leetcode-python-qwen25-coder-7b)
51
- [![Dataset](https://img.shields.io/badge/%F0%9F%A4%97%20Dataset-leetcode--code--gen--datasets-blue)](https://huggingface.co/datasets/AmareshHebbar/leetcode-code-gen-datasets)
52
  [![GGUF](https://img.shields.io/badge/GGUF-quantized-6f42c1)](https://huggingface.co/AmareshHebbar/leetcode-python-qwen25-coder-7b-GGUF)
53
  [![License](https://img.shields.io/badge/license-Apache%202.0-green)](https://www.apache.org/licenses/LICENSE-2.0)
54
  [![Base Model](https://img.shields.io/badge/base-Qwen2.5--Coder--7B-orange)](https://huggingface.co/unsloth/Qwen2.5-Coder-7B-Instruct)
@@ -83,7 +83,7 @@ def twoSum(nums, target):
83
  |---|---|
84
  | **Base model** | [unsloth/Qwen2.5-Coder-7B-Instruct](https://huggingface.co/unsloth/Qwen2.5-Coder-7B-Instruct) |
85
  | **Method** | QDoRA (quantized DoRA, not plain LoRA) |
86
- | **Training data** | [leetcode-code-gen-datasets](https://huggingface.co/datasets/AmareshHebbar/leetcode-code-gen-datasets) config `python` |
87
  | **Data provenance** | scraped from [doocs/leetcode](https://github.com/doocs/leetcode) (3,977 problems), execution-verified, no synthetic/LLM-generated solutions |
88
  | **Data quality** | ~70% of extracted solutions verified (execution-checked against sample I/O) |
89
  | **Weights here** | QDoRA adapter only (~160MB) — load on top of the base model |
@@ -128,7 +128,7 @@ Run `benchmark_suite.py` from the deployment kit to reproduce. All numbers are p
128
  |---|---|---|---|---|
129
  | [HumanEval-X](https://huggingface.co/datasets/THUDM/humaneval-x) | Python | _run benchmark_suite.py_ | _run benchmark_suite.py_ | 164 problems, execution-verified |
130
  | [MultiPL-E](https://huggingface.co/datasets/nuprl/MultiPL-E) (HumanEval subset) | Python | _run benchmark_suite.py_ | — | cross-check vs HumanEval-X |
131
- | Held-out LeetCode test split | Python | _run benchmark_suite.py_ | — | from `leetcode-code-gen-datasets` (`python`) test split, exact I/O match |
132
  | Tokens/sec (fp16, A40) | Python | — | — | latency benchmark, see script |
133
  | Tokens/sec (GGUF q4_k_m, CPU) | Python | — | — | latency benchmark, see script |
134
 
@@ -378,7 +378,7 @@ one-to-many problem-to-solution structure rather than picking a single "canonica
378
  | `build_language_datasets.py` | splits into per-language configs and writes the final train/val/test SFT rows |
379
 
380
  ~70% of extracted solutions verified (execution-checked against sample I/O). Full extraction/verification/build code lives alongside the
381
- [leetcode-code-gen-datasets](https://huggingface.co/datasets/AmareshHebbar/leetcode-code-gen-datasets) dataset card.
382
 
383
  ### Hyperparameters
384
 
 
30
  library_name: peft
31
  pipeline_tag: text-generation
32
  datasets:
33
+ - AmareshHebbar/leetcode-codegen-python
34
  co2_eq_emissions:
35
  emissions: 0
36
  source: "estimate, not measured with a carbon-tracking tool"
 
48
  ### Qwen2.5-Coder-7B, QDoRA fine-tuned to solve LeetCode problems in Python
49
 
50
  [![Hugging Face](https://img.shields.io/badge/%F0%9F%A4%97%20Model-leetcode--python--qwen25--coder--7b-FFD21E)](https://huggingface.co/AmareshHebbar/leetcode-python-qwen25-coder-7b)
51
+ [![Dataset](https://img.shields.io/badge/%F0%9F%A4%97%20Dataset-leetcode--codegen--python-blue)](https://huggingface.co/datasets/AmareshHebbar/leetcode-codegen-python)
52
  [![GGUF](https://img.shields.io/badge/GGUF-quantized-6f42c1)](https://huggingface.co/AmareshHebbar/leetcode-python-qwen25-coder-7b-GGUF)
53
  [![License](https://img.shields.io/badge/license-Apache%202.0-green)](https://www.apache.org/licenses/LICENSE-2.0)
54
  [![Base Model](https://img.shields.io/badge/base-Qwen2.5--Coder--7B-orange)](https://huggingface.co/unsloth/Qwen2.5-Coder-7B-Instruct)
 
83
  |---|---|
84
  | **Base model** | [unsloth/Qwen2.5-Coder-7B-Instruct](https://huggingface.co/unsloth/Qwen2.5-Coder-7B-Instruct) |
85
  | **Method** | QDoRA (quantized DoRA, not plain LoRA) |
86
+ | **Training data** | [leetcode-codegen-python](https://huggingface.co/datasets/AmareshHebbar/leetcode-codegen-python) |
87
  | **Data provenance** | scraped from [doocs/leetcode](https://github.com/doocs/leetcode) (3,977 problems), execution-verified, no synthetic/LLM-generated solutions |
88
  | **Data quality** | ~70% of extracted solutions verified (execution-checked against sample I/O) |
89
  | **Weights here** | QDoRA adapter only (~160MB) — load on top of the base model |
 
128
  |---|---|---|---|---|
129
  | [HumanEval-X](https://huggingface.co/datasets/THUDM/humaneval-x) | Python | _run benchmark_suite.py_ | _run benchmark_suite.py_ | 164 problems, execution-verified |
130
  | [MultiPL-E](https://huggingface.co/datasets/nuprl/MultiPL-E) (HumanEval subset) | Python | _run benchmark_suite.py_ | — | cross-check vs HumanEval-X |
131
+ | Held-out LeetCode test split | Python | _run benchmark_suite.py_ | — | from `leetcode-codegen-python` test split, exact I/O match |
132
  | Tokens/sec (fp16, A40) | Python | — | — | latency benchmark, see script |
133
  | Tokens/sec (GGUF q4_k_m, CPU) | Python | — | — | latency benchmark, see script |
134
 
 
378
  | `build_language_datasets.py` | splits into per-language configs and writes the final train/val/test SFT rows |
379
 
380
  ~70% of extracted solutions verified (execution-checked against sample I/O). Full extraction/verification/build code lives alongside the
381
+ [leetcode-codegen-python](https://huggingface.co/datasets/AmareshHebbar/leetcode-codegen-python) dataset card.
382
 
383
  ### Hyperparameters
384