Update README.md

Browse files

Files changed (1) hide show

README.md +138 -195

README.md CHANGED Viewed

@@ -1,199 +1,142 @@
 ---
-library_name: transformers
-tags: []
 ---
-# Model Card for Model ID
-<!-- Provide a quick summary of what the model is/does. -->
-## Model Details
-### Model Description
-<!-- Provide a longer summary of what this model is. -->
-This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
-- **Developed by:** [More Information Needed]
-- **Funded by [optional]:** [More Information Needed]
-- **Shared by [optional]:** [More Information Needed]
-- **Model type:** [More Information Needed]
-- **Language(s) (NLP):** [More Information Needed]
-- **License:** [More Information Needed]
-- **Finetuned from model [optional]:** [More Information Needed]
-### Model Sources [optional]
-<!-- Provide the basic links for the model. -->
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
-## Uses
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
-### Direct Use
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
-[More Information Needed]
-### Downstream Use [optional]
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
-[More Information Needed]
-### Out-of-Scope Use
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
-[More Information Needed]
-## Bias, Risks, and Limitations
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
-[More Information Needed]
-### Recommendations
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
-## How to Get Started with the Model
-Use the code below to get started with the model.
-[More Information Needed]
-## Training Details
-### Training Data
-<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
-[More Information Needed]
-### Training Procedure
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
-#### Preprocessing [optional]
-[More Information Needed]
-#### Training Hyperparameters
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
-#### Speeds, Sizes, Times [optional]
-<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
-[More Information Needed]
-## Evaluation
-<!-- This section describes the evaluation protocols and provides the results. -->
-### Testing Data, Factors & Metrics
-#### Testing Data
-<!-- This should link to a Dataset Card if possible. -->
-[More Information Needed]
-#### Factors
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
-[More Information Needed]
-#### Metrics
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
-[More Information Needed]
-### Results
-[More Information Needed]
-#### Summary
-## Model Examination [optional]
-<!-- Relevant interpretability work for the model goes here -->
-[More Information Needed]
-## Environmental Impact
-<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
-Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
-- **Hardware Type:** [More Information Needed]
-- **Hours used:** [More Information Needed]
-- **Cloud Provider:** [More Information Needed]
-- **Compute Region:** [More Information Needed]
-- **Carbon Emitted:** [More Information Needed]
-## Technical Specifications [optional]
-### Model Architecture and Objective
-[More Information Needed]
-### Compute Infrastructure
-[More Information Needed]
-#### Hardware
-[More Information Needed]
-#### Software
-[More Information Needed]
-## Citation [optional]
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
-**BibTeX:**
-[More Information Needed]
-**APA:**
-[More Information Needed]
-## Glossary [optional]
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
-[More Information Needed]
-## More Information [optional]
-[More Information Needed]
-## Model Card Authors [optional]
-[More Information Needed]
-## Model Card Contact
-[More Information Needed]

 ---
+base_model: vngrs-ai/Kumru-2B-Base
+model_name: Kumru-2B-EPDK-DPO
+language:
+- tr
+license: apache-2.0
+tags:
+- base_model:adapter:vngrs-ai/Kumru-2B-Base
+- transformers
+- trl
+- peft
+- dpo
+- preference-learning
+- kumru
+- epdk
+- qlora
+- lora
+- causal-lm
+datasets:
+- ogulcanakca/epdk_dpo
+pipeline_tag: text-generation
+library_name: peft
 ---
+# Model Card for Kumru-2B-EPDK-DPO
+Bu adaptör, modele EPDK (Enerji Piyasası Düzenleme Kurumu) alanında hem **(domain knowledge)** hem de **sohbet yeteneğini** kazandırmak amacıyla `vngrs-ai/Kumru-2B-Base` modeli üzerine eğitilmiş bir **DPO (Direct Preference Optimization) LoRA adaptörü** eğitilmiştir.
+Eğitim, Gemini API kullanılarak [`ogulcanakca/epdk_corpus`](https://huggingface.co/datasets/ogulcanakca/epdk_corpus)'tan üretilen sentetik, yüksek kaliteli [`ogulcanakca/epdk_dpo`](https://huggingface.co/datasets/ogulcanakca/epdk_dpo) tercih veri seti (`prompt`, `chosen`, `rejected`) kullanılarak yapılmıştır.
+Model, 4-bit (QLoRA) ile yüklenmeli ve üzerine bu DPO adaptörü eklenmelidir.
+```python
+!pip install -q \
+    "transformers" \
+    "peft" \
+    "accelerate" \
+    "bitsandbytes" \
+    "trl" \
+    "datasets"
+from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
+from peft import PeftModel
+import torch
+base_model_name = "vngrs-ai/Kumru-2B-Base"
+adapter_name = "ogulcanakca/Kumru-2B-EPDK-DPO"
+bnb_config = BitsAndBytesConfig(
+    load_in_4bit=True,
+    bnb_4bit_quant_type="nf4",
+    # T4 için torch.float16, P100/Ampere+ için torch.bfloat16
+    bnb_4bit_compute_dtype=torch.float16
+)
+model = AutoModelForCausalLM.from_pretrained(
+    base_model_name,
+    quantization_config=bnb_config,
+    device_map="auto",
+    trust_remote_code=True
+)
+tokenizer = AutoTokenizer.from_pretrained(base_model_name, trust_remote_code=True)
+if tokenizer.pad_token is None:
+    tokenizer.pad_token = tokenizer.eos_token
+model = PeftModel.from_pretrained(model, adapter_name)
+model.eval()
+prompt_soru = "2007 yılına ait Türkiye Ortalama Elektrik Toptan Satış Fiyatının (TORETOSAF) değeri nedir?"
+messages = [
+    {"role": "user", "content": prompt_soru}
+]
+input_text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
+inputs = tokenizer(input_text, return_tensors="pt", return_token_type_ids=False).to(model.device)
+outputs = model.generate(
+    **inputs,
+    max_new_tokens=150,
+    temperature=0.2,
+    do_sample=True
+)
+print("\n--- Modelin Cevabı ---")
+print(tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True))
+````
+## WandB
+Modelin eğitimi `ogulcanakca/epdk_dpo` veri setinin `train` split'i üzerinde 1 epoch ve `dev` split'i üzerinde her 25 adımda bir değerlendirme yapılarak izlenmiştir.
+[WandB report](https://api.wandb.ai/links/ogulcanakca-none/z06caml6)
+-----
+```json
+{
+  "prompt": "Belirlenen 2007 yılı Türkiye Ortalama Elektrik Toptan Satış Fiyatının (TORETOSAF) değeri nedir?",
+  "Base Model": "Belirlenen 2007 yılı Türkiye Ortalama Elektrik Toptan Satış Fiyatının (TORETOSAF) değeri nedir? 2007 yılı Türkiye Ortalama Elektrik Toptan Satış Fiyatı 1,11 TL/kWh'dir. 2007 yılı Türkiye Ortalama Elektrik Toptan Satış Fiyatı 1,11 TL/kWh'dir. 2007 yılı Türkiye Ortalama Elektrik Toptan Satış Fiyatı 1,11 TL/kWh'dir. 2007 yılı Türkiye Ortalama Elektrik Toptan Satış Fiyatı 1,11 TL/kWh'dir.",
+  "Adapter'lı Model": "2007 yılı toptan satış fiyatı 15,80 krş/kWh olan elektrik enerjisinin fiyatı, 2007 yılı toptan satış fiyatı 15,80 krş/kWh’dir. 2007 yılı toptan satış fiyatı 15,80 krş/kWh olan toptan satış fiyatı 15,80 krş/kWh’dir. 2007 yılı toptan satış fiyatı 15,80 kr"
+}
+```
+## Eğitim Parametreleri (Kaggle P100)
+### DPOConfig
+  * **GPU:** 1x Tesla T4 (16GB)
+  * **Model:** `vngrs-ai/Kumru-2B-Base`
+  * **Veri Seti:** `ogulcanakca/epdk_dpo` (`train` 24.5k, `dev` 1.3k)
+  * **Quantization:** `4-bit (NF4)`
+  * **Compute Precision:** `float16`
+  * **`max_length`:** `1024`
+  * **`max_prompt_length`:** `512`
+  * **`optim`:** `paged_adamw_8bit`
+  * **`gradient_checkpointing`:** `True`
+  * **`learning_rate`:** `5e-5` (LR Tipi: `cosine`)
+  * **`num_train_epochs`:** `1`
+  * **`loss_type`:** `ipo`
+  * **`beta`:** `0.1`
+  * **Batch Size:**
+      * `per_device_train_batch_size`: `1`
+      * `gradient_accumulation_steps`: `8`
+      * **(Effective Batch Size: 8)**
+### LoraConfig
+  * **`r` (Rank):** `16`
+  * **`lora_alpha`:** `32`
+  * **`lora_dropout`:** `0.05`
+  * **`target_modules`:**
+      * `q_proj`
+      * `k_proj`
+      * `v_proj`
+      * `o_proj`
+      * `gate_proj`
+      * `up_proj`
+      * `down_proj`
+```