Qwen2.5 Pipeline

phi3_model/data/source.txt

@@ -0,0 +1,35 @@
+Apollo 11 – Lunar Descent and Moonwalk
+
+As the descent began, Armstrong and Aldrin found themselves passing landmarks on the
+surface two or three seconds early, and reported that they were “long”; they would land
+miles west of their target point. Eagle was traveling too fast. The problem could have been
+mascons—concentrations of high mass in a region or regions of the Moon’s crust that
+contains a gravitational anomaly, potentially altering Eagle’s trajectory.
+
+Five minutes into the descent burn, and 6,000 feet (1,800 m) above the surface of the
+Moon, the LM guidance computer (LGC) distracted the crew with the first of several
+unexpected 1201 and 1202 program alarms. Inside Mission Control Center, computer
+engineer Jack Garman told Guidance Officer Steve Bales it was safe to continue the
+descent, and this was relayed to the crew. The program alarms indicated “executive
+overflows”, meaning the guidance computer could not complete all its tasks in real-time and
+had to postpone some of them. Margaret Hamilton, the Director of Apollo Flight Computer
+Programming at the MIT Charles Stark Draper Laboratory later recalled: “To blame the
+computer for the Apollo 11 problems is like blaming the person who spots a fire and calls
+the fire department. Actually, the computer was programmed to do more than recognize
+error conditions. A complete set of recovery programs was incorporated into the software.
+The software’s action, in this case, was to eliminate lower priority tasks and re-establish
+the more important ones. The computer, rather than almost forcing an abort, prevented an
+abort. If the computer hadn’t recognized this problem and taken recovery action, I doubt if
+Apollo 11 would have been the successful Moon landing it was.”
+
+When Armstrong again looked outside, he saw that the computer’s landing target was in a
+boulder-strewn area just north and east of a 300-foot-diameter (91 m) crater, so he took
+semi-automatic control. Throughout the descent, Aldrin called out navigation data to
+Armstrong, who was busy piloting Eagle. Now 107 feet (33 m) above the surface,
+Armstrong knew their propellant supply was dwindling and was determined to land at the
+first possible landing site.
+
+Armstrong found a clear patch of ground and maneuvered the spacecraft towards it. They
+were now 100 feet (30 m) from the surface, with only 90 seconds of propellant remaining.
+Lunar dust kicked up by the LM’s engine began to impair his ability to determine the
+spacecraft’s motion.

phi3_model/emissions.csv

@@ -0,0 +1,2 @@
+timestamp,project_name,run_id,experiment_id,duration,emissions,emissions_rate,cpu_power,gpu_power,ram_power,cpu_energy,gpu_energy,ram_energy,energy_consumed,water_consumed,country_name,country_iso_code,region,cloud_provider,cloud_region,os,python_version,codecarbon_version,cpu_count,cpu_model,gpu_count,gpu_model,longitude,latitude,ram_total_size,tracking_mode,on_cloud,pue,wue
+2025-11-13T06:09:56,codecarbon,c909bc4e-b083-4230-a78a-99ebf1ebeb1e,5b0fa12a-3dd7-45bb-9766-cc326314d9f1,0.08208100000047125,0.013644424732649086,0.16623121955837222,42.5,0.0,10.0,0.01936782330256946,0.0,0.00455692013805555,0.023924743440625004,0.0,Egypt,EGY,,,,Windows-11-10.0.26100-SP0,3.13.5,3.0.8,12,12th Gen Intel(R) Core(TM) i5-1235U,,,,,7.692127227783203,machine,N,1.0,0.0

phi3_model/emissions_base_c909bc4e-b083-4230-a78a-99ebf1ebeb1e.csv

@@ -0,0 +1,3 @@
+task_name,timestamp,project_name,run_id,duration,emissions,emissions_rate,cpu_power,gpu_power,ram_power,cpu_energy,gpu_energy,ram_energy,energy_consumed,water_consumed,country_name,country_iso_code,region,cloud_provider,cloud_region,os,python_version,codecarbon_version,cpu_count,cpu_model,gpu_count,gpu_model,longitude,latitude,ram_total_size,tracking_mode,on_cloud
+Recursive RAG Query,2025-11-13T06:09:34,codecarbon,c909bc4e-b083-4230-a78a-99ebf1ebeb1e,142.94884759999877,0.001188882070760797,8.316882338340505e-06,42.5,0.0,10.0,0.0016875667336111143,0.0,0.00039707216333332955,0.002084638896944442,0.0,Egypt,EGY,,,,Windows-11-10.0.26100-SP0,3.13.5,3.0.8,12,12th Gen Intel(R) Core(TM) i5-1235U,,,,,7.692127227783203,machine,N
+Recursive RAG Query_f351b95b-3d59-40c2-9d44-ffb9a25be233,2025-11-13T06:09:56,codecarbon,c909bc4e-b083-4230-a78a-99ebf1ebeb1e,0.07393939999928989,6.526370375566776e-07,8.310068383589976e-06,42.5,0.0,10.0,9.271493055622304e-07,0.0,2.172136111108433e-07,1.144362916673941e-06,0.0,Egypt,EGY,,,,Windows-11-10.0.26100-SP0,3.13.5,3.0.8,12,12th Gen Intel(R) Core(TM) i5-1235U,,,,,7.692127227783203,machine,N

phi3_model/requirements.txt

@@ -0,0 +1,8 @@
+huggingface-hub
+sentence-transformers
+transformers
+torch
+faiss-cpu
+python-dotenv
+accelerate
+numpy

qwen_model/README.md

@@ -0,0 +1,57 @@
+<!-- markdownlint-disable MD013 -->
+# QwenModel: Qwen 2.5 + RAG + Recursive Editing
+
+This project experiments with the open-sourced model **Qwen 2.5** (2.5B parameters) combined with **Retrieval-Augmented Generation (RAG)** and **recursive editing** to test its performance across different prompt categories from the Apollo 11 dataset.
+
+## Project Structure
+
+* **main.ipynb** – The core notebook where Qwen 2.5 is set up, RAG is integrated, and recursive editing is implemented.
+* **output.md** – Contains the final prompts and responses generated through recursive editing, along with tracked energy and CO₂ emission metrics.
+
+## Workflow Overview
+
+### 1. Prompt Setup
+
+The notebook defines three structured prompts for iterative refinement:
+
+* **Draft Prompt** – Produces the initial answer.
+* **Critic Prompt** – Analyzes weaknesses in the draft.
+* **Refiner Prompt** – Improves the answer based on the critique.
+
+This setup enables controlled recursive editing, especially useful for summarization, creative writing, and paraphrasing.
+
+### 2. RAG Pipeline
+
+* Connected to the Hugging Face API to generate **embeddings**.
+* Built a **vector index** from the Apollo 11 dataset.
+* Prepared a **retriever** to supply relevant context for Qwen.
+
+### 3. Qwen Integration
+
+A custom function handles:
+
+* Retrieval of context
+* Construction of Qwen API calls
+* Application of recursive editing logic (3 iterations)
+
+### 4. Energy Tracking
+
+Each query is wrapped with **CodeCarbon** to record:
+
+* CPU usage
+* Energy consumption
+* CO₂ emissions
+
+All tracked metrics appear in `output.md`.
+
+## Output
+
+The final markdown file contains:
+
+* The prompts used
+* Recursive-editing improved final responses
+* Per-question energy and emission statistics
+
+## Purpose
+
+This setup provides a compact testing environment to evaluate Qwen 2.5’s performance with RAG and recursive editing, enabling comparison with larger or alternative models.