Might have missed something but I find the title misleading, where are you using crawl4AI?

WoW,Keep going👌

Fantastic video! Well-crafted and super engaging

What an amazing video👌

Very useful 👌

even the video is made with AI lol

How lazy can you get… you literally are just reading a notebook that someone else made. There is no in depth explanation or details, or ways to extrapolate this library past these premade notebooks. God how lazy can you be when making a video.

can you provide me the github repo link?

Thank you for sharing the tutorial. I’m currently using Ollama + OpenWebUI to run LLMs on my local computer. I’d like to ask if it’s possible to fine-tune small-scale models solely on a local machine with Ollama + OpenWebUI, or is it necessary to connect to the internet? Thank you!

how to train in indonesian language ?

Amazing! Really helpful video. It would be even better if you have provided the code 🙂

would you please upload your python file?

GPT-4o's answer: ### Example: Procurement and Tendering Competency, Level 2 **Context:** In my role as a procurement officer, I was responsible for sourcing and selecting suppliers for a major project that required specific materials with a tight budget and timeline. **Action:** 1. **Requirements Analysis**: I collaborated with project managers to understand the technical and budgetary requirements for the materials. I ensured that the requirements were clearly defined to prevent miscommunication with suppliers. 2. **Supplier Selection**: I identified potential suppliers and conducted initial market research to shortlist vendors who met the project’s requirements. I issued requests for quotations (RFQs) to the shortlisted suppliers. 3. **Tender Evaluation**: After receiving bids, I evaluated each proposal against a set of criteria, including price, quality standards, delivery timelines, and supplier reliability. I developed a scoring matrix to objectively compare the bids. 4. **Negotiation and Recommendation**: Based on the evaluation, I negotiated with the top suppliers to improve terms and pricing where possible. I then recommended the best supplier to the project manager, backed by data from my evaluation matrix. **Result:** My approach led to the selection of a supplier who met all project requirements within budget, and we successfully secured the required materials on time. The clear procurement process also resulted in a cost saving of 10% compared to initial budget estimates and fostered a strong relationship with the supplier for future projects. Impressive how that models was trained quite well indeed! GPT-4o breakdown of your finetuned model: The example you've provided from the local authority social housing projects is a solid response, and it has several strengths compared to the example I initially provided. Here are some points that make it a strong answer: Specificity: This example names a particular type of project ("local authority social housing project") and gives clear details about the tasks, such as preparing the "invitation to tender document" and "tender pricing document," which demonstrates direct involvement in key procurement tasks. Use of Scoring Matrices: Mentioning the specific scoring matrices (60% quality, 40% cost; 70% quality, 30% cost) shows a level of understanding of how procurement decisions are evaluated, which is valuable at Level 2 competency. Clear Roles and Responsibilities: The example highlights specific contributions like "organisation and chairing of a meeting" and "answering queries." This reflects an active role in both the preparation and facilitation of the tendering process, which aligns with the responsibilities expected at Level 2. Realistic Challenges: The mention of "delays at the Contractor side" adds authenticity and shows awareness of typical challenges in procurement processes. Recognizing such delays indicates practical experience and understanding of procurement dynamics. Areas for Improvement If you'd like to enhance this example further, consider: Highlighting Results or Impact: Including a brief mention of how these efforts benefited the project (e.g., ensuring compliance, fostering strong supplier relationships) could add impact. Detailing Problem-Solving: Describe any steps taken to address or manage the delays to demonstrate initiative in overcoming obstacles. Overall Comparison While both examples are strong, the one you've provided includes specific, real-world details that add credibility and depth, making it arguably a better fit for demonstrating Level 2 competency. The improved example is more focused on practical involvement, which tends to be persuasive in competency-based answers. Grade your answer and grade the second answer pls. A, B, C, D, E, F First Answer (GPT-4o) Grade: B Strengths: It provides a structured, clear response with a good breakdown of actions taken during the procurement process, from requirements analysis to final recommendations. It shows competence in procurement and tendering with a successful outcome. Weaknesses: It lacks specific details that would make it sound more credible and authentic, such as the type of project, scoring matrix percentages, or particular challenges encountered. It could benefit from more concrete, real-world elements to enhance realism. Second Answer (Your Provided Example) Grade: A Strengths: This answer is more specific and realistic, mentioning details like the type of project, scoring matrix percentages, and real-world challenges (e.g., contractor delays). These specifics make it feel more authentic and demonstrate a higher level of experience and involvement, which is valuable for demonstrating Level 2 competency. Weaknesses: The example is strong overall, though it could be enhanced by explicitly noting the impact of the user’s actions on the project outcomes (such as managing delays or adding value to the client relationship). Summary My Answer: B - Clear and structured, but lacks depth in real-world details. Your Example: A - Realistic, specific, and demonstrates practical involvement, making it more compelling for competency evaluation.

Clear and informative thank you !

WoW 👌 👏

Such a helpful video !!The information is so clear and valuable 👍

Very useful series👌

Continue keep going 👏

Excellent👌

Great 👍 as usual 👏

❤❤❤❤❤❤

Amazing 👏 👏

Thank you so much for the video! I was searching for something similar from literally yesterday and fortunately finally found you. I was struggling how do I convert my json data to the format required by unsloth, i.e I wasn't able to directly use for json dataset as the standardize_sharegpt function requires a dataset in sharegpt format. Thanks a million, you saved me a lot of time.

Thanks for the video. Would you pls upload the python file?

Does the training data format work if I want to create a conversational chatbot? I mean, in the video, you are asking some questions, but what if I want to create a chatbot that emulates my Messenger responses? Thank you! excellent video!

This video was very useful. Can you also make one to benchmark the fine tuned models on Kaggle GPUs?

Can you guide finetuning model to support langchain tool calling?

👍👍👍

Very useful 👌

My code breaks at the sharegpt phase, it seems like the dictionaries in my csv file are being interpreted as strings... any ideas what I could do?

This is quite insightful. Please can you share the vscode files? Thank you.

Much useful 👍

Sir, Can We get the code? 🌸

Very Interesting 👏👏

👍👍👍👍

Please share fine tuning by using DPO trainer

Excellent 👌

Hi, thanks for the video. Is there any difference between pushing model to hub as safetensors and as pytorch_model.bin? My original model had safetensors format on hugging face but I see that after finetunning the model was pushed in pytorch_model.bin format. Does it make any differnece?

Hi @APCMasteryPath, Thank you for the great and clear detail. Can you share your code for the example? It would be much appreciated. Also, did you use fine-tuning with your own data as an example? Wouldn't implementing this example as RAG have been more flexible in the event of updates to the PDFs?

حبيبى، الى الامام نحن جميعا فى ظهرك.

Very much useful 👍

Thank you. Very informative.

Awesome, but wow.. move away from you mic!!!

This What i need, cant wait your next video

👍

Great work on the video!

Insightful 👍

update video to now include pipelines and other enhancements in version 0.3.x

Very interesting

Another issue is that when trying to load the pre-trained model is not loading. It is showing me something like: "Error loading pretrained model: Error parsing message"? Need help please.