RAG implementation

rag/db.py

@@ -0,0 +1,28 @@
+from sentence_transformers import SentenceTransformer
+from datasets import load_dataset
+
+from langchain_community.embeddings import HuggingFaceEmbeddings
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_community.vectorstores import FAISS
+
+### Build and embed a dataset to the FAISS vector db
+
+dataset = load_dataset("databricks/databricks-dolly-15k")
+
+contexts = ''.join(dataset['train']['context'])
+
+text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=150, length_function=len, is_separator_regex=False)
+
+docs = text_splitter.create_documents([contexts])
+
+embedder_path = "rag-embed-model"
+
+embedder = HuggingFaceEmbeddings(
+    model_name=embedder_path,
+    model_kwargs={'device':'cpu'}, 
+    encode_kwargs={'normalize_embeddings': False} 
+)
+
+db = FAISS.from_documents(docs, embedder)
+
+db.save_local("faiss_index")

rag/query.py

@@ -0,0 +1,56 @@
+from sentence_transformers import SentenceTransformer
+from datasets import load_dataset
+
+from langchain_community.embeddings import HuggingFaceEmbeddings
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_community.vectorstores import FAISS
+from langchain.chains import RetrievalQA
+from langchain.prompts import ChatPromptTemplate
+from langchain.chat_models import ChatOpenAI
+from langchain import HuggingFacePipeline
+from langchain.schema.runnable import RunnablePassthrough
+from langchain.schema.output_parser import StrOutputParser
+from transformers import AutoTokenizer, AutoModelForQuestionAnswering, pipeline
+import torch
+
+embedder_path = "rag-embed-model"
+
+embedder = HuggingFaceEmbeddings(
+    model_name=embedder_path,     
+    model_kwargs={'device':'cpu'}, 
+    encode_kwargs={'normalize_embeddings': False} 
+)
+
+db = FAISS.load_local("faiss_index", embedder)
+retriever = db.as_retriever(search_kwargs={"k": 2})
+
+
+#################
+query = "What is on top of the Main Building at the University of Notre Dame?"
+model_name = "Intel/dynamic_tinybert" 
+#################
+
+
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForQuestionAnswering.from_pretrained(model_name)
+
+def run(query):
+    context = ''.join(c.page_content for c in retriever.get_relevant_documents(query))
+    inputs = tokenizer(query, context, return_tensors="pt")
+
+    outputs = model(**inputs)
+
+
+    start_logits = outputs.start_logits
+    end_logits = outputs.end_logits
+
+    # Get the answer span (start and end positions)
+    start_index = torch.argmax(start_logits)
+    end_index = torch.argmax(end_logits)
+
+    # Decode the answer from the token indices
+    answer = tokenizer.decode(inputs["input_ids"][0, start_index:end_index+1])
+
+    return answer
+
+print(run(query))

rag/rag-embed-model/1_Pooling/config.json

@@ -0,0 +1,9 @@
+{
+  "word_embedding_dimension": 384,
+  "pooling_mode_cls_token": false,
+  "pooling_mode_mean_tokens": true,
+  "pooling_mode_max_tokens": false,
+  "pooling_mode_mean_sqrt_len_tokens": false,
+  "pooling_mode_weightedmean_tokens": false,
+  "pooling_mode_lasttoken": false
+}

rag/rag-embed-model/README.md

@@ -0,0 +1,92 @@
+---
+library_name: sentence-transformers
+pipeline_tag: sentence-similarity
+tags:
+- sentence-transformers
+- feature-extraction
+- sentence-similarity
+
+---
+
+# {MODEL_NAME}
+
+This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 384 dimensional dense vector space and can be used for tasks like clustering or semantic search.
+
+<!--- Describe your model here -->
+
+## Usage (Sentence-Transformers)
+
+Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed:
+
+```
+pip install -U sentence-transformers
+```
+
+Then you can use the model like this:
+
+```python
+from sentence_transformers import SentenceTransformer
+sentences = ["This is an example sentence", "Each sentence is converted"]
+
+model = SentenceTransformer('{MODEL_NAME}')
+embeddings = model.encode(sentences)
+print(embeddings)
+```
+
+
+
+## Evaluation Results
+
+<!--- Describe how your model was evaluated -->
+
+For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name={MODEL_NAME})
+
+
+## Training
+The model was trained with the parameters:
+
+**DataLoader**:
+
+`torch.utils.data.dataloader.DataLoader` of length 5475 with parameters:
+```
+{'batch_size': 16, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'}
+```
+
+**Loss**:
+
+`sentence_transformers.losses.ContrastiveLoss.ContrastiveLoss` with parameters:
+  ```
+  {'distance_metric': 'SiameseDistanceMetric.COSINE_DISTANCE', 'margin': 0.5, 'size_average': True}
+  ```
+
+Parameters of the fit()-Method:
+```
+{
+    "epochs": 1,
+    "evaluation_steps": 0,
+    "evaluator": "NoneType",
+    "max_grad_norm": 1,
+    "optimizer_class": "<class 'torch.optim.adamw.AdamW'>",
+    "optimizer_params": {
+        "lr": 2e-05
+    },
+    "scheduler": "WarmupLinear",
+    "steps_per_epoch": null,
+    "warmup_steps": 100,
+    "weight_decay": 0.01
+}
+```
+
+
+## Full Model Architecture
+```
+SentenceTransformer(
+  (0): Transformer({'max_seq_length': 256, 'do_lower_case': False}) with Transformer model: BertModel 
+  (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False})
+  (2): Normalize()
+)
+```
+
+## Citing & Authors
+
+<!--- Describe where people can find more information -->

rag/rag-embed-model/config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "sentence-transformers/all-MiniLM-L6-v2",
+  "architectures": [
+    "BertModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 384,
+  "initializer_range": 0.02,
+  "intermediate_size": 1536,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 6,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.37.2",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

rag/rag-embed-model/config_sentence_transformers.json

@@ -0,0 +1,7 @@
+{
+  "__version__": {
+    "sentence_transformers": "2.0.0",
+    "transformers": "4.6.1",
+    "pytorch": "1.8.1"
+  }
+}

rag/rag-embed-model/modules.json

@@ -0,0 +1,20 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  },
+  {
+    "idx": 2,
+    "name": "2",
+    "path": "2_Normalize",
+    "type": "sentence_transformers.models.Normalize"
+  }
+]

rag/rag-embed-model/sentence_bert_config.json

@@ -0,0 +1,4 @@
+{
+  "max_seq_length": 256,
+  "do_lower_case": false
+}

rag/rag-embed-model/special_tokens_map.json

@@ -0,0 +1,37 @@
+{
+  "cls_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}