chore: push model from local

.gitignore

@@ -206,5 +206,5 @@ marimo/_static/
 marimo/_lsp/
 __marimo__/
 
-# Kafka Commands
-command.txt
+# Temp files
+temp.py

app/models/twitter-roberta-base-sentiment-latest/config.json

@@ -0,0 +1,37 @@
+{
+  "_name_or_path": "/home/jupyter/misc/tweeteval/TweetEval_models/sentiment/sentiment_latest_2021/",
+  "architectures": [
+    "RobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "negative",
+    "1": "neutral",
+    "2": "positive"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "negative": 0,
+    "neutral": 1,
+    "positive": 2
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.13.0.dev0",
+  "type_vocab_size": 1,
+  "vocab_size": 50265
+}

app/models/twitter-roberta-base-sentiment-latest/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d24a3e32a88ed1c4e5b789fc6644e2e767500554e954b27dccf52a8e762cbae
+size 501045531

app/models/twitter-roberta-base-sentiment-latest/special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}}

app/services/sentiment_service.py

@@ -1,6 +1,8 @@
 import os
+from pathlib import Path
 from typing import List, Dict, Any
-from app.core.config import settings
+
+# from app.core.config import settings
 
 import torch
 import numpy as np
@@ -27,14 +29,29 @@ class SentimentService:
             print("GPU not found. Loading model onto CPU.")
 
         # Load model, tokenizer, and config (for id2label mapping)
-        model_name = settings.SENTIMENT_MODEL
-        cache_dir = os.getenv("TRANSFORMERS_CACHE", "/tmp/hf_cache")
 
-        self.tokenizer = AutoTokenizer.from_pretrained(model_name, cache_dir=cache_dir)
-        self.config = AutoConfig.from_pretrained(model_name)
-        self.model = AutoModelForSequenceClassification.from_pretrained(
-            model_name, cache_dir=cache_dir
-        ).to(self.device)
+        BASE_DIR = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+        MODEL_DIR = os.path.join(
+            BASE_DIR, "models", "twitter-roberta-base-sentiment-latest"
+        )
+        print(MODEL_DIR)
+
+        if not os.path.exists(MODEL_DIR):
+            raise FileNotFoundError(f"Model folder not found: {MODEL_DIR}")
+
+        # Load tokenizer, config, model
+        self.tokenizer = AutoTokenizer.from_pretrained(MODEL_DIR)
+        self.config = AutoConfig.from_pretrained(MODEL_DIR)
+        self.model = AutoModelForSequenceClassification.from_pretrained(MODEL_DIR).to(
+            self.device
+        )
+
+        # self.tokenizer = AutoTokenizer.from_pretrained(model_source)
+        # self.config = AutoConfig.from_pretrained(model_source)
+        # self.model = AutoModelForSequenceClassification.from_pretrained(
+        #     model_source
+        # ).to(self.device)
+
         self.model.eval()  # set model to inference mode
         print("Sentiment model loaded successfully.")