Optimum-Intel Notebook (#484)

* Corrections to quantization description

* fix json structure

* Update setfit-optimum-intel.ipynb

* Update setfit-optimum-intel.ipynb

notebooks/setfit-optimum-intel.ipynb

@@ -36,7 +36,7 @@
    "source": [
     "* Linux OS\n",
     "* Python 3.9+\n",
-    "* An Intel Xeon Processor which supports [bfloat16](https://www.intel.com/content/www/us/en/developer/articles/technical/pytorch-on-xeon-processors-with-bfloat16.html))\n",
+    "* An Intel Xeon Processor which supports [bfloat16](https://www.intel.com/content/www/us/en/developer/articles/technical/pytorch-on-xeon-processors-with-bfloat16.html)\n",
     "\n",
     "Since Google Colab doesn't meet the hardware requirement, it's incompatible with this notebook and will not see speedups from our optimizations.\n",
     "\n",
@@ -508,7 +508,7 @@
    "source": [
     "from setfit import SetFitModel\n",
     "\n",
-    "small_model = SetFitModel.from_pretrained(\"moshew/bge-small-en-v1.5_setfit-sst2-english\")\n",
+    "small_model = SetFitModel.from_pretrained(\"dkorat/bge-small-en-v1.5_setfit-sst2-english\")\n",
     "pb = PerformanceBenchmark(small_model, test_dataset, \"bge-small (PyTorch)\", batch_sizes=BATCH_SIZES)\n",
     "perf_metrics = pb.run_benchmark()"
    ]
@@ -610,7 +610,7 @@
     "    )\n",
     "    tokenizer.save_pretrained(output_path)\n",
     "    \n",
-    "model_name = \"moshew/bge-small-en-v1.5_setfit-sst2-english\"\n",
+    "model_name = \"dkorat/bge-small-en-v1.5_setfit-sst2-english\"\n",
     "calibration_set = load_qasper_calibration_set(sample_size=100)\n",
     "optimum_model_path = f\"/tmp/{model_name}_opt\""
    ]
@@ -747,8 +747,8 @@
       "2024-02-01 10:03:34 [INFO] [Strategy] Found the model meets accuracy requirements, ending the tuning process.\n",
       "2024-02-01 10:03:34 [INFO] Specified timeout or max trials is reached! Found a quantized model which meet accuracy goal. Exit.\n",
       "2024-02-01 10:03:34 [INFO] Save deploy yaml to /home/sdp/dkorat/setfit/notebooks/nc_workspace/2024-02-01_10-03-17/deploy.yaml\n",
-      "Model weights saved to /tmp/moshew/bge-small-en-v1.5_setfit-sst2-english_opt/pytorch_model.bin\n",
-      "Configuration saved in /tmp/moshew/bge-small-en-v1.5_setfit-sst2-english_opt/inc_config.json\n"
+      "Model weights saved to /tmp/dkorat/bge-small-en-v1.5_setfit-sst2-english_opt/pytorch_model.bin\n",
+      "Configuration saved in /tmp/dkorat/bge-small-en-v1.5_setfit-sst2-english_opt/inc_config.json\n"
      ]
     }
    ],
@@ -836,7 +836,7 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "loading configuration file /tmp/moshew/bge-small-en-v1.5_setfit-sst2-english_opt/inc_config.json\n",
+      "loading configuration file /tmp/dkorat/bge-small-en-v1.5_setfit-sst2-english_opt/inc_config.json\n",
       "INCConfig {\n",
       "  \"distillation\": {},\n",
       "  \"neural_compressor_version\": \"2.4.1\",\n",
@@ -852,7 +852,7 @@
       "}\n",
       "\n",
       "intel_extension_for_pytorch version is 2.1.100+cpu\n",
-      "2024-02-01 10:03:35,206 - sentence_transformers.SentenceTransformer - INFO - Load pretrained SentenceTransformer: moshew/bge-small-en-v1.5_setfit-sst2-english\n",
+      "2024-02-01 10:03:35,206 - sentence_transformers.SentenceTransformer - INFO - Load pretrained SentenceTransformer: dkorat/bge-small-en-v1.5_setfit-sst2-english\n",
       "2024-02-01 10:03:36,508 - sentence_transformers.SentenceTransformer - INFO - Use pytorch device_name: cpu\n"
      ]
     },
@@ -870,7 +870,7 @@
    "source": [
     "tokenizer = AutoTokenizer.from_pretrained(optimum_model_path, model_max_length=512)\n",
     "optimum_model = optimum.intel.INCModel.from_pretrained(optimum_model_path)\n",
-    "model = SetFitModel.from_pretrained(\"moshew/bge-small-en-v1.5_setfit-sst2-english\")\n",
+    "model = SetFitModel.from_pretrained(\"dkorat/bge-small-en-v1.5_setfit-sst2-english\")\n",
     "optimum_setfit_model = OptimumSetFitModel(optimum_model, tokenizer, model.model_head)\n",
     "\n",
     "# Reload test dataset\n",