push files

deployment/config.json

@@ -0,0 +1,34 @@
+{
+  "_name_or_path": "/home/alexandre/research/distilbert/pruned80_vnni/zoomodels/framework",
+  "activation": "gelu",
+  "architectures": [
+    "DistilBertForSequenceClassification"
+  ],
+  "attention_dropout": 0.1,
+  "dim": 768,
+  "dropout": 0.1,
+  "finetuning_task": "sst2",
+  "hidden_dim": 3072,
+  "id2label": {
+    "0": "negative",
+    "1": "positive"
+  },
+  "initializer_range": 0.02,
+  "label2id": {
+    "negative": 0,
+    "positive": 1
+  },
+  "max_position_embeddings": 512,
+  "model_type": "distilbert",
+  "n_heads": 12,
+  "n_layers": 6,
+  "pad_token_id": 0,
+  "problem_type": "single_label_classification",
+  "qa_dropout": 0.1,
+  "seq_classif_dropout": 0.2,
+  "sinusoidal_pos_embds": false,
+  "tie_weights_": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.18.0.dev0",
+  "vocab_size": 30522
+}

deployment/model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8f814a1a6b4f818e07d1183e2204eedd0fb8c8fdd708326e5d97ce4ee44c3e5
+size 67197076

deployment/tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "/home/alexandre/research/bert_base/sst2/framework", "tokenizer_class": "BertTokenizer"}

dev.py

@@ -0,0 +1,8 @@
+import shutil
+from sparsezoo import Model
+
+stub = "zoo:nlp/sentiment_analysis/distilbert-none/pytorch/huggingface/sst2/pruned80_quant-none-vnni"
+model = Model(stub, download_path=".")
+
+# Downloads and prints the download path of the model 
+print(model.deployment.path)

handler.py

@@ -0,0 +1,24 @@
+from typing import Dict, Any
+from deepsparse import Pipeline
+from time import perf_counter
+
+class EndpointHandler:
+
+    def __init__(self, path=""):
+
+        self.pipeline = Pipeline.create(task="text-classification", model_path=path)
+
+    def __call__(self, data: Dict[str, Any]) -> Dict[str, str]:
+        """
+        Args:
+            data (:obj:):
+                includes the deserialized image file as PIL.Image
+        """
+        inputs = data.pop("inputs", data)
+
+        start = perf_counter()
+        prediction = self.pipeline(inputs)
+        end = perf_counter()
+        delta = end - start
+
+        return prediction.json(), "latency: " + str(delta) + " secs."

requirements.txt

@@ -0,0 +1 @@
+deepsparse>=1.2.0