Revert to deepset model only

config.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:affcab423cf457c5772f3964841cde223b6e8bf0beb34624ed70174bc3752f35
-size 1707
+oid sha256:e1d8e25d7e2253e2cdf052086eb65616b5c239cfa17febbeb22c28cb9601b9c8
+size 1703

decoder_model.onnx

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:ff3b4a6bb62ddd4fe4f9b541a9c6197326118fb855ea2785bb66d1687b0d86d8
-size 1222666655

decoder_with_past_model.onnx

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:ea74662431a57df2675e9d950d629e42aab2fa18c929094135e7c80203f6f627
-size 1121888515

encoder_model.onnx

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:1b1a5e7021e95441c44514c23c4316548c4bd1e428559aea2c9359787ca67a7f
-size 814841685

old/export_model.py → export_model.py

@@ -1,11 +1,22 @@
+'''
+A script to convert this normal/deepset model into an onnx model for further optimization.
+Use the local .env and install the requirements
+Make sure the model checkpoint name is correct (or convert to do this locally)
+Make sure the auth token has access to the model
+'''
+
+
 from optimum.onnxruntime import ORTModelForSeq2SeqLM
 from transformers import AutoTokenizer
+
 model_checkpoint = "getvector/earnings-transcript-summary"
 save_directory = "tmp/onnx/"
 huggingface_auth_token = "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"
+
 # Load a model from transformers and export it to ONNX
 ort_model = ORTModelForSeq2SeqLM.from_pretrained(model_checkpoint, from_transformers=True, use_auth_token=huggingface_auth_token)
 tokenizer = AutoTokenizer.from_pretrained(model_checkpoint, use_auth_token=huggingface_auth_token)
+
 # Save the onnx model and tokenizer
 ort_model.save_pretrained(save_directory)
 tokenizer.save_pretrained(save_directory)

handler.py

@@ -1,17 +0,0 @@
-from optimum.onnxruntime import ORTModelForSeq2SeqLM
-from transformers import pipeline, AutoTokenizer
-
-class EndpointHandler():
-    def __init__(self, path=""):
-        model = ORTModelForSeq2SeqLM.from_pretrained(path)
-        tokenizer = AutoTokenizer.from_pretrained(path)
-        self.pipeline = pipeline("summarization", model=model, tokenizer=tokenizer)
-
-    def __call__(self, data):
-        inputs = data.pop("inputs", data)
-        parameters = data.pop("parameters", None)
-        if parameters is not None:
-            summary = self.pipeline(inputs, **parameters)
-        else:
-            summary = self.pipeline(inputs)
-        return summary

old-req.txt

@@ -1,24 +0,0 @@
-certifi==2022.9.24
-charset-normalizer==2.1.1
-coloredlogs==15.0.1
-filelock==3.8.0
-huggingface-hub==0.10.0
-humanfriendly==10.0
-idna==3.4
-mpmath==1.2.1
-numpy==1.23.3
-optimum==1.4.0
-packaging==21.3
-protobuf==3.20.1
-pyparsing==3.0.9
-PyYAML==6.0
-regex==2022.9.13
-requests==2.28.1
-sentencepiece==0.1.97
-sympy==1.11.1
-tokenizers==0.12.1
-torch==1.12.1
-tqdm==4.64.1
-transformers==4.22.2
-typing_extensions==4.3.0
-urllib3==1.26.12

old/config.json

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:e1d8e25d7e2253e2cdf052086eb65616b5c239cfa17febbeb22c28cb9601b9c8
-size 1703

old/special_tokens_map.json

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:06e405a36dfe4b9604f484f6a1e619af1a7f7d09e34a8555eb0b77b66318067f
-size 280

old/tokenizer.json

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:4dcc90e1c6bea98fde74f85a2a307a1935fe9067bd4c9d6623c520965efce58f
-size 2108689

old/tokenizer_config.json

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:c3d09c165ebe460bc0c338606208e725ce9e1c24df798a156d43cd854fff12db
-size 375

old/vocab.json

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:ed19656ea1707df69134c4af35c8ceda2cc9860bf2c3495026153a133670ab5e
-size 798293

requirements.txt

@@ -1 +1,24 @@
-optimum[onnxruntime]==1.4.0
+certifi==2022.9.24
+charset-normalizer==2.1.1
+coloredlogs==15.0.1
+filelock==3.8.0
+huggingface-hub==0.10.0
+humanfriendly==10.0
+idna==3.4
+mpmath==1.2.1
+numpy==1.23.3
+optimum==1.4.0
+packaging==21.3
+protobuf==3.20.1
+pyparsing==3.0.9
+PyYAML==6.0
+regex==2022.9.13
+requests==2.28.1
+sentencepiece==0.1.97
+sympy==1.11.1
+tokenizers==0.12.1
+torch==1.12.1
+tqdm==4.64.1
+transformers==4.22.2
+typing_extensions==4.3.0
+urllib3==1.26.12

tokenizer_config.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9c88e5132bac58e796a8cd076e7c9d3bd93f701c1bf5454486f7d24846150025
-size 403
+oid sha256:c3d09c165ebe460bc0c338606208e725ce9e1c24df798a156d43cd854fff12db
+size 375