Add application file

.gitignore

@@ -0,0 +1,3 @@
+trained_model/
+dist/
+.idea

Dockerfile

@@ -0,0 +1,16 @@
+# read the doc: https://huggingface.co/docs/hub/spaces-sdks-docker
+# you will also find guides on how best to write your Dockerfile
+
+FROM python:3.9
+
+RUN useradd -m -u 1000 user
+
+WORKDIR /app
+
+COPY --chown=user ./requirements.txt requirements.txt
+
+RUN pip install --no-cache-dir --upgrade -r requirements.txt
+
+COPY --chown=user . /app
+
+CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]

README.md

@@ -1,11 +1,7 @@
----
-title: Restricted Item Detector
-emoji: 👁
-colorFrom: gray
-colorTo: green
-sdk: docker
-pinned: false
-license: mit
----
+# hackweek2024-sup-genai-tools
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+## Poetry setup
+to start, first install dependency by
+`$ poetry install`
+then you can start the venv by
+`$ poetry shell`

app.py

@@ -0,0 +1,7 @@
+from fastapi import FastAPI
+
+app = FastAPI()
+
+@app.get("/")
+def greet_json():
+  return {"Hello": "World!"}

csv/block_items.csv

@@ -0,0 +1,34 @@
+Date,Host,Service,Blocked Cart Items,Message
+"2024-06-25T21:51:35.523Z","""i-0ec7dc300989b0f58""","""cash-commerce-browser""","""[{\""name\"":\""Promotional Email GiftCard $10\"",\""image_url\"":\""https://target.scene7.com/is/image/Target//GUEST_337c880d-e2c0-4794-9784-962c63cf0646?qlt=80&fmt=pjpeg\""}, {\""name\"":\""Promotional Email GiftCard $10\"",\""image_url\"":\""https://target.scene7.com/is/image/Target//GUEST_337c880d-e2c0-4794-9784-962c63cf0646?qlt=80&fmt=pjpeg\""}]""","Found restricted items: [ShoppingCartProduct{name=Promotional Email GiftCard $10, image_url=https://target.scene7.com/is/image/Target//GUEST_337c880d-e2c0-4794-9784-962c63cf0646?qlt=80&fmt=pjpeg}, ShoppingCartProduct{name=Promotional Email GiftCard $10, image_url=https://target.scene7.com/is/image/Target//GUEST_337c880d-e2c0-4794-9784-962c63cf0646?qlt=80&fmt=pjpeg}]; {customer_token=C_gqed91mw3, flow_token=4s6ZZrxNlMdUWqbN59nXHURXi, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_3mfvyukp, sup_token=BRAND_57fqf0l740hpnbxev71c6d56e}"
+"2024-06-25T21:43:14.348Z","""i-0ec7dc300989b0f58""","""cash-commerce-browser""","""[{\""name\"":\""$100 Vanilla® Visa® Gift Box Gift Card (plus $5.44 Purchase Fee)\"",\""url\"":\""https://www.walmart.com/ip/seort/55844027\""}]""","Found restricted items: [ShoppingCartProduct{name=$100 Vanilla® Visa® Gift Box Gift Card (plus $5.44 Purchase Fee), url=https://www.walmart.com/ip/seort/55844027}]; {customer_token=C_86dpxgy6b, flow_token=4c546519-bc7c-4ae7-bbf0-5daf158351b1, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_fopbedhc, sup_token=null}"
+"2024-06-25T21:43:05.816Z","""i-044ba56b80a95063d""","""cash-commerce-browser""","""[{\""name\"":\""Promotional Email GiftCard $10\"",\""image_url\"":\""https://target.scene7.com/is/image/Target//GUEST_337c880d-e2c0-4794-9784-962c63cf0646?qlt=80&fmt=pjpeg\""}]""","Found restricted items: [ShoppingCartProduct{name=Promotional Email GiftCard $10, image_url=https://target.scene7.com/is/image/Target//GUEST_337c880d-e2c0-4794-9784-962c63cf0646?qlt=80&fmt=pjpeg}]; {customer_token=C_fsexdpmbh, flow_token=fRqPMaEwTcPsgLuLQ8mCvf9df, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_3mfvyukp, sup_token=BRAND_57fqf0l740hpnbxev71c6d56e}"
+"2024-06-25T21:35:36.381Z","""i-0ec7dc300989b0f58""","""cash-commerce-browser""","""[{\""name\"":\""GPS7000 GPS Tracker for Vehicles - Hidden Tracking Device for Any Vehicle - Easy Installation on Car's Battery- 10 Days of Service - Subscription Required - Low Cost Subscription Plan Options\"",\""url\"":\""https://www.amazon.com/gp/aw/d/B0BYK99LZC/ref=ox_sc_act_title_delete_4?smid=AWU3BDL6BD5T5&psc=1\"",\""image_url\"":\""https://m.media-amazon.com/images/I/61E6Cex5dUL._AC_AA135_.jpg\""}]""","Found restricted items: [ShoppingCartProduct{name=GPS7000 GPS Tracker for Vehicles - Hidden Tracking Device for Any Vehicle - Easy Installation on Car's Battery- 10 Days of Service - Subscription Required - Low Cost Subscription Plan Options, url=https://www.amazon.com/gp/aw/d/B0BYK99LZC/ref=ox_sc_act_title_delete_4?smid=AWU3BDL6BD5T5&psc=1, image_url=https://m.media-amazon.com/images/I/61E6Cex5dUL._AC_AA135_.jpg}]; {customer_token=C_5cpf4hygz, flow_token=73658682-d99e-4742-b6b4-8ebb7f1543ad, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_ih4h4tzb, sup_token=BRAND_djwngg0n5zluxwp131plevfvw}"
+"2024-06-25T21:35:24.733Z","""i-044ba56b80a95063d""","""cash-commerce-browser""","""[{\""name\"":\""GPS7000 GPS Tracker for Vehicles - Hidden Tracking Device for Any Vehicle - Easy Installation on Car's Battery- 10 Days of Service - Subscription Required - Low Cost Subscription Plan Options\"",\""url\"":\""https://www.amazon.com/gp/aw/d/B0BYK99LZC/ref=ox_sc_act_title_delete_4?smid=AWU3BDL6BD5T5&psc=1\"",\""image_url\"":\""https://m.media-amazon.com/images/I/61E6Cex5dUL._AC_AA135_.jpg\""}]""","Found restricted items: [ShoppingCartProduct{name=GPS7000 GPS Tracker for Vehicles - Hidden Tracking Device for Any Vehicle - Easy Installation on Car's Battery- 10 Days of Service - Subscription Required - Low Cost Subscription Plan Options, url=https://www.amazon.com/gp/aw/d/B0BYK99LZC/ref=ox_sc_act_title_delete_4?smid=AWU3BDL6BD5T5&psc=1, image_url=https://m.media-amazon.com/images/I/61E6Cex5dUL._AC_AA135_.jpg}]; {customer_token=C_5cpf4hygz, flow_token=73658682-d99e-4742-b6b4-8ebb7f1543ad, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_ih4h4tzb, sup_token=BRAND_djwngg0n5zluxwp131plevfvw}"
+"2024-06-25T21:35:20.107Z","""i-003fbf26706b7af3e""","""cash-commerce-browser""","""[{\""name\"":\""GPS7000 GPS Tracker for Vehicles - Hidden Tracking Device for Any Vehicle - Easy Installation on Car's Battery- 10 Days of Service - Subscription Required - Low Cost Subscription Plan Options\"",\""url\"":\""https://www.amazon.com/gp/aw/d/B0BYK99LZC/ref=ox_sc_act_title_delete_4?smid=AWU3BDL6BD5T5&psc=1\"",\""image_url\"":\""https://m.media-amazon.com/images/I/61E6Cex5dUL._AC_AA135_.jpg\""}]""","Found restricted items: [ShoppingCartProduct{name=GPS7000 GPS Tracker for Vehicles - Hidden Tracking Device for Any Vehicle - Easy Installation on Car's Battery- 10 Days of Service - Subscription Required - Low Cost Subscription Plan Options, url=https://www.amazon.com/gp/aw/d/B0BYK99LZC/ref=ox_sc_act_title_delete_4?smid=AWU3BDL6BD5T5&psc=1, image_url=https://m.media-amazon.com/images/I/61E6Cex5dUL._AC_AA135_.jpg}]; {customer_token=C_5cpf4hygz, flow_token=73658682-d99e-4742-b6b4-8ebb7f1543ad, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_ih4h4tzb, sup_token=BRAND_djwngg0n5zluxwp131plevfvw}"
+"2024-06-25T21:35:01.987Z","""i-003fbf26706b7af3e""","""cash-commerce-browser""","""[{\""name\"":\""GPS7000 GPS Tracker for Vehicles - Hidden Tracking Device for Any Vehicle - Easy Installation on Car's Battery- 10 Days of Service - Subscription Required - Low Cost Subscription Plan Options\"",\""url\"":\""https://www.amazon.com/gp/aw/d/B0BYK99LZC/ref=ox_sc_act_title_delete_4?smid=AWU3BDL6BD5T5&psc=1\"",\""image_url\"":\""https://m.media-amazon.com/images/I/61E6Cex5dUL._AC_AA135_.jpg\""}]""","Found restricted items: [ShoppingCartProduct{name=GPS7000 GPS Tracker for Vehicles - Hidden Tracking Device for Any Vehicle - Easy Installation on Car's Battery- 10 Days of Service - Subscription Required - Low Cost Subscription Plan Options, url=https://www.amazon.com/gp/aw/d/B0BYK99LZC/ref=ox_sc_act_title_delete_4?smid=AWU3BDL6BD5T5&psc=1, image_url=https://m.media-amazon.com/images/I/61E6Cex5dUL._AC_AA135_.jpg}]; {customer_token=C_5cpf4hygz, flow_token=73658682-d99e-4742-b6b4-8ebb7f1543ad, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_ih4h4tzb, sup_token=BRAND_djwngg0n5zluxwp131plevfvw}"
+"2024-06-25T21:34:07.013Z","""i-044ba56b80a95063d""","""cash-commerce-browser""","""[{\""name\"":\""$50 Vanilla Visa eGift Card\"",\""url\"":\""https://www.samsclub.com/p/vanilla-e-gift-visa-various-amount/prod25810992\"",\""image_url\"":\""https://scene7.samsclub.com/is/image/samsclub/0079936655659_A?wid=200&hei=200\""}]""","Found restricted items: [ShoppingCartProduct{name=$50 Vanilla Visa eGift Card, url=https://www.samsclub.com/p/vanilla-e-gift-visa-various-amount/prod25810992, image_url=https://scene7.samsclub.com/is/image/samsclub/0079936655659_A?wid=200&hei=200}]; {customer_token=C_haed91mp3, flow_token=Q81SuACF2yWsd6Fn6PRWDsBK4, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_dsjbdob6, sup_token=null}"
+"2024-06-25T21:33:34.552Z","""i-0ec7dc300989b0f58""","""cash-commerce-browser""","""[{\""name\"":\""GPS7000 GPS Tracker for Vehicles - Hidden Tracking Device for Any Vehicle - Easy Installation on Car's Battery- 10 Days of Service - Subscription Required - Low Cost Subscription Plan Options\"",\""url\"":\""https://www.amazon.com/gp/aw/d/B0BYK99LZC/ref=ox_sc_act_title_delete_4?smid=AWU3BDL6BD5T5&psc=1\"",\""image_url\"":\""https://m.media-amazon.com/images/I/61E6Cex5dUL._AC_AA135_.jpg\""}]""","Found restricted items: [ShoppingCartProduct{name=GPS7000 GPS Tracker for Vehicles - Hidden Tracking Device for Any Vehicle - Easy Installation on Car's Battery- 10 Days of Service - Subscription Required - Low Cost Subscription Plan Options, url=https://www.amazon.com/gp/aw/d/B0BYK99LZC/ref=ox_sc_act_title_delete_4?smid=AWU3BDL6BD5T5&psc=1, image_url=https://m.media-amazon.com/images/I/61E6Cex5dUL._AC_AA135_.jpg}]; {customer_token=C_5cpf4hygz, flow_token=73658682-d99e-4742-b6b4-8ebb7f1543ad, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_ih4h4tzb, sup_token=BRAND_djwngg0n5zluxwp131plevfvw}"
+"2024-06-25T21:31:29.788Z","""i-0c92416a905681db6""","""cash-commerce-browser""","""[{\""name\"":\""$200 Vanilla Visa Shiny Bow Gift Card (plus $6.88 Purchase Fee)\"",\""url\"":\""https://www.walmart.com/ip/seort/1042177970\""}]""","Found restricted items: [ShoppingCartProduct{name=$200 Vanilla Visa Shiny Bow Gift Card (plus $6.88 Purchase Fee), url=https://www.walmart.com/ip/seort/1042177970}]; {customer_token=C_smecd0mhm, flow_token=wnHQN78zduV3T55TsN8s0CKHA, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_fopbedhc, sup_token=BRAND_4sxo1et5hf8sug8lh500413rj}"
+"2024-06-25T21:30:38.963Z","""i-044ba56b80a95063d""","""cash-commerce-browser""","""[{\""name\"":\""$200 Vanilla Visa Shiny Bow Gift Card (plus $6.88 Purchase Fee)\"",\""image_url\"":\""https://i5.walmartimages.com/seo/200-Vanilla-Visa-Shiny-Bow-Gift-Card-plus-6-88-Purchase-Fee_1cc86583-2dcb-4677-bca4-16f685df6170.d43209d754cbb7ab3c09354d6cf906a7.jpeg?odnHeight=48&odnWidth=48&odnBg=FFFFFF\""}]""","Found restricted items: [ShoppingCartProduct{name=$200 Vanilla Visa Shiny Bow Gift Card (plus $6.88 Purchase Fee), image_url=https://i5.walmartimages.com/seo/200-Vanilla-Visa-Shiny-Bow-Gift-Card-plus-6-88-Purchase-Fee_1cc86583-2dcb-4677-bca4-16f685df6170.d43209d754cbb7ab3c09354d6cf906a7.jpeg?odnHeight=48&odnWidth=48&odnBg=FFFFFF}]; {customer_token=C_smecd0mhm, flow_token=CCvc7KfXlUWnhdqtcxORrG4Sp, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_fopbedhc, sup_token=BRAND_4sxo1et5hf8sug8lh500413rj}"
+"2024-06-25T21:29:54.170Z","""i-0ec7dc300989b0f58""","""cash-commerce-browser""","""[{\""name\"":\""$200 Vanilla Visa Shiny Bow Gift Card (plus $6.88 Purchase Fee)\"",\""url\"":\""https://www.walmart.com/ip/seort/1042177970\""}]""","Found restricted items: [ShoppingCartProduct{name=$200 Vanilla Visa Shiny Bow Gift Card (plus $6.88 Purchase Fee), url=https://www.walmart.com/ip/seort/1042177970}]; {customer_token=C_smecd0mhm, flow_token=oucI3kCcOLy76Hiv8Tb2vxgfa, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_fopbedhc, sup_token=BRAND_4sxo1et5hf8sug8lh500413rj}"
+"2024-06-25T21:26:19.296Z","""i-0c92416a905681db6""","""cash-commerce-browser""","""[{\""name\"":\""$200 Vanilla® Visa® Gift Box Gift Card (plus $6.88 Purchase Fee)\"",\""url\"":\""https://www.walmart.com/ip/seort/20625678\""}]""","Found restricted items: [ShoppingCartProduct{name=$200 Vanilla® Visa® Gift Box Gift Card (plus $6.88 Purchase Fee), url=https://www.walmart.com/ip/seort/20625678}]; {customer_token=C_002v20mt1, flow_token=iNJAgMr3PvHc07lbZAkdrnEIn, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_fopbedhc, sup_token=BRAND_4sxo1et5hf8sug8lh500413rj}"
+"2024-06-25T21:20:06.661Z","""i-003fbf26706b7af3e""","""cash-commerce-browser""","""[{\""name\"":\""Promotional Email GiftCard $20\"",\""image_url\"":\""https://target.scene7.com/is/image/Target//GUEST_337c880d-e2c0-4794-9784-962c63cf0646?qlt=80&fmt=pjpeg\""}, {\""name\"":\""Promotional Email GiftCard $10\"",\""image_url\"":\""https://target.scene7.com/is/image/Target//GUEST_337c880d-e2c0-4794-9784-962c63cf0646?qlt=80&fmt=pjpeg\""}]""","Found restricted items: [ShoppingCartProduct{name=Promotional Email GiftCard $20, image_url=https://target.scene7.com/is/image/Target//GUEST_337c880d-e2c0-4794-9784-962c63cf0646?qlt=80&fmt=pjpeg}, ShoppingCartProduct{name=Promotional Email GiftCard $10, image_url=https://target.scene7.com/is/image/Target//GUEST_337c880d-e2c0-4794-9784-962c63cf0646?qlt=80&fmt=pjpeg}]; {customer_token=C_ssdwxqyqz, flow_token=roXsSw6mIgIqe5XjF8bO8tH2o, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_3mfvyukp, sup_token=BRAND_57fqf0l740hpnbxev71c6d56e}"
+"2024-06-25T21:18:25.022Z","""i-044ba56b80a95063d""","""cash-commerce-browser""","""[{\""name\"":\""Xbox $100 Gift Card - [Digital]\"",\""url\"":\""https://www.walmart.com/ip/seort/48695417\""}]""","Found restricted items: [ShoppingCartProduct{name=Xbox $100 Gift Card - \[Digital\], url=https://www.walmart.com/ip/seort/48695417}]; {customer_token=C_h9ccehmmq, flow_token=ZichXe9bjvCk2NYseQpasHyxW, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_fopbedhc, sup_token=BRAND_4sxo1et5hf8sug8lh500413rj}"
+"2024-06-25T21:15:07.437Z","""i-0ec7dc300989b0f58""","""cash-commerce-browser""","""[{\""name\"":\""$25 Vanilla Visa Shiny Bow Gift Card (plus $3.44 Purchase Fee), quantity 2\"",\""image_url\"":\""https://i5.walmartimages.com/seo/25-Vanilla-Visa-Shiny-Bow-Gift-Card-plus-3-44-Purchase-Fee_852b53b6-7090-407d-a47b-06618b04d244.3c1e4f0aae77e8bf05fdd0660a354377.jpeg?odnHeight=48&odnWidth=48&odnBg=FFFFFF\""}, {\""name\"":\""$50 Vanilla Visa Shiny Bow Gift Card (plus $3.94 Purchase Fee), quantity 2\"",\""image_url\"":\""https://i5.walmartimages.com/seo/50-Vanilla-Visa-Shiny-Bow-Gift-Card-plus-3-94-Purchase-Fee_97c67191-c411-43a8-8174-5bed3ccfa2ed.bfc91450d0b09abcfbab003e6edf9fcf.jpeg?odnHeight=48&odnWidth=48&odnBg=FFFFFF\""}]""","Found restricted items: [ShoppingCartProduct{name=$25 Vanilla Visa Shiny Bow Gift Card (plus $3.44 Purchase Fee)\, quantity 2, image_url=https://i5.walmartimages.com/seo/25-Vanilla-Visa-Shiny-Bow-Gift-Card-plus-3-44-Purchase-Fee_852b53b6-7090-407d-a47b-06618b04d244.3c1e4f0aae77e8bf05fdd0660a354377.jpeg?odnHeight=48&odnWidth=48&odnBg=FFFFFF}, ShoppingCartProduct{name=$50 Vanilla Visa Shiny Bow Gift Card (plus $3.94 Purchase Fee)\, quantity 2, image_url=https://i5.walmartimages.com/seo/50-Vanilla-Visa-Shiny-Bow-Gift-Card-plus-3-94-Purchase-Fee_97c67191-c411-43a8-8174-5bed3ccfa2ed.bfc91450d0b09abcfbab003e6edf9fcf.jpeg?odnHeight=48&odnWidth=48&odnBg=FFFFFF}]; {customer_token=C_gawfj8yrr, flow_token=08e30143-ac4d-428a-963a-f9bbd10cfdb5, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_fopbedhc, sup_token=BRAND_4sxo1et5hf8sug8lh500413rj}"
+"2024-06-25T21:15:01.385Z","""i-003fbf26706b7af3e""","""cash-commerce-browser""","""[{\""name\"":\""$25 Vanilla Visa Shiny Bow Gift Card (plus $3.44 Purchase Fee)\"",\""image_url\"":\""https://i5.walmartimages.com/seo/25-Vanilla-Visa-Shiny-Bow-Gift-Card-plus-3-44-Purchase-Fee_852b53b6-7090-407d-a47b-06618b04d244.3c1e4f0aae77e8bf05fdd0660a354377.jpeg?odnHeight=48&odnWidth=48&odnBg=FFFFFF\""}, {\""name\"":\""$50 Vanilla Visa Shiny Bow Gift Card (plus $3.94 Purchase Fee), quantity 2\"",\""image_url\"":\""https://i5.walmartimages.com/seo/50-Vanilla-Visa-Shiny-Bow-Gift-Card-plus-3-94-Purchase-Fee_97c67191-c411-43a8-8174-5bed3ccfa2ed.bfc91450d0b09abcfbab003e6edf9fcf.jpeg?odnHeight=48&odnWidth=48&odnBg=FFFFFF\""}]""","Found restricted items: [ShoppingCartProduct{name=$25 Vanilla Visa Shiny Bow Gift Card (plus $3.44 Purchase Fee), image_url=https://i5.walmartimages.com/seo/25-Vanilla-Visa-Shiny-Bow-Gift-Card-plus-3-44-Purchase-Fee_852b53b6-7090-407d-a47b-06618b04d244.3c1e4f0aae77e8bf05fdd0660a354377.jpeg?odnHeight=48&odnWidth=48&odnBg=FFFFFF}, ShoppingCartProduct{name=$50 Vanilla Visa Shiny Bow Gift Card (plus $3.94 Purchase Fee)\, quantity 2, image_url=https://i5.walmartimages.com/seo/50-Vanilla-Visa-Shiny-Bow-Gift-Card-plus-3-94-Purchase-Fee_97c67191-c411-43a8-8174-5bed3ccfa2ed.bfc91450d0b09abcfbab003e6edf9fcf.jpeg?odnHeight=48&odnWidth=48&odnBg=FFFFFF}]; {customer_token=C_gawfj8yrr, flow_token=08e30143-ac4d-428a-963a-f9bbd10cfdb5, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_fopbedhc, sup_token=BRAND_4sxo1et5hf8sug8lh500413rj}"
+"2024-06-25T21:14:58.239Z","""i-003fbf26706b7af3e""","""cash-commerce-browser""","""[{\""name\"":\""$25 Vanilla Visa Shiny Bow Gift Card (plus $3.44 Purchase Fee)\"",\""image_url\"":\""https://i5.walmartimages.com/seo/25-Vanilla-Visa-Shiny-Bow-Gift-Card-plus-3-44-Purchase-Fee_852b53b6-7090-407d-a47b-06618b04d244.3c1e4f0aae77e8bf05fdd0660a354377.jpeg?odnHeight=48&odnWidth=48&odnBg=FFFFFF\""}, {\""name\"":\""$50 Vanilla Visa Shiny Bow Gift Card (plus $3.94 Purchase Fee)\"",\""image_url\"":\""https://i5.walmartimages.com/seo/50-Vanilla-Visa-Shiny-Bow-Gift-Card-plus-3-94-Purchase-Fee_97c67191-c411-43a8-8174-5bed3ccfa2ed.bfc91450d0b09abcfbab003e6edf9fcf.jpeg?odnHeight=48&odnWidth=48&odnBg=FFFFFF\""}]""","Found restricted items: [ShoppingCartProduct{name=$25 Vanilla Visa Shiny Bow Gift Card (plus $3.44 Purchase Fee), image_url=https://i5.walmartimages.com/seo/25-Vanilla-Visa-Shiny-Bow-Gift-Card-plus-3-44-Purchase-Fee_852b53b6-7090-407d-a47b-06618b04d244.3c1e4f0aae77e8bf05fdd0660a354377.jpeg?odnHeight=48&odnWidth=48&odnBg=FFFFFF}, ShoppingCartProduct{name=$50 Vanilla Visa Shiny Bow Gift Card (plus $3.94 Purchase Fee), image_url=https://i5.walmartimages.com/seo/50-Vanilla-Visa-Shiny-Bow-Gift-Card-plus-3-94-Purchase-Fee_97c67191-c411-43a8-8174-5bed3ccfa2ed.bfc91450d0b09abcfbab003e6edf9fcf.jpeg?odnHeight=48&odnWidth=48&odnBg=FFFFFF}]; {customer_token=C_gawfj8yrr, flow_token=08e30143-ac4d-428a-963a-f9bbd10cfdb5, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_fopbedhc, sup_token=BRAND_4sxo1et5hf8sug8lh500413rj}"
+"2024-06-25T21:13:44.815Z","""i-0c92416a905681db6""","""cash-commerce-browser""","""[{\""name\"":\""Xbox $100 Gift Card - [Digital]\"",\""url\"":\""https://www.walmart.com/ip/seort/48695417\""}]""","Found restricted items: [ShoppingCartProduct{name=Xbox $100 Gift Card - \[Digital\], url=https://www.walmart.com/ip/seort/48695417}]; {customer_token=C_h9ccehmmq, flow_token=W9fV7QLzgEeyuDVEf2101TEnm, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_fopbedhc, sup_token=BRAND_4sxo1et5hf8sug8lh500413rj}"
+"2024-06-25T21:13:36.779Z","""i-0ec7dc300989b0f58""","""cash-commerce-browser""","""[{\""name\"":\""Home Depot Birthday Cupcake eGift\"",\""url\"":\""https://www.homedepot.com/gift-cards/p/Home-Depot-Birthday-Cupcake-eGift/53L1QLAXWKVF06WJ8DLHBYT6S8/5J0QS6XYWSCJ3H9HG3F6YSRL2C\"",\""image_url\"":\""https://images.thdstatic.com/giftcards/catalog/53L1QLAXWKVF06WJ8DLHBYT6S8/xxlarge/5J0QS6XYWSCJ3H9HG3F6YSRL2C_0608202315:26:21.PNG\""}]""","Found restricted items: [ShoppingCartProduct{name=Home Depot Birthday Cupcake eGift, url=https://www.homedepot.com/gift-cards/p/Home-Depot-Birthday-Cupcake-eGift/53L1QLAXWKVF06WJ8DLHBYT6S8/5J0QS6XYWSCJ3H9HG3F6YSRL2C, image_url=https://images.thdstatic.com/giftcards/catalog/53L1QLAXWKVF06WJ8DLHBYT6S8/xxlarge/5J0QS6XYWSCJ3H9HG3F6YSRL2C_0608202315:26:21.PNG}]; {customer_token=C_ssd2wpm6w, flow_token=V53QOpItzVeaNwXwPC6cewpnt, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_kiqt4xl6, sup_token=BRAND_78qp2zjkqjto3v3xxr5ibueyv}"
+"2024-06-25T21:12:33.161Z","""i-003fbf26706b7af3e""","""cash-commerce-browser""","""[{\""name\"":\""$100 Vanilla® Visa® Gift Box Gift Card (plus $5.44 Purchase Fee)\"",\""url\"":\""https://www.walmart.com/ip/seort/55844027\"",\""image_url\"":\""https://i5.walmartimages.com/seo/100-Vanilla-Visa-Gift-Box-Gift-Card-plus-5-44-Purchase-Fee_d1b5b130-1ae0-4690-a1ee-4e06ff34661a.1e16d3785d608d00f8636a7a91f40182.jpeg?odnHeight=72&odnWidth=72&odnBg=FFFFFF\""}, {\""name\"":\""$50 Vanilla Visa Shiny Bow Gift Card (plus $3.94 Purchase Fee)\"",\""url\"":\""https://www.walmart.com/ip/seort/1529427774\"",\""image_url\"":\""https://i5.walmartimages.com/seo/50-Vanilla-Visa-Shiny-Bow-Gift-Card-plus-3-94-Purchase-Fee_97c67191-c411-43a8-8174-5bed3ccfa2ed.bfc91450d0b09abcfbab003e6edf9fcf.jpeg?odnHeight=72&odnWidth=72&odnBg=FFFFFF\""}, {\""name\"":\""$25 Vanilla Visa Shiny Bow Gift Card (plus $3.44 Purchase Fee)\"",\""url\"":\""https://www.walmart.com/ip/seort/1314241502\"",\""image_url\"":\""https://i5.walmartimages.com/seo/25-Vanilla-Visa-Shiny-Bow-Gift-Card-plus-3-44-Purchase-Fee_852b53b6-7090-407d-a47b-06618b04d244.3c1e4f0aae77e8bf05fdd0660a354377.jpeg?odnHeight=72&odnWidth=72&odnBg=FFFFFF\""}]""","Found restricted items: [ShoppingCartProduct{name=$100 Vanilla® Visa® Gift Box Gift Card (plus $5.44 Purchase Fee), url=https://www.walmart.com/ip/seort/55844027, image_url=https://i5.walmartimages.com/seo/100-Vanilla-Visa-Gift-Box-Gift-Card-plus-5-44-Purchase-Fee_d1b5b130-1ae0-4690-a1ee-4e06ff34661a.1e16d3785d608d00f8636a7a91f40182.jpeg?odnHeight=72&odnWidth=72&odnBg=FFFFFF}, ShoppingCartProduct{name=$50 Vanilla Visa Shiny Bow Gift Card (plus $3.94 Purchase Fee), url=https://www.walmart.com/ip/seort/1529427774, image_url=https://i5.walmartimages.com/seo/50-Vanilla-Visa-Shiny-Bow-Gift-Card-plus-3-94-Purchase-Fee_97c67191-c411-43a8-8174-5bed3ccfa2ed.bfc91450d0b09abcfbab003e6edf9fcf.jpeg?odnHeight=72&odnWidth=72&odnBg=FFFFFF}, ShoppingCartProduct{name=$25 Vanilla Visa Shiny Bow Gift Card (plus $3.44 Purchase Fee), url=https://www.walmart.com/ip/seort/1314241502, image_url=https://i5.walmartimages.com/seo/25-Vanilla-Visa-Shiny-Bow-Gift-Card-plus-3-44-Purchase-Fee_852b53b6-7090-407d-a47b-06618b04d244.3c1e4f0aae77e8bf05fdd0660a354377.jpeg?odnHeight=72&odnWidth=72&odnBg=FFFFFF}]; {customer_token=C_gawfj8yrr, flow_token=08e30143-ac4d-428a-963a-f9bbd10cfdb5, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_fopbedhc, sup_token=BRAND_4sxo1et5hf8sug8lh500413rj}"
+"2024-06-25T21:09:57.657Z","""i-003fbf26706b7af3e""","""cash-commerce-browser""","""[{\""name\"":\""Birthday Celebration Target Giftcard\"",\""url\"":\""https://www.target.com/p/birthday-celebration-target-giftcard--no-aasa/-/A-81480862\"",\""image_url\"":\""https://target.scene7.com/is/image/Target//GUEST_c2ea8817-d6da-437c-9d85-c39b2e96953d?qlt=80&fmt=pjpeg\""}]""","Found restricted items: [ShoppingCartProduct{name=Birthday Celebration Target Giftcard, url=https://www.target.com/p/birthday-celebration-target-giftcard--no-aasa/-/A-81480862, image_url=https://target.scene7.com/is/image/Target//GUEST_c2ea8817-d6da-437c-9d85-c39b2e96953d?qlt=80&fmt=pjpeg}]; {customer_token=C_8xdwa1m6n, flow_token=9RBhrYZk8g7j2WDYyM3GbfdES, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_3mfvyukp, sup_token=BRAND_57fqf0l740hpnbxev71c6d56e}"
+"2024-06-25T21:09:49.051Z","""i-003fbf26706b7af3e""","""cash-commerce-browser""","""[{\""name\"":\""Birthday Celebration Target Giftcard\"",\""url\"":\""https://www.target.com/p/birthday-celebration-target-giftcard--no-aasa/-/A-81480862\"",\""image_url\"":\""https://target.scene7.com/is/image/Target//GUEST_c2ea8817-d6da-437c-9d85-c39b2e96953d?qlt=80&fmt=pjpeg\""}]""","Found restricted items: [ShoppingCartProduct{name=Birthday Celebration Target Giftcard, url=https://www.target.com/p/birthday-celebration-target-giftcard--no-aasa/-/A-81480862, image_url=https://target.scene7.com/is/image/Target//GUEST_c2ea8817-d6da-437c-9d85-c39b2e96953d?qlt=80&fmt=pjpeg}]; {customer_token=C_8xdwa1m6n, flow_token=J7liXzVSAPSDIaK4CWVLC7bBi, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_3mfvyukp, sup_token=BRAND_57fqf0l740hpnbxev71c6d56e}"
+"2024-06-25T21:09:35.160Z","""i-003fbf26706b7af3e""","""cash-commerce-browser""","""[{\""name\"":\""Promotional Email GiftCard $10\"",\""image_url\"":\""https://target.scene7.com/is/image/Target//GUEST_337c880d-e2c0-4794-9784-962c63cf0646?qlt=80&fmt=pjpeg\""}]""","Found restricted items: [ShoppingCartProduct{name=Promotional Email GiftCard $10, image_url=https://target.scene7.com/is/image/Target//GUEST_337c880d-e2c0-4794-9784-962c63cf0646?qlt=80&fmt=pjpeg}]; {customer_token=C_fsexdpmbh, flow_token=jVDz28RrrN9tAkim71XsZGZMD, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_3mfvyukp, sup_token=BRAND_57fqf0l740hpnbxev71c6d56e}"
+"2024-06-25T21:09:35.022Z","""i-0ec7dc300989b0f58""","""cash-commerce-browser""","""[{\""name\"":\""Birthday Celebration Target Giftcard\"",\""url\"":\""https://www.target.com/p/birthday-celebration-target-giftcard--no-aasa/-/A-81480862\"",\""image_url\"":\""\""}]""","Found restricted items: [ShoppingCartProduct{name=Birthday Celebration Target Giftcard, url=https://www.target.com/p/birthday-celebration-target-giftcard--no-aasa/-/A-81480862, image_url=}]; {customer_token=C_8xdwa1m6n, flow_token=9pZoMnaqRmVXuWQ9P9pSLefbO, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_3mfvyukp, sup_token=BRAND_57fqf0l740hpnbxev71c6d56e}"
+"2024-06-25T21:09:07.723Z","""i-044ba56b80a95063d""","""cash-commerce-browser""","""[{\""name\"":\""Nintendo Switch Family Online 12 Month Gift Card [Digital]\"",\""url\"":\""https://www.walmart.com/ip/seort/875981752\""}]""","Found restricted items: [ShoppingCartProduct{name=Nintendo Switch Family Online 12 Month Gift Card \[Digital\], url=https://www.walmart.com/ip/seort/875981752}]; {customer_token=C_gq2ekammv, flow_token=03c0b437-4fc5-4967-9692-3efddd21a0d6, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_fopbedhc, sup_token=BRAND_4sxo1et5hf8sug8lh500413rj}"
+"2024-06-25T21:08:46.049Z","""i-044ba56b80a95063d""","""cash-commerce-browser""","""[{\""name\"":\""Promotional Email GiftCard $10\"",\""image_url\"":\""https://target.scene7.com/is/image/Target//GUEST_337c880d-e2c0-4794-9784-962c63cf0646?qlt=80&fmt=pjpeg\""}]""","Found restricted items: [ShoppingCartProduct{name=Promotional Email GiftCard $10, image_url=https://target.scene7.com/is/image/Target//GUEST_337c880d-e2c0-4794-9784-962c63cf0646?qlt=80&fmt=pjpeg}]; {customer_token=C_fsexdpmbh, flow_token=e5xlrZ3F45YirfY5AHNFlAvtV, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_3mfvyukp, sup_token=BRAND_57fqf0l740hpnbxev71c6d56e}"
+"2024-06-25T20:55:28.994Z","""i-044ba56b80a95063d""","""cash-commerce-browser""","""[{\""name\"":\""$50 Vanilla® Mastercard® Celebration Dots Gift Card (plus $3.94 Purchase Fee)\"",\""image_url\"":\""https://i5.walmartimages.com/seo/50-Vanilla-Mastercard-Celebration-Dots-Gift-Card-plus-3-94-Purchase-Fee_04e596b9-6d0a-41c9-96ed-7bdf11468cd2.d9453dcf7909a341bf31ee4dce04c833.jpeg?odnHeight=48&odnWidth=48&odnBg=FFFFFF\""}]""","Found restricted items: [ShoppingCartProduct{name=$50 Vanilla® Mastercard® Celebration Dots Gift Card (plus $3.94 Purchase Fee), image_url=https://i5.walmartimages.com/seo/50-Vanilla-Mastercard-Celebration-Dots-Gift-Card-plus-3-94-Purchase-Fee_04e596b9-6d0a-41c9-96ed-7bdf11468cd2.d9453dcf7909a341bf31ee4dce04c833.jpeg?odnHeight=48&odnWidth=48&odnBg=FFFFFF}]; {customer_token=C_nywqkaygt, flow_token=IOAzdzUmK2FLIN1KiCt8xQZQd, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_fopbedhc, sup_token=BRAND_4sxo1et5hf8sug8lh500413rj}"
+"2024-06-25T20:55:26.016Z","""i-0c92416a905681db6""","""cash-commerce-browser""","""[{\""name\"":\""$50 Vanilla® Mastercard® Celebration Dots Gift Card (plus $3.94 Purchase Fee)\"",\""url\"":\""https://www.walmart.com/ip/seort/46518449\""}]""","Found restricted items: [ShoppingCartProduct{name=$50 Vanilla® Mastercard® Celebration Dots Gift Card (plus $3.94 Purchase Fee), url=https://www.walmart.com/ip/seort/46518449}]; {customer_token=C_nywqkaygt, flow_token=dEIHJrVnTIH6xsmTYspEfBcle, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_fopbedhc, sup_token=BRAND_4sxo1et5hf8sug8lh500413rj}"
+"2024-06-25T20:55:13.996Z","""i-0c92416a905681db6""","""cash-commerce-browser""","""[{\""name\"":\""$50 Vanilla® Mastercard® Celebration Dots Gift Card (plus $3.94 Purchase Fee)\"",\""url\"":\""https://www.walmart.com/ip/seort/46518449\""}, {\""name\"":\""$50 Vanilla® Visa® eGift Card (plus $3.94 Purchase Fee)\"",\""url\"":\""https://www.walmart.com/ip/seort/761861368\""}]""","Found restricted items: [ShoppingCartProduct{name=$50 Vanilla® Mastercard® Celebration Dots Gift Card (plus $3.94 Purchase Fee), url=https://www.walmart.com/ip/seort/46518449}, ShoppingCartProduct{name=$50 Vanilla® Visa® eGift Card (plus $3.94 Purchase Fee), url=https://www.walmart.com/ip/seort/761861368}]; {customer_token=C_nywqkaygt, flow_token=MJgyeF7Eew00jGOPMLCG2rHyD, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_fopbedhc, sup_token=BRAND_4sxo1et5hf8sug8lh500413rj}"
+"2024-06-25T20:52:39.628Z","""i-0c92416a905681db6""","""cash-commerce-browser""","""[{\""name\"":\""$50 Vanilla® Visa® eGift Card (plus $3.94 Purchase Fee)\"",\""url\"":\""https://www.walmart.com/ip/seort/761861368\""}]""","Found restricted items: [ShoppingCartProduct{name=$50 Vanilla® Visa® eGift Card (plus $3.94 Purchase Fee), url=https://www.walmart.com/ip/seort/761861368}]; {customer_token=C_nywqkaygt, flow_token=WJm0vCQG5TcMjVjTvpMqrfOGs, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_fopbedhc, sup_token=BRAND_4sxo1et5hf8sug8lh500413rj}"
+"2024-06-25T20:52:36.059Z","""i-0ec7dc300989b0f58""","""cash-commerce-browser""","""[{\""name\"":\""$50 Vanilla Visa Shiny Bow Gift Card (plus $3.94 Purchase Fee)\"",\""url\"":\""https://www.walmart.com/ip/seort/1529427774\""}, {\""name\"":\""$50 Vanilla® Visa® eGift Card (plus $3.94 Purchase Fee)\"",\""url\"":\""https://www.walmart.com/ip/seort/761861368\""}]""","Found restricted items: [ShoppingCartProduct{name=$50 Vanilla Visa Shiny Bow Gift Card (plus $3.94 Purchase Fee), url=https://www.walmart.com/ip/seort/1529427774}, ShoppingCartProduct{name=$50 Vanilla® Visa® eGift Card (plus $3.94 Purchase Fee), url=https://www.walmart.com/ip/seort/761861368}]; {customer_token=C_nywqkaygt, flow_token=a5YVHg3kkZn8fb3e3TT1b9GlZ, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_fopbedhc, sup_token=BRAND_4sxo1et5hf8sug8lh500413rj}"
+"2024-06-25T20:52:02.686Z","""i-0ec7dc300989b0f58""","""cash-commerce-browser""","""[{\""name\"":\""Delta Airlines Wedding $250 Gift Card (Email Delivery)\"",\""url\"":\""https://www.target.com/p/delta-airlines-wedding--250-gift-card--email-delivery---no-aasa/-/A-89537719\"",\""image_url\"":\""https://target.scene7.com/is/image/Target//GUEST_0de9a4fb-7f0e-4d85-806f-e841d7664a59?qlt=80&fmt=pjpeg\""}]""","Found restricted items: [ShoppingCartProduct{name=Delta Airlines Wedding $250 Gift Card (Email Delivery), url=https://www.target.com/p/delta-airlines-wedding--250-gift-card--email-delivery---no-aasa/-/A-89537719, image_url=https://target.scene7.com/is/image/Target//GUEST_0de9a4fb-7f0e-4d85-806f-e841d7664a59?qlt=80&fmt=pjpeg}]; {customer_token=C_80exppmwa, flow_token=pnOyvjetUVkuZfhIOQwOQCINH, payment_method=PAYMENT_METHOD_SINGLE_USE_PAYMENT, merchant_token=M_3mfvyukp, sup_token=BRAND_57fqf0l740hpnbxev71c6d56e}"

genai_SDK/Seq2Seq.py

@@ -0,0 +1,250 @@
+#!/usr/bin/env python
+# coding: utf-8
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, random_split
+import pandas as pd
+from tqdm import tqdm
+import time
+from .Utilities import LanguageDataset
+
+class Seq2Seq():
+    """
+    Base class for Seq2Seq (text-generation models). This class will be inherited by wrappers of transformers like GPT2
+    and T5.
+
+    Attributes:
+
+    Methods:
+
+    """
+
+    def __init__(self, gpu=0, max_length=0, model_path=None):
+
+        # Load Seq2Seq to device based on available hardware
+        if torch.cuda.is_available():
+            self.device = torch.device('cuda')
+        else:
+            try:
+                self.device = torch.device('mps') # Apple Silicon
+            except Exception:
+                self.device = torch.device('cpu')
+
+        # GPU that model will run on
+        self.gpu = gpu
+
+        # Model specs
+        if model_path: self.model = torch.load(model_path).to(self.device)
+        else: self.model = None
+        self.model_name = ""
+        self.tokenizer = None
+        self.max_length = max_length
+
+        # Training specs
+        self.train_loader = None
+        self.valid_loader = None
+        self.results = pd.DataFrame(columns=['epoch', 'model_arch', 'batch_size', 'gpu', 'training_loss', 'validation_loss', 'epoch_duration_sec'])
+
+    def load_data(self, df, batch_size, train_ratio=0.8):
+        self.batch_size = batch_size
+        dataset = LanguageDataset(df, self.tokenizer)
+        train_size = int(0.8*len(dataset))
+        valid_size = len(dataset) - train_size
+        train_data, valid_data = random_split(dataset, [train_size, valid_size])
+        self.max_length = dataset.max_length
+        self.train_loader = DataLoader(train_data, batch_size=self.batch_size, shuffle=True)
+        self.valid_loader = DataLoader(valid_data, batch_size=self.batch_size)
+
+    """ Return training results """
+    def summary(self):
+        return self.results
+
+    """ Save model to path """
+    def to_pt(self, path):
+        torch.save(self.model, path)
+
+
+class GPT2(Seq2Seq):
+    """
+    This is the GPT2 implementation of Seq2Seq.
+    """
+
+    def __init__(self, gpu, model_name, batch_size=16):
+        super().__init__(gpu, max_length=0)
+        from transformers import GPT2Tokenizer, GPT2LMHeadModel
+        self.model_name = model_name
+        self.model = GPT2LMHeadModel.from_pretrained(self.model_name).to(self.device)
+        self.tokenizer = GPT2Tokenizer.from_pretrained(self.model_name)
+        self.tokenizer.pad_token = self.tokenizer.eos_token
+
+    def train(self, num_epochs=3, train_ratio=0.8):
+        criterion = nn.CrossEntropyLoss(ignore_index=self.tokenizer.pad_token_id)
+        optimizer = optim.Adam(self.model.parameters(), lr=5e-4)
+
+        # Init a results dataframe
+
+        results = pd.DataFrame(columns=['epoch', 'transformer', 'batch_size', 'gpu',
+                                        'training_loss', 'validation_loss', 'epoch_duration_sec'])
+        # The training loop
+        for epoch in range(num_epochs):
+            start_time = time.time()  # Start the timer for the epoch
+
+            # Training
+            ## This line tells the self.model we're in 'learning mode'
+            self.model.train()
+            epoch_training_loss = 0
+            train_iterator = tqdm(self.train_loader,
+                                  desc=f"Training Epoch {epoch + 1}/{num_epochs} Batch Size: {self.batch_size}, Transformer: {self.model_name}")
+            for batch in train_iterator:
+                optimizer.zero_grad()
+                inputs = batch['input_ids'].squeeze(1).to(self.device)
+                targets = inputs.clone()
+                outputs = self.model(input_ids=inputs, labels=targets)
+                loss = outputs.loss
+                loss.backward()
+                optimizer.step()
+                train_iterator.set_postfix({'Training Loss': loss.item()})
+                epoch_training_loss += loss.item()
+            avg_epoch_training_loss = epoch_training_loss / len(train_iterator)
+
+            # Validation
+            ## This line below tells the self.model to 'stop learning'
+            self.model.eval()
+            epoch_validation_loss = 0
+            total_loss = 0
+            valid_iterator = tqdm(self.valid_loader, desc=f"Validation Epoch {epoch + 1}/{num_epochs}")
+            with torch.no_grad():
+                for batch in valid_iterator:
+                    inputs = batch['input_ids'].squeeze(1).to(self.device)
+                    targets = inputs.clone()
+                    outputs = self.model(input_ids=inputs, labels=targets)
+                    loss = outputs.loss
+                    total_loss += loss
+                    valid_iterator.set_postfix({'Validation Loss': loss.item()})
+                    epoch_validation_loss += loss.item()
+
+            avg_epoch_validation_loss = epoch_validation_loss / len(self.valid_loader)
+
+            end_time = time.time()  # End the timer for the epoch
+            epoch_duration_sec = end_time - start_time  # Calculate the duration in seconds
+
+            new_row = {'transformer': self.model_name,
+                       'batch_size': self.batch_size,
+                       'gpu': self.gpu,
+                       'epoch': epoch + 1,
+                       'training_loss': avg_epoch_training_loss,
+                       'validation_loss': avg_epoch_validation_loss,
+                       'epoch_duration_sec': epoch_duration_sec}  # Add epoch_duration to the dataframe
+
+            self.results.loc[len(self.results)] = new_row
+            print(f"Epoch: {epoch + 1}, Validation Loss: {total_loss / len(self.valid_loader)}")
+
+    def generate_text(self, input_str, top_k=16, top_p=0.95, temperature=1.0, repetition_penalty=1.2):
+        # Encode string to tokens
+        input_ids= self.tokenizer.encode(input_str, return_tensors='pt').to(self.device)
+
+        # Feed tokens to model and get outcome tokens
+        output = self.model.generate(
+            input_ids,
+            max_length=self.max_length,
+            num_return_sequences=1,
+            do_sample=True,
+            top_k=top_k,
+            top_p=top_p,
+            temperature=temperature,
+            repetition_penalty=repetition_penalty
+        )
+
+        # Decode tokens to string
+        decoded_output = self.tokenizer.decode(output[0], skip_special_tokens=True)
+        return decoded_output
+
+class FlanT5(Seq2Seq):
+    """
+    This is the T5 implementation of Seq2Seq - it is designed to support T5 models of various sizes.
+    """
+    def __init__(self, gpu, model_name, batch_size=16):
+        super().__init__(gpu, max_length=0)
+        from transformers import T5ForConditionalGeneration, T5Tokenizer
+        self.model_name = model_name
+        self.model = T5ForConditionalGeneration.from_pretrained(self.model_name).to(self.device)
+        self.tokenizer = T5Tokenizer.from_pretrained(self.model_name)
+        self.tokenizer.pad_token = self.tokenizer.eos_token
+
+    def train(self, num_epochs=3, train_ratio=0.8):
+        criterion = nn.CrossEntropyLoss(ignore_index=self.tokenizer.pad_token_id)
+        optimizer = optim.Adam(self.model.parameters(), lr=5e-4)
+
+        # Init a results dataframe
+
+        self.results = pd.DataFrame(columns=['epoch', 'transformer', 'batch_size', 'gpu',
+                                        'training_loss', 'validation_loss', 'epoch_duration_sec'])
+        # The training loop
+        for epoch in range(num_epochs):
+            start_time = time.time()  # Start the timer for the epoch
+
+            # Training
+            ## This line tells the model we're in 'learning mode'
+            self.model.train()
+            epoch_training_loss = 0
+            train_iterator = tqdm(self.train_loader,
+                                  desc=f"Training Epoch {epoch + 1}/{num_epochs} Batch Size: {self.batch_size}, Transformer: {self.model_name}")
+            for batch in train_iterator:
+                optimizer.zero_grad()
+                inputs = batch['input_ids'].squeeze(1).to(self.device)
+                targets = batch['labels'].squeeze(1).to(self.device)
+                outputs = self.model(input_ids=inputs, labels=targets)
+                loss = outputs.loss
+                loss.backward()
+                optimizer.step()
+                train_iterator.set_postfix({'Training Loss': loss.item()})
+                epoch_training_loss += loss.item()
+            avg_epoch_training_loss = epoch_training_loss / len(train_iterator)
+
+            # Validation
+            ## This line below tells the model to 'stop learning'
+            self.model.eval()
+            epoch_validation_loss = 0
+            total_loss = 0
+            valid_iterator = tqdm(self.valid_loader, desc=f"Validation Epoch {epoch + 1}/{num_epochs}")
+            with torch.no_grad():
+                for batch in valid_iterator:
+                    inputs = batch['input_ids'].squeeze(1).to(self.device)
+                    targets = batch['labels'].squeeze(1).to(self.device)
+                    outputs = self.model(input_ids=inputs, labels=targets)
+                    loss = outputs.loss
+                    total_loss += loss
+                    valid_iterator.set_postfix({'Validation Loss': loss.item()})
+                    epoch_validation_loss += loss.item()
+
+            avg_epoch_validation_loss = epoch_validation_loss / len(self.valid_loader)
+
+            end_time = time.time()  # End the timer for the epoch
+            epoch_duration_sec = end_time - start_time  # Calculate the duration in seconds
+
+            new_row = {'transformer': self.model_name,
+                       'batch_size': self.batch_size,
+                       'gpu': self.gpu,
+                       'epoch': epoch + 1,
+                       'training_loss': avg_epoch_training_loss,
+                       'validation_loss': avg_epoch_validation_loss,
+                       'epoch_duration_sec': epoch_duration_sec}  # Add epoch_duration to the dataframe
+
+            self.results.loc[len(self.results)] = new_row
+            print(f"Epoch: {epoch + 1}, Validation Loss: {total_loss / len(self.valid_loader)}")
+
+    def generate_text(self, input_str, top_k=16, top_p=0.95, temperature=1.0, repetition_penalty=1.2):
+        # Encode input string into tensors via the FlanT5 tokenizer
+        input_ids = self.tokenizer.encode(input_str, return_tensors='pt', max_length=self.max_length, truncation=True).to(self.device)
+        # Run tensors through model to get output tensor values
+        output_ids = self.model.generate(input_ids,
+            max_length=self.max_length,
+            do_sample=True,
+            top_k=top_k,
+            top_p=top_p,
+            temperature=temperature,
+            repetition_penalty=repetition_penalty)
+        # Decode output tensors to text vi
+        output_str = self.tokenizer.decode(output_ids[0], skip_special_tokens=True)
+        return output_str

genai_SDK/Utilities.py

@@ -0,0 +1,563 @@
+#!/usr/bin/env python
+# coding: utf-8
+import pandas as pd
+from torch.utils.data import Dataset
+
+class LanguageDataset(Dataset):
+    def __init__(self, df, tokenizer):
+        # Make sure data is compatible
+        if len(df.columns) !=2:
+            raise Exception("Dataset can only have two columns!")
+
+        self.data = df.to_dict(orient='records')
+        self.tokenizer = tokenizer
+
+        # set the length of smallest square needed
+        self.max_length = smallest_square_length(df)
+        self.labels = df.columns
+
+    def __len__(self):
+        return len(self.data)
+    def __getitem__(self, i):
+        X = self.data[i][self.labels[0]]
+        Y = self.data[i][self.labels[1]]
+        if str(type(self.tokenizer)) == "<class 'transformers.models.gpt2.tokenization_gpt2.GPT2Tokenizer'>":
+            return self.tokenizer.encode_plus(X + ' | ' + Y,
+                                                return_tensors='pt',
+                                                max_length = self.max_length,
+                                                padding='max_length',
+                                                truncation=True)
+        elif str(type(self.tokenizer)) == "<class 'transformers.models.t5.tokenization_t5.T5Tokenizer'>":
+            input_tokens = self.tokenizer.encode_plus(
+                X,
+                max_length=self.max_length,
+                padding='max_length',
+                truncation=True,
+                return_tensors='pt'
+            )
+            target_tokens = self.tokenizer.encode_plus(
+                Y,
+                max_length=self.max_length,
+                padding='max_length',
+                truncation=True,
+                return_tensors='pt'
+            )
+            return {
+                'input_ids': input_tokens['input_ids'].squeeze(),
+                # 'attention_mask': input_tokens['attention_mask'].squeeze(),
+                'labels': target_tokens['input_ids'].squeeze()
+            }
+
+
+
+def smallest_square_length(df):
+    col1 = df[df.columns[0]].astype(str).apply(lambda x: len(x)).max()
+    col2 = df[df.columns[1]].astype(str).apply(lambda x: len(x)).max()
+
+    max_length = max(col1, col2)
+
+    x = 2
+    while x < max_length:
+        x = x * 2
+
+    return x
+
+def levenshtein_distance(str1, str2):
+    """
+    Computes the Levenshtein distance between two strings.
+    Parameters:
+        str1 (str): The first string.
+        str2 (str): The second string.
+    Returns:
+        int: The Levenshtein distance between the two strings.
+    """
+    m, n = len(str1), len(str2)
+    dp = [[0] * (n + 1) for _ in range(m + 1)]
+
+    for i in range(m + 1):
+        dp[i][0] = i
+
+    for j in range(n + 1):
+        dp[0][j] = j
+
+    for i in range(1, m + 1):
+        for j in range(1, n + 1):
+            if str1[i - 1] == str2[j - 1]:
+                dp[i][j] = dp[i - 1][j - 1]
+            else:
+                dp[i][j] = 1 + min(dp[i - 1][j], dp[i][j - 1], dp[i - 1][j - 1])
+
+    return dp[m][n]
+
+def grid_search(model, tokenizer, input_str, topK_values, topP_values, temperature_values, repetition_penalty_values, expected_output):
+    """
+     Conducts a grid search over specified hyperparameters to find the best text generation settings (GPT series).
+     Parameters:
+     - model: The pre-trained model used for text generation.
+     - tokenizer: The tokenizer associated with the model.
+     - input_str: The input string to the model for text generation.
+     - topK_values: A list of integer values for the topK sampling hyperparameter.
+     - topP_values: A list of float values for the topP (nucleus) sampling hyperparameter.
+     - temperature_values: A list of float values for the temperature setting of the model.
+     - repetition_penalty_values: A list of float values for penalizing repetitions in the generated text.
+     - expected_output: The expected output string against which generated texts are evaluated using the Levenshtein distance.
+     Returns:
+     - results: A pandas DataFrame containing the combination of hyperparameters, the generated output for each combination, and its Levenshtein distance from the expected output.
+     Notes:
+     - The function prints out the best hyperparameters found during the search, based on the smallest Levenshtein distance.
+     - Levenshtein distance measures the number of edits required to transform one string into another.
+     """
+    results = pd.DataFrame(columns=['topK', 'topP', 'temperature', 'repetition_penalty', 'generated_output', 'levenshtein_distance'])
+    min_distance = 9999999
+    for topK in topK_values:
+        for topP in topP_values:
+            for temperature in temperature_values:
+                for repetition_penalty in repetition_penalty_values:
+                    # try:
+                    generated_output = model.generate_text(input_str, topK, topP, temperature, repetition_penalty)
+                    # print(generated_output)
+                    distance = levenshtein_distance(generated_output, expected_output)
+                    if distance < min_distance:
+                        print(f'topK={topK}, topP={topP}, temperature={temperature}, repetition_penalty={repetition_penalty}, levenshtein_distance={distance}')
+                        min_distance = distance
+
+                    new_row = {'topK': topK,
+                               'topP': topP,
+                               'temperature': temperature,
+                               'repetition_penalty': repetition_penalty,
+                               'generated_output': generated_output,
+                               'levenshtein_distance': distance
+                               }
+                    results.loc[len(results)] = new_row
+
+    return results.sort_values(by='levenshtein_distance', ascending=True)
+
+
+def to_coreml(gpt_model, path=''):
+    import torch
+
+    device = torch.device('mps')
+
+    if torch.cuda.is_available():
+        device = torch.device('cuda')
+    else:
+        try:
+            device = torch.device('mps')  # Apple Silicon
+        except Exception:
+            device = torch.device('cpu')
+    if path != '': lm_head_model = torch.load(path, map_location=device)
+    else: lm_head_model = gpt_model.model
+
+    """
+    Recreate the Core ML model from scratch using
+    coremltools' neural_network.NeuralNetworkBuilder
+    """
+    import coremltools
+    import coremltools.models.datatypes as datatypes
+    from coremltools.models import neural_network as neural_network
+    from coremltools.models.utils import save_spec
+    import numpy as np
+    import torch
+    model_name = 'model'
+
+    model = lm_head_model.transformer
+
+    wte = model.wte.weight.data.cpu().numpy().transpose() # shape (768, 50257) /!\ i hate this
+    wpe = model.wpe.weight.data.cpu().numpy().transpose() # shape (768, 1024)
+
+    sequence_length = 128
+    steps = model.config.n_layer
+
+    # build model
+    input_features = [
+        ('input_ids', datatypes.Array(sequence_length)),
+        ('position_ids', datatypes.Array(sequence_length)),
+    ]
+    output_features = [('output_logits', None)]
+
+    builder = neural_network.NeuralNetworkBuilder(
+        input_features,
+        output_features,
+        mode=None,
+        disable_rank5_shape_mapping=True,
+    )
+    builder.add_expand_dims(
+        name='input_ids_expanded_to_rank5',
+        input_name='input_ids',
+        output_name='input_ids_expanded_to_rank5',
+        axes=(1, 2, 3, 4)
+    )
+    builder.add_expand_dims(
+        name='position_ids_expanded_to_rank5',
+        input_name='position_ids',
+        output_name='position_ids_expanded_to_rank5',
+        axes=(1, 2, 3, 4)
+    )
+    builder.add_embedding(
+        name='token_embeddings',
+        input_name='input_ids_expanded_to_rank5',
+        output_name='token_embeddings',
+        W=wte,
+        b=None,
+        input_dim=50257,
+        output_channels=768,
+        has_bias=False,
+    )
+    builder.add_embedding(
+        name='positional_embeddings',
+        input_name='position_ids_expanded_to_rank5',
+        output_name='positional_embeddings',
+        W=wpe,
+        b=None,
+        input_dim=1024,
+        output_channels=768,
+        has_bias=False,
+    )
+
+    # Input:, Output: (seq, 1, 768, 1, 1)
+    builder.add_add_broadcastable(
+        name='embeddings_addition',
+        input_names=['token_embeddings', 'positional_embeddings'],
+        output_name=f'{0}_previous_block'
+    )
+
+    for i in range(steps):
+        print(i)
+        ln_weight = model.h[i].ln_1.weight.data.cpu().numpy().reshape((1, 1, 768, 1, 1))
+        ln_bias = model.h[i].ln_1.bias.data.cpu().numpy().reshape((1, 1, 768, 1, 1))
+        ln_epsilon = model.h[i].ln_1.eps
+
+        builder.add_mvn(
+            name=f"{i}_block_ln_1",
+            input_name=f"{i}_previous_block",
+            # output_name=f"{i}_block_ln_1_output",
+            output_name=f"{i}_block_ln_1",
+            across_channels=True,
+            normalize_variance=True,
+            epsilon=ln_epsilon
+        )
+
+        builder.add_scale(
+            name=f"{i}_block_ln_1_scaled",
+            input_name=f"{i}_block_ln_1",
+            output_name=f"{i}_block_ln_1_scaled",
+            W=ln_weight,
+            b=ln_bias,
+            has_bias=True,
+            shape_scale=[768],
+            shape_bias=[768]
+        )
+
+        builder.add_transpose(
+            name=f"{i}_block_ln_1_reshape",
+            input_name=f"{i}_block_ln_1_scaled",
+            output_name=f"{i}_block_ln_1_scaled_transposed",
+            axes=(1, 0, 2, 3, 4)
+        )
+
+
+        conv_1D_bias = model.h[i].attn.c_attn.bias.data.cpu().numpy().reshape((1, 1, 2304, 1, 1))
+        conv_1D_weights = model.h[i].attn.c_attn.weight.cpu().data.numpy().transpose().reshape((1, 768, 2304, 1, 1))
+
+        builder.add_inner_product(
+            name=f"{i}_block_attn_conv",
+            input_name=f"{i}_block_ln_1_scaled_transposed",
+            output_name=f"{i}_block_attn_conv",
+            input_channels=768,
+            output_channels=2304,
+            W=conv_1D_weights,
+            b=conv_1D_bias,
+            has_bias=True
+        )
+
+        builder.add_split(
+            name=f"{i}_block_attn_qkv_split",
+            input_name=f"{i}_block_attn_conv",
+            output_names=[f"{i}_block_attn_q", f"{i}_block_attn_k", f"{i}_block_attn_v"]
+        )
+
+        builder.add_rank_preserving_reshape(
+            name=f"{i}_block_attn_q_reshape",
+            input_name=f"{i}_block_attn_q",
+            output_name=f"{i}_block_attn_q_reshape",
+            output_shape=(1, 1, sequence_length, 12, 64)
+        )
+
+        builder.add_transpose(
+            name=f"{i}_block_attn_q_reshape_permuted",
+            input_name=f"{i}_block_attn_q_reshape",
+            output_name=f"{i}_block_attn_q_reshape_permuted",
+            axes=(0, 1, 3, 2, 4)
+        )
+
+        builder.add_rank_preserving_reshape(
+            name=f"{i}_block_attn_k_reshape",
+            input_name=f"{i}_block_attn_k",
+            output_name=f"{i}_block_attn_k_reshape",
+            output_shape=(1, 1, sequence_length, 12, 64)
+        )
+
+        builder.add_transpose(
+            name=f"{i}_block_attn_k_reshape_permuted",
+            input_name=f"{i}_block_attn_k_reshape",
+            output_name=f"{i}_block_attn_k_reshape_permuted",
+            axes=(0, 1, 3, 4, 2)
+        )
+
+        builder.add_rank_preserving_reshape(
+            name=f"{i}_block_attn_v_reshape",
+            input_name=f"{i}_block_attn_v",
+            output_name=f"{i}_block_attn_v_reshape",
+            output_shape=(1, 1, sequence_length, 12, 64)
+        )
+
+        builder.add_transpose(
+            name=f"{i}_block_attn_v_reshape_permuted",
+            input_name=f"{i}_block_attn_v_reshape",
+            output_name=f"{i}_block_attn_v_reshape_permuted",
+            axes=(0, 1, 3, 2, 4)
+        )
+
+        builder.add_batched_mat_mul(
+            name=f"{i}_block_attn_qv_matmul",
+            input_names=[f"{i}_block_attn_q_reshape_permuted", f"{i}_block_attn_k_reshape_permuted"],
+            output_name=f"{i}_block_attn_qv_matmul"
+        )
+
+        builder.add_scale(
+            name=f"{i}_block_attn_qv_matmul_scaled",
+            input_name=f"{i}_block_attn_qv_matmul",
+            output_name=f"{i}_block_attn_qv_matmul_scaled",
+            W=np.array(1/8),
+            b=0,
+            has_bias=False
+        )
+
+        bias_0 = model.h[i].attn.bias
+        nd = ns = sequence_length
+        b = (model.h[i].attn.bias[:, :, ns-nd:ns, :ns]).unsqueeze(0)
+
+        builder.add_scale(
+            name=f"{i}_block_attn_bias",
+            input_name=f"{i}_block_attn_qv_matmul_scaled",
+            output_name=f"{i}_block_attn_bias",
+            W=b,
+            b=None,
+            has_bias=False,
+            shape_scale=[1, sequence_length, sequence_length]
+        )
+
+        bias_constant_0 = -1e4 * torch.logical_not(b)
+
+        builder.add_bias(
+            name=f"{i}_block_attn_afterbias",
+            input_name=f"{i}_block_attn_bias",
+            output_name=f"{i}_block_attn_afterbias",
+            # output_name=f"output_logits",
+            b=bias_constant_0,
+            shape_bias=[1, sequence_length, sequence_length],
+        )
+
+        builder.add_squeeze(
+            name=f"{i}_squeezit",
+            input_name=f"{i}_block_attn_afterbias",
+            output_name=f"{i}_squeezit",
+            axes=[0, 1]
+        )
+
+        builder.add_softmax(
+            name=f"{i}_block_attn_softmax",
+            input_name=f"{i}_squeezit",
+            output_name=f"{i}_block_attn_softmax",
+        )
+
+        builder.add_expand_dims(
+            name=f"{i}_expandit",
+            input_name=f"{i}_block_attn_softmax",
+            output_name=f"{i}_expandit",
+            axes=[0, 1]
+        )
+
+        builder.add_batched_mat_mul(
+            name=f"{i}_block_full_attention",
+            input_names=[f"{i}_expandit", f"{i}_block_attn_v_reshape_permuted"],
+            output_name=f"{i}_block_full_attention"
+        )
+
+        builder.add_transpose(
+            name=f"{i}_block_full_attention_merged_t",
+            input_name=f"{i}_block_full_attention",
+            output_name=f"{i}_block_full_attention_merged_t",
+            axes=[0, 1, 3, 2, 4]
+        )
+
+        builder.add_rank_preserving_reshape(
+            name=f"{i}_block_full_attention_merged",
+            input_name=f"{i}_block_full_attention_merged_t",
+            output_name=f"{i}_block_full_attention_merged",
+            output_shape=[1, 1, 1, sequence_length, 768]
+        )
+
+        builder.add_transpose(
+            name=f"{i}_block_attn_conv_proj_t",
+            input_name=f"{i}_block_full_attention_merged",
+            output_name=f"{i}_block_attn_conv_proj_t",
+            axes=[0, 3, 4, 1, 2]
+        )
+
+        conv_1D_proj_bias = model.h[i].attn.c_proj.bias.data.cpu().numpy().reshape((1, 1, 768, 1, 1))
+        conv_1D_proj_weights = model.h[i].attn.c_proj.weight.data.cpu().numpy().transpose().reshape((1, 768, 768, 1, 1))
+
+        # Input:, Output: (1, 3, 768, 1, 1)
+        builder.add_inner_product(
+            name=f"{i}_block_attn_conv_proj",
+            input_name=f"{i}_block_attn_conv_proj_t",
+            output_name=f"{i}_block_attn_conv_proj",
+            input_channels=768,
+            output_channels=768,
+            W=conv_1D_proj_weights,
+            b=conv_1D_proj_bias,
+            has_bias=True
+        )
+
+        # Input: (seq, 1, 768, 1, 1), Output: (1, seq, 768, 1, 1)
+        builder.add_transpose(
+            name=f"{i}_previous_block_t",
+            input_name=f'{i}_previous_block',
+            output_name=f"{i}_previous_block_t",
+            axes=[1, 0, 2, 3, 4]
+        )
+
+        # Input: [(1, seq, 768, 1, 1), (1, seq, 768, 1, 1)], Output: (1, seq, 768, 1, 1)
+        builder.add_add_broadcastable(
+            name=f"{i}_block_xa_sum",
+            input_names=[f"{i}_previous_block_t", f"{i}_block_attn_conv_proj"],
+            output_name=f"{i}_block_xa_sum",
+            # output_name=f"output_logits"
+        )
+
+        ln_2_weight = model.h[i].ln_2.weight.data.cpu().numpy().reshape((1, 1, 768, 1, 1))
+        ln_2_bias = model.h[i].ln_2.bias.data.cpu().numpy().reshape((1, 1, 768, 1, 1))
+        ln_2_epsilon = model.h[i].ln_2.eps
+
+        # Input: (1, seq, 768, 1, 1), Output:
+        builder.add_mvn(
+            name=f"{i}_block_ln_2",
+            input_name=f"{i}_block_xa_sum",
+            output_name=f"{i}_block_ln_2",
+            across_channels=True,
+            normalize_variance=True,
+            epsilon=ln_2_epsilon
+        )
+
+        builder.add_scale(
+            name=f"{i}_block_ln_2_scaled",
+            input_name=f"{i}_block_ln_2",
+            # output_name=f"output_logits",
+            output_name=f"{i}_block_ln_2_scaled",
+            W=ln_2_weight,
+            b=ln_2_bias,
+            has_bias=True,
+            shape_scale=[768],
+            shape_bias=[768]
+        )
+
+        mlp_conv_1D_fc_bias = model.h[i].mlp.c_fc.bias.data.cpu().numpy().reshape((1, 1, 3072, 1, 1))
+        mlp_conv_1D_fc_weights = model.h[i].mlp.c_fc.weight.data.cpu().numpy().transpose().reshape((1, 768, 3072, 1, 1))
+
+        # Input:, Output: (1, 3, 3072, 1, 1)
+        builder.add_inner_product(
+            name=f"{i}_block_mlp_conv_fc",
+            input_name=f"{i}_block_ln_2_scaled",
+            output_name=f"{i}_block_mlp_conv_fc",
+            # output_name=f"output_logits",
+            input_channels=768,
+            output_channels=3072,
+            W=mlp_conv_1D_fc_weights,
+            b=mlp_conv_1D_fc_bias,
+            has_bias=True
+        )
+
+        builder.add_gelu(
+            name=f"{i}_block_mlp_gelu",
+            input_name=f"{i}_block_mlp_conv_fc",
+            output_name=f"{i}_block_mlp_gelu",
+            # output_name=f"output_logits",
+            mode='TANH_APPROXIMATION'
+        )
+
+        mlp_conv_1D_proj_bias = model.h[i].mlp.c_proj.bias.data.cpu().numpy().reshape((1, 1, 768, 1, 1))
+        mlp_conv_1D_proj_weights = model.h[i].mlp.c_proj.weight.data.cpu().numpy().transpose().reshape((1, 3072, 768, 1, 1))
+
+        # Input:, Output: (1, 3, 3072, 1, 1)
+        builder.add_inner_product(
+            name=f"{i}_block_mlp_conv_proj",
+            input_name=f"{i}_block_mlp_gelu",
+            output_name=f"{i}_block_mlp_conv_proj",
+            # output_name=f"output_logits",
+            input_channels=3072,
+            output_channels=768,
+            W=mlp_conv_1D_proj_weights,
+            b=mlp_conv_1D_proj_bias,
+            has_bias=True
+        )
+
+        builder.add_add_broadcastable(
+            name=f"{i}_block_xm_sum",
+            input_names=[f"{i}_block_xa_sum", f"{i}_block_mlp_conv_proj"],
+            # output_name=f"output_logits"
+            output_name=f"{i + 1}_previous_block_final"
+        )
+
+        builder.add_transpose(
+            name=f"{i}_block_xm_sum_t",
+            input_name=f"{i + 1}_previous_block_final",
+            output_name=f"{i + 1}_previous_block",
+            axes=[1, 0, 2, 3, 4]
+        )
+
+
+    ln_f_weight = model.ln_f.weight.data.cpu().numpy().reshape((1, 1, 768, 1, 1))
+    ln_f_bias = model.ln_f.bias.data.cpu().numpy().reshape((1, 1, 768, 1, 1))
+    ln_f_epsilon = model.ln_f.eps
+
+    # Input: (1, seq, 768, 1, 1), Output:
+    builder.add_mvn(
+        name=f"ln_f",
+        input_name=f"{steps}_previous_block_final",
+        output_name=f"ln_f",
+        # output_name=f"output_logits",
+        across_channels=True,
+        normalize_variance=True,
+        epsilon=ln_f_epsilon
+    )
+
+    builder.add_scale(
+        name=f"ln_f_scaled",
+        input_name=f"ln_f",
+        output_name=f"ln_f_scaled",
+        # output_name=f"output_logits",
+        W=ln_f_weight,
+        b=ln_f_bias,
+        has_bias=True,
+        shape_scale=[768],
+        shape_bias=[768]
+    )
+
+    lm_head_weights = lm_head_model.lm_head.weight.data.cpu().numpy().reshape((1, 50257, 768, 1, 1))
+
+    builder.add_inner_product(
+        name="lm_head",
+        input_name="ln_f_scaled",
+        output_name="output_logits",
+        input_channels=768,
+        output_channels=50257,
+        W=lm_head_weights,
+        b=None,
+        has_bias=False
+    )
+
+    # compile spec to model
+    mlmodel = coremltools.models.MLModel(builder.spec)
+
+    save_spec(builder.spec, f'{model_name}-{sequence_length}-{steps}.mlmodel')

genai_SDK/__init__.py

@@ -0,0 +1,2 @@
+from .Utilities import *
+from .Seq2Seq import *

main.py

@@ -0,0 +1,24 @@
+from datasets import load_dataset, DatasetDict, Dataset
+import pandas as pd
+import ast
+
+import genai_SDK
+from genai_SDK.Seq2Seq import GPT2
+
+# Load dataset from huggingface
+dataset = load_dataset("erwanlc/cocktails_recipe_no_brand")
+
+# Convert to a pandas dataframe
+data = [{'title': item['title'], 'raw_ingredients': item['raw_ingredients']} for item in dataset['train']]
+df = pd.DataFrame(data)
+
+# Just extract the ingredient names, nothing else
+df.raw_ingredients = df.raw_ingredients.apply(lambda x: ', '.join([y[1] for y in ast.literal_eval(x)]))
+#display(df.head())
+
+model = GPT2(gpu=0, model_name="distilgpt2")
+model.load_data(df=df, batch_size=8)
+
+model.train(num_epochs=2)
+
+print(model.generate_text("Annual Planning"))

pyproject.toml

@@ -0,0 +1,25 @@
+[tool.poetry]
+name = "hackweek2024-sup-genai-tools"
+version = "0.1.0"
+description = ""
+authors = ["Your Name <you@example.com>"]
+readme = "README.md"
+
+[tool.poetry.dependencies]
+python = "^3.12"
+torch = "^2.3.1"
+torchtext = "^0.18.0"
+transformers = "^4.41.2"
+sentencepiece = "^0.2.0"
+pandas = "^2.2.2"
+tqdm = "^4.66.4"
+datasets = "^2.20.0"
+scikit-learn = "^1.5.0"
+accelerate = "^0.31.0"
+fastapi = "^0.111.0"
+uvicorn = "^0.30.1"
+
+
+[build-system]
+requires = ["poetry-core"]
+build-backend = "poetry.core.masonry.api"

requirement.txt

@@ -0,0 +1,11 @@
+fastapi
+uvicorn[standard]
+torch
+torchtext
+transformers
+sentencepiece
+pandas
+tqdm
+datasets
+scikit
+accelerate

restrictedItems/parse.py

@@ -0,0 +1,11 @@
+import csv, json
+
+with open('csv/block_items.csv', newline='') as csvfile:
+  reader = csv.reader(csvfile, delimiter=',', quotechar='"')
+  for i, row in enumerate(reader):
+    if i == 0:
+      continue
+    items = json.loads(json.loads(row[3]))
+
+    for item in items:
+      print(item['name'])

restrictedItems/predict.py

@@ -0,0 +1,39 @@
+from transformers import BertTokenizer, BertForSequenceClassification
+import torch
+
+# Load the trained model and tokenizer
+model = BertForSequenceClassification.from_pretrained("/Users/slei/hackweek2024-sup-genai-tools/hackweek2024-sup-genai-tools/trained_model")
+tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
+
+# Function to predict the class of a single input text
+def predict(text):
+  # Preprocess the input text
+  inputs = tokenizer(text, return_tensors='pt', truncation=True, padding=True)
+
+  # Make predictions
+  with torch.no_grad():
+    outputs = model(**inputs)
+
+  # Get the predicted class
+  logits = outputs.logits
+  predicted_class = torch.argmax(logits, dim=1).item()
+
+  return predicted_class
+
+label_map = {0: 'Allowed Item', 1: 'Restricted Item'}
+
+def main():
+  while True:
+    # Prompting the user for input
+    user_input = input("Enter something: ")
+
+    predicted_class = predict(user_input)
+
+    # Map the predicted class to a human-readable label
+    predicted_label = label_map[predicted_class]
+
+    # Displaying the user input
+    print(f'The item "{user_input}" is classified as: "{predicted_label}"')
+
+if __name__ == "__main__":
+  main()

restrictedItems/train.py

@@ -0,0 +1,132 @@
+from transformers import BertTokenizer, BertForSequenceClassification, Trainer, TrainingArguments
+from sklearn.model_selection import train_test_split
+import torch
+
+# Load dataset
+# Assume dataset is a list of tuples (text, label)
+# dataset = [("item description 1", 0), ("item description 2", 1), ...]
+
+restricted_dataset = [
+  ("Promotional Email GiftCard $10", 1),
+  ("$100 Vanilla® Visa", 1),
+  ("Promotional Email GiftCard $10", 1),
+  ("GPS7000 GPS Tracker for", 1),
+  ("$50 Vanilla Visa eGift", 1),
+  ("GPS7000 GPS Tracker for", 1),
+  ("$200 Vanilla Visa Shiny", 1),
+  ("$200 Vanilla Visa Shiny", 1),
+  ("$200 Vanilla Visa Shiny", 1),
+  ("$200 Vanilla® Visa", 1),
+  ("Promotional Email GiftCard $20", 1),
+  ("Xbox $100 Gift Card", 1),
+  ("$25 Vanilla Visa Shiny", 1),
+  ("$25 Vanilla Visa Shiny", 1),
+  ("$25 Vanilla Visa Shiny", 1),
+  ("Xbox $100 Gift Card", 1),
+  ("Home Depot Birthday Cupcake", 1),
+  ("$100 Vanilla® Visa", 1),
+  ("Birthday Celebration Target Giftcard", 1),
+  ("Birthday Celebration Target Giftcard", 1),
+  ("Promotional Email GiftCard $10", 1),
+  ("Birthday Celebration Target Giftcard", 1),
+  ("Nintendo Switch Family Online", 1),
+  ("Promotional Email GiftCard $10", 1),
+  ("$50 Vanilla® Mastercard", 1),
+  ("$50 Vanilla® Mastercard", 1),
+  ("$50 Vanilla® Mastercard", 1),
+  ("$50 Vanilla® Visa", 1),
+  ("$50 Vanilla Visa Shiny", 1),
+  ("Delta Airlines Wedding $250", 1)
+]
+
+
+normal_dataset =[
+  ("Kerrygold Grass-Fed Pure Irish Garlic & Herb Butter Stick, 3", 0),
+   ("bettergoods Garlic, Parmesan, & Basil Butter, 3 oz", 0),
+    ("Birds Eye Savory Herb Riced Cauliflower, 10 oz (Frozen)", 0),
+     ("Great Value Root Blend, Beets, Carrots, Parsnips and Sweet Potatoes", 0),
+      ("Fresh Blueberries, 18 oz Container", 0),
+       ("Mixpresso 3 Piece Black Canisters Sets For The Kitchen, Kitchen Jars With", 0),
+        ("Freshness Guaranteed Chicken Breast Tenderloins, 2.25 - 3.2", 0),
+         ("Kiolbassa Smoked Meats Beef Hickory Smoked Sausage, 4 links - 13oz", 0),
+          ("Hot Pockets Frozen Snacks, Pepperoni Pizza Buttery Crust, 5 Sandwiches", 0),
+           ("Kool Aid Jammers Tropical Punch Kids Drink 0% Juice Box Pouches, 10", 0),
+            ("Frito-Lay Flavor Mix Variety Pack Snack Chips, 1oz Bags, 18 Count", 0),
+             ("State Fair Classic Corn Dogs, 42.7 oz, 16 Count", 0),
+              ("ASURION 2 Year Sporting Goods Protection Plan ($175 - $199.99)", 0),
+               ("6% Incline Walking Pad Treadmill 320+ lb Capacity, Under The Desk", 0),
+                ("Renpure Biotin & Collagen Thickening Conditioner for All Hair Types, 32 fl", 0),
+                 ("Renpure Biotin & Collagen Thickening Hair Shampoo for All Hair Types, 32", 0),
+                  ("eos Shea Better Body Lotion for Dry Skin, Vanilla Cashmere, 16 fl", 0),
+                   ("Degree Ultra Clear Long Lasting Men's Antiperspirant Deodorant Dry Spray,", 0),
+                    ("Tide PODS Liquid Laundry Detergent, Original Scent, HE Compatible, 42 Count", 0),
+                     ("DEER PARK Brand 100% Natural Spring Water, 16.9-ounce", 0),
+                      ("Great Value Milk Whole Vitamin D Gallon Plastic Jug", 0),
+                       ("Jumbo Russet Potatoes Whole Fresh, 8 lb Bag", 0),
+                        ("Great Value Butter Pecan Flavored Ice Cream, 16 fl oz", 0),
+                         ("Beef Lean Stew Meat, 1.0 - 1.5 lb Tray", 0),
+                          ("Great Value Spaghetti 16oz", 0),
+                           ("Great Value Flavored with Meat Pasta Sauce, 24 oz", 0),
+                            ("Kentucky Kernel Original Seasoned Flour, Coating Mix for Frying, Value Size", 0)
+  ]
+
+dataset = restricted_dataset + normal_dataset
+
+# Split dataset
+train_texts, val_texts, train_labels, val_labels = train_test_split([item[0] for item in dataset], [item[1] for item in dataset], test_size=0.2)
+import pdb; pdb.set_trace()
+
+# Load pre-trained BERT tokenizer
+tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
+
+# Tokenize data
+train_encodings = tokenizer(train_texts, truncation=True, padding=True)
+val_encodings = tokenizer(val_texts, truncation=True, padding=True)
+
+# Convert to torch Dataset
+class ShoppingCartDataset(torch.utils.data.Dataset):
+  def __init__(self, encodings, labels):
+    self.encodings = encodings
+    self.labels = labels
+
+  def __getitem__(self, idx):
+    item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
+    item['labels'] = torch.tensor(self.labels[idx])
+    return item
+
+  def __len__(self):
+    return len(self.labels)
+
+train_dataset = ShoppingCartDataset(train_encodings, train_labels)
+val_dataset = ShoppingCartDataset(val_encodings, val_labels)
+
+# Load pre-trained BERT model
+model = BertForSequenceClassification.from_pretrained('bert-base-uncased')
+
+# Training arguments
+training_args = TrainingArguments(
+  output_dir='../results',
+  num_train_epochs=3,
+  per_device_train_batch_size=16,
+  per_device_eval_batch_size=16,
+  warmup_steps=500,
+  weight_decay=0.01,
+  logging_dir='./logs',
+  logging_steps=10,
+)
+
+# Trainer
+trainer = Trainer(
+  model=model,
+  args=training_args,
+  train_dataset=train_dataset,
+  eval_dataset=val_dataset,
+)
+
+# Train model
+trainer.train()
+
+# Evaluate model
+trainer.evaluate()
+
+model.save_pretrained('trained_model')