From 1d3506ae720e6cdd3ab08931a65ee7a1a996746c Mon Sep 17 00:00:00 2001
From: CSY-ModelCloud <csy@modelcloud.ai>
Date: Sat, 22 Jun 2024 13:48:04 +0800
Subject: [PATCH 01/17] auto find good 4090

---
 .github/workflows/run_tests.yml | 19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)

diff --git a/.github/workflows/run_tests.yml b/.github/workflows/run_tests.yml
index d6f07f40..23f0aa3e 100644
--- a/.github/workflows/run_tests.yml
+++ b/.github/workflows/run_tests.yml
@@ -13,9 +13,7 @@ concurrency:
   group: ${{ github.workflow }}-${{ github.ref }}
 
 env:
-  DAY_OF_WEEK: Monday
   CUDA_DEVICE_ORDER: PCI_BUS_ID
-  CUDA_VISIBLE_DEVICES: 0
 
 jobs:
   build:
@@ -27,6 +25,23 @@ jobs:
       - name: Checkout code
         uses: actions/checkout@v4
 
+      - name: Find suitable GPU and set CUDA_VISIBLE_DEVICES
+        run: |
+          suitable_gpu=$(nvidia-smi -L | grep "RTX 4090" | awk -F': ' '{print $1}' | sed 's/GPU //g' | while read gpu_id
+          do
+            mem_total=$(nvidia-smi --query-gpu=memory.total --format=csv,noheader,nounits -i $gpu_id)
+            mem_used=$(nvidia-smi --query-gpu=memory.used --format=csv,noheader,nounits -i $gpu_id)
+            mem_used_pct=$((100 * mem_used / mem_total))
+            if [ $mem_used_pct -lt 2 ]; then
+              echo $gpu_id
+              break
+            fi
+          done)
+          echo "CUDA_VISIBLE_DEVICES=$suitable_gpu" >> $GITHUB_ENV
+
+      - name: Display Environment Variable
+        run: echo "CUDA_VISIBLE_DEVICES set to ${{ env.CUDA_VISIBLE_DEVICES }}"
+
       - name: Compile
         run: |
           pip install -U pytest ninja parameterized protobuf

From 4649e43010c7f49ed502a05d83b87c3e4d93f1ca Mon Sep 17 00:00:00 2001
From: CSY-ModelCloud <csy@modelcloud.ai>
Date: Sat, 22 Jun 2024 14:03:46 +0800
Subject: [PATCH 02/17] move to top

---
 .github/workflows/run_tests.yml | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/.github/workflows/run_tests.yml b/.github/workflows/run_tests.yml
index 23f0aa3e..c474d924 100644
--- a/.github/workflows/run_tests.yml
+++ b/.github/workflows/run_tests.yml
@@ -22,17 +22,14 @@ jobs:
       image: pytorch/pytorch:2.3.1-cuda12.1-cudnn8-devel
 
     steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: Find suitable GPU and set CUDA_VISIBLE_DEVICES
+      - name: Find suitable GPU
         run: |
           suitable_gpu=$(nvidia-smi -L | grep "RTX 4090" | awk -F': ' '{print $1}' | sed 's/GPU //g' | while read gpu_id
           do
             mem_total=$(nvidia-smi --query-gpu=memory.total --format=csv,noheader,nounits -i $gpu_id)
             mem_used=$(nvidia-smi --query-gpu=memory.used --format=csv,noheader,nounits -i $gpu_id)
             mem_used_pct=$((100 * mem_used / mem_total))
-            if [ $mem_used_pct -lt 2 ]; then
+            if [ $mem_used_pct -lt 2 ]; then # 2 -> 98% free
               echo $gpu_id
               break
             fi
@@ -42,6 +39,9 @@ jobs:
       - name: Display Environment Variable
         run: echo "CUDA_VISIBLE_DEVICES set to ${{ env.CUDA_VISIBLE_DEVICES }}"
 
+      - name: Checkout code
+        uses: actions/checkout@v4
+
       - name: Compile
         run: |
           pip install -U pytest ninja parameterized protobuf

From 62a6647e493c52034289c8f9b86b42f39982b442 Mon Sep 17 00:00:00 2001
From: CSY-ModelCloud <csy@modelcloud.ai>
Date: Sat, 22 Jun 2024 14:13:45 +0800
Subject: [PATCH 03/17] move find gpu to top of the test

---
 .github/workflows/run_tests.yml | 27 +++++++++++++++------------
 1 file changed, 15 insertions(+), 12 deletions(-)

diff --git a/.github/workflows/run_tests.yml b/.github/workflows/run_tests.yml
index c474d924..8bb79c3a 100644
--- a/.github/workflows/run_tests.yml
+++ b/.github/workflows/run_tests.yml
@@ -22,6 +22,14 @@ jobs:
       image: pytorch/pytorch:2.3.1-cuda12.1-cudnn8-devel
 
     steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Compile
+        run: |
+          pip install -U pytest ninja parameterized protobuf
+          pip install -v --no-build-isolation .
+
       - name: Find suitable GPU
         run: |
           suitable_gpu=$(nvidia-smi -L | grep "RTX 4090" | awk -F': ' '{print $1}' | sed 's/GPU //g' | while read gpu_id
@@ -34,18 +42,13 @@ jobs:
               break
             fi
           done)
-          echo "CUDA_VISIBLE_DEVICES=$suitable_gpu" >> $GITHUB_ENV
-
-      - name: Display Environment Variable
-        run: echo "CUDA_VISIBLE_DEVICES set to ${{ env.CUDA_VISIBLE_DEVICES }}"
-
-      - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: Compile
-        run: |
-          pip install -U pytest ninja parameterized protobuf
-          pip install -v --no-build-isolation .
+          if [ -z "$suitable_gpu" ]; then
+            echo "No suitable GPU found. Exiting with error."
+            exit 1
+          else
+            echo "CUDA_VISIBLE_DEVICES=$suitable_gpu" >> $GITHUB_ENV
+          echo "CUDA_VISIBLE_DEVICES set to ${{ env.CUDA_VISIBLE_DEVICES }}"
+          fi
 
       - name: test_lm_head.py
         if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_lm_head') }}

From 34f2887612ad1186dd38e916c23940cd16cf594e Mon Sep 17 00:00:00 2001
From: CSY-ModelCloud <csy@modelcloud.ai>
Date: Sat, 22 Jun 2024 14:26:23 +0800
Subject: [PATCH 04/17] test env was set

---
 .github/workflows/run_tests.yml | 18 ++++++++++--------
 1 file changed, 10 insertions(+), 8 deletions(-)

diff --git a/.github/workflows/run_tests.yml b/.github/workflows/run_tests.yml
index 8bb79c3a..2301c182 100644
--- a/.github/workflows/run_tests.yml
+++ b/.github/workflows/run_tests.yml
@@ -22,13 +22,6 @@ jobs:
       image: pytorch/pytorch:2.3.1-cuda12.1-cudnn8-devel
 
     steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: Compile
-        run: |
-          pip install -U pytest ninja parameterized protobuf
-          pip install -v --no-build-isolation .
 
       - name: Find suitable GPU
         run: |
@@ -47,9 +40,18 @@ jobs:
             exit 1
           else
             echo "CUDA_VISIBLE_DEVICES=$suitable_gpu" >> $GITHUB_ENV
-          echo "CUDA_VISIBLE_DEVICES set to ${{ env.CUDA_VISIBLE_DEVICES }}"
+            echo "CUDA_VISIBLE_DEVICES set to ${{ env.CUDA_VISIBLE_DEVICES }}"
           fi
 
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Compile
+        run: |
+          pip install -U pytest ninja parameterized protobuf
+          pip install -v --no-build-isolation .
+
+
       - name: test_lm_head.py
         if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_lm_head') }}
         run: pytest tests/test_lm_head.py

From d08165caf880dedae982158756082db5bd1031a1 Mon Sep 17 00:00:00 2001
From: CSY-ModelCloud <csy@modelcloud.ai>
Date: Sat, 22 Jun 2024 14:33:36 +0800
Subject: [PATCH 05/17] remove action/checkout

---
 .github/workflows/run_tests.yml | 31 +++++++++++++++----------------
 1 file changed, 15 insertions(+), 16 deletions(-)

diff --git a/.github/workflows/run_tests.yml b/.github/workflows/run_tests.yml
index 2301c182..d894421b 100644
--- a/.github/workflows/run_tests.yml
+++ b/.github/workflows/run_tests.yml
@@ -22,10 +22,9 @@ jobs:
       image: pytorch/pytorch:2.3.1-cuda12.1-cudnn8-devel
 
     steps:
-
       - name: Find suitable GPU
         run: |
-          suitable_gpu=$(nvidia-smi -L | grep "RTX 4090" | awk -F': ' '{print $1}' | sed 's/GPU //g' | while read gpu_id
+         suitable_gpu=$(nvidia-smi -L | grep "RTX 4090" | awk -F': ' '{print $1}' | sed 's/GPU //g' | while read gpu_id
           do
             mem_total=$(nvidia-smi --query-gpu=memory.total --format=csv,noheader,nounits -i $gpu_id)
             mem_used=$(nvidia-smi --query-gpu=memory.used --format=csv,noheader,nounits -i $gpu_id)
@@ -44,61 +43,61 @@ jobs:
           fi
 
       - name: Checkout code
-        uses: actions/checkout@v4
+        run: git clone --depth=1 https://github.com/ModelCloud/GPTQModel
 
       - name: Compile
         run: |
           pip install -U pytest ninja parameterized protobuf
-          pip install -v --no-build-isolation .
+          pip install -v --no-build-isolation GPTQModel
 
 
       - name: test_lm_head.py
         if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_lm_head') }}
-        run: pytest tests/test_lm_head.py
+        run: pytest GPTQModel/tests/test_lm_head.py
 
       - name: test_q4_exallama.py
         if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_q4_exallama') }}
-        run: pytest tests/test_q4_exallama.py
+        run: pytest GPTQModel/tests/test_q4_exallama.py
 
       - name: test_q4_exallama_v2.py
         if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_q4_exallama_v2') }}
-        run: pytest tests/test_q4_exallama_v2.py
+        run: pytest GPTQModel/tests/test_q4_exallama_v2.py
 
       - name: test_q4_marlin.py
         if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_q4_marlin') }}
-        run: pytest tests/test_q4_marlin.py
+        run: pytest GPTQModel/tests/test_q4_marlin.py
 
       - name: test_q4_triton.py
         if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_q4_triton') }}
-        run: pytest tests/test_q4_triton.py
+        run: pytest GPTQModel/tests/test_q4_triton.py
 
       - name: test_repacking.py
         if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_repacking') }}
-        run: pytest tests/test_repacking.py
+        run: pytest GPTQModel/tests/test_repacking.py
 
       - name: test_serialization.py
         if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_serialization') }}
-        run: pytest tests/test_serialization.py
+        run: pytest GPTQModel/tests/test_serialization.py
 
       - name: test_sharded.py
         if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_sharded') }}
-        run: pytest tests/test_sharded.py
+        run: pytest GPTQModel/tests/test_sharded.py
 
       - name: test_triton.py
         if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_triton') }}
-        run: pytest tests/test_triton.py
+        run: pytest GPTQModel/tests/test_triton.py
 
       - name: test_quant_formats.py
         if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_quant_formats') }}
-        run: pytest tests/test_quant_formats.py
+        run: pytest GPTQModel/tests/test_quant_formats.py
 
       - name: test_q4_cuda.py
         if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_q4_cuda') }}
-        run: pytest tests/test_q4_cuda.py
+        run: pytest GPTQModel/tests/test_q4_cuda.py
 
       - name: test_perplexity.py
         if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_perplexity') }}
-        run: pytest tests/test_perplexity.py
+        run: pytest GPTQModel/tests/test_perplexity.py
 
 
         

From da5a5f13bb32f7892292c79fec7238cd7b0db5d1 Mon Sep 17 00:00:00 2001
From: CSY-ModelCloud <csy@modelcloud.ai>
Date: Sat, 22 Jun 2024 14:37:34 +0800
Subject: [PATCH 06/17] Update run_tests.yml

---
 .github/workflows/run_tests.yml | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/.github/workflows/run_tests.yml b/.github/workflows/run_tests.yml
index d894421b..e3df60ac 100644
--- a/.github/workflows/run_tests.yml
+++ b/.github/workflows/run_tests.yml
@@ -42,8 +42,11 @@ jobs:
             echo "CUDA_VISIBLE_DEVICES set to ${{ env.CUDA_VISIBLE_DEVICES }}"
           fi
 
+      - name: Display Environment Variable
+        run: echo "CUDA_VISIBLE_DEVICES set to ${{ env.CUDA_VISIBLE_DEVICES }}"
+
       - name: Checkout code
-        run: git clone --depth=1 https://github.com/ModelCloud/GPTQModel
+        run: apt update && apt install git && git clone --depth=1 https://github.com/ModelCloud/GPTQModel
 
       - name: Compile
         run: |

From b7fed291b05692e785a098f0198340e2855d82f5 Mon Sep 17 00:00:00 2001
From: CSY-ModelCloud <csy@modelcloud.ai>
Date: Sat, 22 Jun 2024 14:41:54 +0800
Subject: [PATCH 07/17] use action/checkout

---
 .github/workflows/run_tests.yml | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/.github/workflows/run_tests.yml b/.github/workflows/run_tests.yml
index e3df60ac..e024a96a 100644
--- a/.github/workflows/run_tests.yml
+++ b/.github/workflows/run_tests.yml
@@ -39,14 +39,11 @@ jobs:
             exit 1
           else
             echo "CUDA_VISIBLE_DEVICES=$suitable_gpu" >> $GITHUB_ENV
-            echo "CUDA_VISIBLE_DEVICES set to ${{ env.CUDA_VISIBLE_DEVICES }}"
+            echo "CUDA_VISIBLE_DEVICES set to $suitable_gpu"
           fi
 
-      - name: Display Environment Variable
-        run: echo "CUDA_VISIBLE_DEVICES set to ${{ env.CUDA_VISIBLE_DEVICES }}"
-
       - name: Checkout code
-        run: apt update && apt install git && git clone --depth=1 https://github.com/ModelCloud/GPTQModel
+        uses: actions/checkout@v4
 
       - name: Compile
         run: |

From e86e4b3e66a1002fd5ed1424fe49329bb58ee9fb Mon Sep 17 00:00:00 2001
From: CSY-ModelCloud <csy@modelcloud.ai>
Date: Sat, 22 Jun 2024 15:56:03 +0800
Subject: [PATCH 08/17] add matrix

---
 .github/workflows/run_tests.yml | 81 +++++++++------------------------
 1 file changed, 22 insertions(+), 59 deletions(-)

diff --git a/.github/workflows/run_tests.yml b/.github/workflows/run_tests.yml
index e024a96a..bce5e1d2 100644
--- a/.github/workflows/run_tests.yml
+++ b/.github/workflows/run_tests.yml
@@ -3,14 +3,6 @@ name: Unit Tests (GPU)
 on:
   repository_dispatch:
   workflow_dispatch:
-    inputs:
-      test_names:
-        description: 'Input Test(s) to Run (default all)'
-        required: false
-        default: ''
-
-concurrency:
-  group: ${{ github.workflow }}-${{ github.ref }}
 
 env:
   CUDA_DEVICE_ORDER: PCI_BUS_ID
@@ -18,6 +10,11 @@ env:
 jobs:
   build:
     runs-on: self-hosted
+    strategy:
+      matrix:
+        test-group:
+          - gpu_group1
+          - gpu_group2
     container:
       image: pytorch/pytorch:2.3.1-cuda12.1-cudnn8-devel
 
@@ -50,54 +47,20 @@ jobs:
           pip install -U pytest ninja parameterized protobuf
           pip install -v --no-build-isolation GPTQModel
 
-
-      - name: test_lm_head.py
-        if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_lm_head') }}
-        run: pytest GPTQModel/tests/test_lm_head.py
-
-      - name: test_q4_exallama.py
-        if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_q4_exallama') }}
-        run: pytest GPTQModel/tests/test_q4_exallama.py
-
-      - name: test_q4_exallama_v2.py
-        if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_q4_exallama_v2') }}
-        run: pytest GPTQModel/tests/test_q4_exallama_v2.py
-
-      - name: test_q4_marlin.py
-        if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_q4_marlin') }}
-        run: pytest GPTQModel/tests/test_q4_marlin.py
-
-      - name: test_q4_triton.py
-        if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_q4_triton') }}
-        run: pytest GPTQModel/tests/test_q4_triton.py
-
-      - name: test_repacking.py
-        if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_repacking') }}
-        run: pytest GPTQModel/tests/test_repacking.py
-
-      - name: test_serialization.py
-        if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_serialization') }}
-        run: pytest GPTQModel/tests/test_serialization.py
-
-      - name: test_sharded.py
-        if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_sharded') }}
-        run: pytest GPTQModel/tests/test_sharded.py
-
-      - name: test_triton.py
-        if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_triton') }}
-        run: pytest GPTQModel/tests/test_triton.py
-
-      - name: test_quant_formats.py
-        if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_quant_formats') }}
-        run: pytest GPTQModel/tests/test_quant_formats.py
-
-      - name: test_q4_cuda.py
-        if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_q4_cuda') }}
-        run: pytest GPTQModel/tests/test_q4_cuda.py
-
-      - name: test_perplexity.py
-        if: ${{ !github.event.inputs.test_names || contains(github.event.inputs.test_names, 'test_perplexity') }}
-        run: pytest GPTQModel/tests/test_perplexity.py
-
-
-        
+      - name: Run tests
+        if: matrix.test-group == 'gpu_group2'
+        run: pytest tests/test_perplexity.py
+      - name: Run tests
+        if: matrix.test-group == 'gpu_group1'
+        run: |
+          pytest tests/test_lm_head.py
+          pytest tests/test_q4_exallama.py
+          pytest tests/test_q4_exallama_v2.py
+          pytest tests/test_q4_marlin.py
+          pytest tests/test_q4_triton.py
+          pytest tests/test_repacking.py
+          pytest tests/test_serialization.py
+          pytest tests/test_sharded.py
+          pytest tests/test_triton.py
+          pytest tests/test_quant_formats.py
+          pytest tests/test_q4_cuda.py

From bf368992cb240314cd0a30d98f00baff039c28a3 Mon Sep 17 00:00:00 2001
From: CSY-ModelCloud <csy@modelcloud.ai>
Date: Sat, 22 Jun 2024 16:57:06 +0800
Subject: [PATCH 09/17] Update run_tests.yml

---
 .github/workflows/run_tests.yml | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/.github/workflows/run_tests.yml b/.github/workflows/run_tests.yml
index bce5e1d2..d3d66185 100644
--- a/.github/workflows/run_tests.yml
+++ b/.github/workflows/run_tests.yml
@@ -44,6 +44,8 @@ jobs:
 
       - name: Compile
         run: |
+          apt update && apt install net-tools iproute2 -y
+          ip link show
           pip install -U pytest ninja parameterized protobuf
           pip install -v --no-build-isolation GPTQModel
 

From 86fe23cb7c8d59bf1130064f295ed53f0ae90c12 Mon Sep 17 00:00:00 2001
From: CSY-ModelCloud <csy@modelcloud.ai>
Date: Sat, 22 Jun 2024 17:00:09 +0800
Subject: [PATCH 10/17] modelcloud/gptqmodel

---
 .github/workflows/run_tests.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/run_tests.yml b/.github/workflows/run_tests.yml
index d3d66185..0be30806 100644
--- a/.github/workflows/run_tests.yml
+++ b/.github/workflows/run_tests.yml
@@ -16,7 +16,7 @@ jobs:
           - gpu_group1
           - gpu_group2
     container:
-      image: pytorch/pytorch:2.3.1-cuda12.1-cudnn8-devel
+      image: modelcloud/gptqmodel
 
     steps:
       - name: Find suitable GPU

From 977c08b9c48d9aa2ba4b77d7f1bb1cb8f27a28d0 Mon Sep 17 00:00:00 2001
From: CSY-ModelCloud <csy@modelcloud.ai>
Date: Sat, 22 Jun 2024 17:07:05 +0800
Subject: [PATCH 11/17] Update run_tests.yml

---
 .github/workflows/run_tests.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/run_tests.yml b/.github/workflows/run_tests.yml
index 0be30806..d3d66185 100644
--- a/.github/workflows/run_tests.yml
+++ b/.github/workflows/run_tests.yml
@@ -16,7 +16,7 @@ jobs:
           - gpu_group1
           - gpu_group2
     container:
-      image: modelcloud/gptqmodel
+      image: pytorch/pytorch:2.3.1-cuda12.1-cudnn8-devel
 
     steps:
       - name: Find suitable GPU

From f41fdc49a1cfd3ff45978c9a1530df49a129df18 Mon Sep 17 00:00:00 2001
From: CSY-ModelCloud <csy@modelcloud.ai>
Date: Sat, 22 Jun 2024 17:08:44 +0800
Subject: [PATCH 12/17] remove test

---
 .github/workflows/run_tests.yml | 2 --
 1 file changed, 2 deletions(-)

diff --git a/.github/workflows/run_tests.yml b/.github/workflows/run_tests.yml
index d3d66185..bce5e1d2 100644
--- a/.github/workflows/run_tests.yml
+++ b/.github/workflows/run_tests.yml
@@ -44,8 +44,6 @@ jobs:
 
       - name: Compile
         run: |
-          apt update && apt install net-tools iproute2 -y
-          ip link show
           pip install -U pytest ninja parameterized protobuf
           pip install -v --no-build-isolation GPTQModel
 

From 71440f865348032ea4ba14beba7734c6a37ca256 Mon Sep 17 00:00:00 2001
From: CSY-ModelCloud <csy@modelcloud.ai>
Date: Sat, 22 Jun 2024 17:10:01 +0800
Subject: [PATCH 13/17] fix path

---
 .github/workflows/run_tests.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/run_tests.yml b/.github/workflows/run_tests.yml
index bce5e1d2..f0dab36d 100644
--- a/.github/workflows/run_tests.yml
+++ b/.github/workflows/run_tests.yml
@@ -45,7 +45,7 @@ jobs:
       - name: Compile
         run: |
           pip install -U pytest ninja parameterized protobuf
-          pip install -v --no-build-isolation GPTQModel
+          pip install -v --no-build-isolation .
 
       - name: Run tests
         if: matrix.test-group == 'gpu_group2'

From ec4a5716007a97649c1e7ce6ce7ad749b1df43db Mon Sep 17 00:00:00 2001
From: CSY-ModelCloud <csy@modelcloud.ai>
Date: Sat, 22 Jun 2024 17:16:10 +0800
Subject: [PATCH 14/17] modelcloud/gptqmodel:github-ci-v1

---
 .github/workflows/run_tests.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/run_tests.yml b/.github/workflows/run_tests.yml
index f0dab36d..f9203d1c 100644
--- a/.github/workflows/run_tests.yml
+++ b/.github/workflows/run_tests.yml
@@ -16,7 +16,7 @@ jobs:
           - gpu_group1
           - gpu_group2
     container:
-      image: pytorch/pytorch:2.3.1-cuda12.1-cudnn8-devel
+      image: modelcloud/gptqmodel:github-ci-v1
 
     steps:
       - name: Find suitable GPU

From 38803e496bee5a2f27395944003b8b2504e32104 Mon Sep 17 00:00:00 2001
From: CSY-ModelCloud <csy@modelcloud.ai>
Date: Sat, 22 Jun 2024 17:20:00 +0800
Subject: [PATCH 15/17] Update run_tests.yml

---
 .github/workflows/run_tests.yml | 30 +++++++++++++++++++-----------
 1 file changed, 19 insertions(+), 11 deletions(-)

diff --git a/.github/workflows/run_tests.yml b/.github/workflows/run_tests.yml
index f9203d1c..d7b845d6 100644
--- a/.github/workflows/run_tests.yml
+++ b/.github/workflows/run_tests.yml
@@ -53,14 +53,22 @@ jobs:
       - name: Run tests
         if: matrix.test-group == 'gpu_group1'
         run: |
-          pytest tests/test_lm_head.py
-          pytest tests/test_q4_exallama.py
-          pytest tests/test_q4_exallama_v2.py
-          pytest tests/test_q4_marlin.py
-          pytest tests/test_q4_triton.py
-          pytest tests/test_repacking.py
-          pytest tests/test_serialization.py
-          pytest tests/test_sharded.py
-          pytest tests/test_triton.py
-          pytest tests/test_quant_formats.py
-          pytest tests/test_q4_cuda.py
+          run_test() {
+              echo -e "\033[1;34mRunning $1...\033[0m"
+              echo "========================================"
+              pytest tests/$1
+              echo -e "\033[1;32mFinished $1\033[0m"
+              echo "========================================"
+          }
+
+          run_test test_lm_head.py
+          run_test test_q4_exallama.py
+          run_test test_q4_exallama_v2.py
+          run_test test_q4_marlin.py
+          run_test test_q4_triton.py
+          run_test test_repacking.py
+          run_test test_serialization.py
+          run_test test_sharded.py
+          run_test test_triton.py
+          run_test test_quant_formats.py
+          run_test test_q4_cuda.py

From c7e2c2bf9942807e113d0cf395ff3e221838f327 Mon Sep 17 00:00:00 2001
From: CSY-ModelCloud <csy@modelcloud.ai>
Date: Sat, 22 Jun 2024 17:20:45 +0800
Subject: [PATCH 16/17] pytorch/pytorch:2.3.1-cuda12.1-cudnn8-devel

---
 .github/workflows/run_tests.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/run_tests.yml b/.github/workflows/run_tests.yml
index d7b845d6..d9bfa138 100644
--- a/.github/workflows/run_tests.yml
+++ b/.github/workflows/run_tests.yml
@@ -16,7 +16,7 @@ jobs:
           - gpu_group1
           - gpu_group2
     container:
-      image: modelcloud/gptqmodel:github-ci-v1
+      image: pytorch/pytorch:2.3.1-cuda12.1-cudnn8-devel
 
     steps:
       - name: Find suitable GPU

From e2c09d803b1f5a1a6c13b209a9ec53ccf8690376 Mon Sep 17 00:00:00 2001
From: CSY-ModelCloud <csy@modelcloud.ai>
Date: Sat, 22 Jun 2024 17:27:13 +0800
Subject: [PATCH 17/17] remove -e

---
 .github/workflows/run_tests.yml | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/.github/workflows/run_tests.yml b/.github/workflows/run_tests.yml
index d9bfa138..5641dd26 100644
--- a/.github/workflows/run_tests.yml
+++ b/.github/workflows/run_tests.yml
@@ -54,11 +54,11 @@ jobs:
         if: matrix.test-group == 'gpu_group1'
         run: |
           run_test() {
-              echo -e "\033[1;34mRunning $1...\033[0m"
+              echo "\033[1;34mRunning $1...\033[0m"
               echo "========================================"
               pytest tests/$1
-              echo -e "\033[1;32mFinished $1\033[0m"
-              echo "========================================"
+              echo "\033[1;32mFinished $1\033[0m"
+              echo "========================================\n\n\n"
           }
 
           run_test test_lm_head.py